/diy/ - Do It Yourself

Old thread >>1138806
reached bump limit

I'm making an electric vehicle with an onboard computer and some pretty expensive sensors. The voltage of the sensors is 12V (while the battery is 24V) and I want to add some protection to assure that the sensors will never be damaged by some misfortune.
Is a voltage regulator and a fuse enough to protect my devices? I thought about using a voltage clipper to remove any possible overvoltage and protect against reversed polarity

>>1144976
you mean sense side protection? use a zener diode and resistor. value for R1 is the lowest it can be without exceeding current rating of the diode.

as for supply side, just run another circuit on a regulator, a decent one with a high output rating will already have plenty of output suppression built in.

curious as to why you are using 12V sensors when the industry norm is to use 5V for everything besides mechanically switched circuits(lights, HVAC, etc). you could probably save a lot of cost and complexity that way.

>>1144887
>If I make another charger from one does it matter if it's not 300ma, probably more?

yes it matters. in order to make these chargers a few pennies cheaper, they use the fact that the transformer itself will limit the current to the batteries. if you use a charger with a higher current rating, you'll send too much current to the batteries, making them hot-and-bothered, and shortening their lives.

you can still use those chargers, but you gotta find a way to limit the current, using a series resistor, or preferably, an LM317 in constant current mode.

>>1144976
They make automotive rated regulators that will handle a lot of abuse, y'know.

>>1144983
Thank you, anon.

It is well-known that the resistance of a light-dependent resistor (LDR) varies exponentially with respect to the intensity of the light falling on the photo-sensitive surface.

Would it be possible to use a logarithmic amplifier circuit (or IC) to linearize the response of the LDR? This would make it more useful for interfacing with an A/D-converter or microcontroller chip.

>>1145097
why not just curve fit the data in software?

>>1145104
To utilize most of the dynamic range of an A/D converter.
And, to experiment.

>>1144912
Would 1N4007 diodes work well for this? Or should I use some other diode?

>>1144988
the thing is I don't know when a regulator is good or not. most things I find in my country are chinese and unbranded, and importing from the us can get quite expensive.
I was thinking about maybe using a power inverter and a PC power supply to get 5V and 12V.

Could someone help me identify this component? I've already tried google to no avail. I suppose it's a coil. Rumor has it, that it was used as a filter with a capacitor. Thanks in advance.

>>1144981
I don't quite understand what you mean by sense side protection, but the pic you posted is exactly what I was thinking about using, with a fuse in series with the resistor.

>curious as to why you are using 12V sensors
kinect v2, hokuyo UST-10LX

>>1145253
It makes me think of those 3-lead varistor-capacitor combinations, like Murata's DSN9H series, but I'm not too sure.

>>1145253
it doesn't look quite right but could it be a ceramic resonator?

>>1145197
yes, it will be fine

>>1145253
>>1145267
the datasheet has equivalent circuit diagrams for all of the components in that series so you could possibly use a multimeter to try and figure out if it's what anon suggested it could be

>>1145267
>>1145273
>>1145285
Thanks a lot for the replies. As >>1145267
suggested; it seems indeed like a varistor. However, as I previously wrote, looking for the code yields no results on google.

I ignore the behaviour or testing of these; but here is what I have done so far.

Measuring Resistance from a side lead and the middle lead shows 8.8-10 ohms.

Measuring from both side leads, shows an increasing value that's only readable on the maximum dial, and then becomes unreadable again( OL).

Finally measuring from the other side lead and the middle one shows once again an OL.

If someone could share a good reference link/ video on the topic I would really appreciate.

It's resistometer anon again. AGAIN. So I wrote a little python code to simulate the measurement of the voltage across the test resistor (actually across the reference resistor, but same difference) by the ADC, the rounding by the ADC, and comparing this to the actual value.

Now I thought the values of test resistor near the value of the reference resistor would all be perfectly accurate, or at least near enough so that a 24b ADC will display it perfectly on a 4_digit display. But no, with some very small step sizes, I found that 527.85Ω is rounded up to 527.9 by just rounding the ideal resistance, but is rounded down to 527.8 by rounding the ADC approximation. I'm not sure if /diy/ is the right place to be asking this, but is this one of those "python takes 527.85 as 527.84999999999 and rounds it down" things? Or is this an actual product of the ADC?

I'm just doing this to figure out what values the ADC can read accurately enough to not flash a "you're shit" light at me or switch range to a more appropriate resistor.

Is this okay for a beginner? I am just a 1st year EE student who wants to practice some of his labs and independent projects outside of class.

https://www.amazon.com/dp/B00ZBCLJSY/ref=wl_it_dp_o_pC_nS_ttl?_encoding=UTF8&colid=Z0FSG1M43Y8X&coliid=I2UA1WQDUP7L2U

>>1145438
Personally I think it's too much, unless you want to already start you lab equipment collection. You could Try a linear voltage power supply for small breadboard projects.

http://www.banggood.com/EU-220V-DIY-LM317-Adjustable-Voltage-Power-Supply-Board-Kit-With-Case-p-973391.html?rmmds=search

>>1145440
Hm I didn't consider variable versions of these. I saw a little one like that, that only output 5.0V This is pretty cool, it should be adequate for most of my labs.

I don't have a soldering kit though.It's a digital course and we are only using breadboards atm. Is it a bad idea to cheap out on a soldering kit?

>>1145446
http://www.banggood.com/Mustool-MT223-60W-Adjustable-Temperature-Electric-Soldering-Iron-with-5pcs-Solder-Tips-p-1077119.html?rmmds=search

Banggood and aliexpress are your friends here. Welcome aboard.

>>1145453
I appreciate all the help. Time to learn how to solder.

>>1145446
You honestly don't need any more than a soldering iron that doesn't melt in your hands for basic hobby stuff. High heat capacity is nice too. You'll probably also want some helping hands if you're doing a lot of wire-to-wire soldering, and a board-vice too if you're doing PCB soldering. I don't have either, just a strip of galvanised steel with steep triangles cut out of it which I jam wires into, and it does the trick.

Hey /diy/, what's the series resistance of a linear voltage regulator as opposed to a voltage reference like? Is there any other choice for a voltage source with very low internal resistance? I'd like to be able to handle high currents.

>>1145097
>It is well-known that the resistance of a light-dependent resistor (LDR) varies exponentially
This depends on the LDR type, but generally not very and the parameters or curves shown in the datasheets aren't that accurate either.
Why not use a photodiode and a logarithmic amplifier? There are plenty of schematics available for such circuits, some with excellent performance over huge range.

One such circuit: http://www.linear.com/solutions/1598

>>1145400
It would be much easier to answer if you told what exact values you used and in what equation.

>>1145540
This is specified in the datasheet. Sometimes the output resistance is specified directly, sometimes you need to calculate it from the stated load or output voltage regulation value.
For example, one 7805 datasheet says that the output voltage changes typically 15mV, when the output current changes from 5mA to 1.5A. This gives you a typical output resistance of 10mR.The same datasheet also tells that the output resistance is typically 17mR at 1kHz.

For comparison, LTC6652 (a rather nice reference) has a stated load regulation value of 20ppm/mA. For 5V version this means 100mR output resistance.

Point being, it is very unfair to compare a regulator meant to deliver amps to a reference meant to deliver milliamps (or preferably less). Also, high output current regulators generally have (must have) lower output resistance. Finally, you can build your own regulator around that LTC6652 with superior performance.

>>1145594
Right. I saw the "at 1kHz" stat, and while I'm not sure what frequency has to do with a DC voltage anything, I guess it's a decent ballpark for the actual DC figure. Shame it isn't an explicitly stated variable I can sort by on Mouser, because it's by far the most important spec I need to minimise.

>build your own regulator around that LTC6652
Like pic related?

>>1145600
It is just traditional to measure the output resistance at 1kHz, partly because it is much easier to measure it that way. DC output resistance is generally lower than the 1kHz value.

>explicitly stated variable I can sort by on Mouser
They list the load regulation value, though. Too bad the listing format isn't consistent.

>because it's by far the most important spec I need to minimise
That's kinda difficult to believe. Anyway, you could try checking the regulators supporting remote sensing, like ON Semi's CS5253B-8.

>Like pic related?
That's one way if you don't need much current, but I was thinking something bit fancier. And you should use remote sensing, if possible.

>>1145606
Thanks for that, I just needed to confirm that regulators have less resistance than references, which I need because I might be measuring the resistance of things potentially as low as 0.1Ω with a reference/current limiting resistor and an ADC. Since I've got 4 digits of information, I need the resistance though the voltage reference to be significantly low as to not change this, or at least be constant enough that I can compensate for it in the code.

Nothing but this regulator's resistance matters, apart from the voltage being in the range of 5V and maximum current output being >0.2A. The accuracy and temperature coefficient don't matter because the regulator's output potential is put directly across the ADC's ground and reference point, so no matter what the voltage is the fractional voltage across the reference resistor remains constant.

Is it a bad idea to use SPDT relay for disconnecting battery when there is wall adapter plugged in?

>>1145774
No, it's no more abrupt than switching it off at the wall.

You could try to use a DPDT relay to switch the mains power within the wall adapter off instead, it will lose less power that way buy you'd have to isolate your 9V wires pretty well.

Might I ask the purpose of your circuit? It looks like you might be able to use a latching relay instead.

>>1145774
Nothing wrong with that but a little increased power consumption when plugged in

>>1145781
I think the purpose is to save the batteries when not needed and to keep the 12v from them when plugged in

>>1145774
>>1145796
>Nothing wrong with that
After a cup of coffee, I take that back
The relay coil is in series with the load, it should be strait across the 12V. (Move the bottom side of the coil in the pic to the other side of the resistor)

Wired like that it will buzz.
-relay activates
-coil pulls contacts closed
-contacts short coil
-coil deactivates
-contacts open
-relay activates...

>>1145781
>>1145796

So the thing I am doing is a switchbox for Arduino, where I could choose between usb and wall adapter and also have a multimeter in that same box too, so if I wanted to measure current there would be no need for extra leads.
The reason for battery would be to have the box also work also as independent battery powered multimeter if needed.

>>1145806
I drew the circuit diagram wrong. Relay should disconnect the battery when power plug is in and close the another contact.

Here is another rough sketch. Still going to need to figure out how to switch between the power going to ammeter.

I've been thinking about analog computation, and I'm wondering if there is a simpler way of performing voltage division using op amps.

so, for example, lets say I have two inputs that I want to divide, both are just DC voltages. The only way I can think to accomplish this would be pic related. Seems a little convoluted though.

I know log and exponential amps aren't very common configurations, but they do exist.
https://en.wikipedia.org/wiki/Operational_amplifier_applications#Exponential_output

>>1145967
I swear I posted pic

sry for the bad quality img. i need help with this one, the circuit works BUT not with the motor that i want to use. it works because i can measure voltage on the furthest 8550 bjt, around 3-4volts. i suspect that the current is inadequate.

is there a fun starter kit to test the waters with this hobby?

>>1145967
If you drop the requirement for opamp-only solution, then you can use either a ready-made division chip or a multiplier and an opamp.

And Google gave me this: http://www.edn.com/design/analog/4317160/Analog-divider-uses-few-components

>>1146043
Kinda depends on your idea of fun. If you play guitar, maybe a pedal?

>>1145980

there's something seriously wrong with that circuit. you need to add a resistor, say around 500-1000 ohms between the base of the 2nd PNP and the collector of the NPN, coz otherwise, you're essentially shorting the battery thru 2 diode drops.

i believe in a proper sziklai pair, the motor would be connected to the emitter of the NPN and collector of the PNP. that would also work to stop the short-circuit. and you wouldnt need any extra resistor.

>s there a fun starter kit to test the waters with this hobby?

not a kit, but this seems like a fun project you can emulate:

"Latest sperm collection technology @ China International Medical Equipment Fair 2011"
https://www.youtube.com/watch?v=OjaWysfHyUY

>>1146119
>China International Medical Equipment

>chops off your dick
>autocircumcize
>skin stripper

>>1146101
Ready made solutions are probably the way to go. I just wanted to get a sense of how computation was done in ye olde times.

Pic is interesting, although I suppose the output is limited to 5V. I'll save that one, it could be useful for some kind of microcontroller related project.

>>1146119
>>1146171

>>1145852
It's still going to buzz when plugged in. The coil is in parallel with the contacts.
Is the power in 5 or 12 volts?
The output is 5V, right?
Are you trying to measure current or voltage?
If current, 220 ohms is going to be a problem at 5 volts due to the voltage drop.

Using a ground symbol can simplify and clarify schematics

It's St. Patrick's day and I don't know if I'll be here later or tomorrow, so here's this for both 5 & 12 V.
Meters thrown in in multiple places cause I'm not sure what you're measuring

>>1146178
Afaik all the analog divider ICs and the most of the analog multiplier ICs are based on adding/subtracting logarithms, like your pic suggested. Some multipliers use different methods, though.

>>1146119
Unnecessary. Everyone knows Anon collects sperm using his mouth.

Is there a rule of thumb for what kind of resistor I have to use to limit the current going into an opto-isolator? What value from the spec sheet do I need to check to find the necessary value to calculate the resistor?

>>1146317
It's basically an LED and a transistor

For the LED, look up Forward Voltage(Vf) and forward current(If)
Subtract the Vf from the voltage you're using to find the resistor voltage(Vr).
Current is the same through R and LED, so, Ir=If

Use ohms law to find resistor value
R = Vr/I r
Round R up to an avalable value

Formula: R = (Vapplied - Vforward) / Iforward

Basically the same thing for the transistor.
lookup Vce at saturation, subtract it from V+ to find Vr. Choose resistor to limit current to less than max collector current (Ic). This is a minimum value an can be increased if you dont need to supply that much current.
If you don't need the max current then disregard Vce and just estimate the minimum R = Vapplied / Ic

>>1146327
>max collector current (Ic)
No. You start from the minimum expected LED current and multiply it by the minimum applicable current transfer ratio (CTR). Then you add a reasonable safety factor to account CTR loss due to aging, for example halve the expected Ic. And if/when you're using the transistor as a saturated switch, you should overdrive it, say, 5 times.

I have an FPC connector that loosens itself due to vibrations. Any ideas on fixing that?

>>1146400
silastic

>>1146419
looks good, cheers

I clumsily fucked up a 47uF 25V electrolytic capacitor from a kit. Can i replace it with a 47uF 35 or 50V one? Those are the only ones i've got lying around. Generally, can electrolytic capacitors be replaced with ones with higher voltage rating?

>>1146400
>>1146419
>>1146421
A bit of Kapton over the connector would probably work, too.

>>1146429
>Generally, can electrolytic capacitors be replaced with ones with higher voltage rating?
Generally, yes.

>>1146443
Thanks, can throw it together now. It's just a buck/boost converter so it shouldn't be that critical i hope.

>tfw you burn a 555
ayyy

>>1146470
you did it

>try to use regulator on chinese knockoff arduino nano
>fries immediately in datasheet compliant circuit
that was my first experience with chink parts

Complete ignorant on the issue

What's a TIP and whats the difference between a 30 and a 32 and what does the C stands for?

>>1146481
TIP30/32 are part numbers for a particular PNP switching transistor. the former appears to be able to handle more current (3A vs 1A continuous, will need heatsinking at that)

here's datasheets for each:
https://www.egr.msu.edu/eceshop/Parts_Inventory/datasheets/TIP32.pdf
http://www.mouser.com/ds/2/68/tip30-12170.pdf

should i be able to get decent yield soldering these with paste and a butane iron with the a wide hot air tip?

>>1146481
Congrats on nearly incomprehensible question.
TIP32C (etc.) is a transistor, originally made by Texas Instruments (TI). TIP is just a part of the type designator, the difference between TIP30 and TIP32 is the current rating and the letter at the end specifies the voltage rating.

Download the datasheet for more info.

>>1145267
could also be a varistor/fuse combo with a monitor lead. Can we see the other side? Fuse usually looks like a small bulge...

>>1146486
paste and a hotplate, yes. Iron, no.

>>1146486
Oh, butane with an air tip.

I'd say no--- there's no regulating temp, and those burn *hot*. Anything over 300-350C and you're going to be burning the FR4. Even controlled IR or air will burn the PCB slightly, still takes 100-300 seconds to get up to temp.

>>1145518
Okay, thanks.

I want to build an eight-key touch keyboard with a size of about 38cm x 10 cm, but instead of just on/off, I need to be able to know the number of fingers touching each key. I figure the "easiest" way to do this would be if I could just buy a multitouch touchpad that supports lots of touches, and just figure out what keys are touched from where the fingers are, but I have no idea where I would find/build one. Also, I'd like the device to expose a standard touchscreen interface to the operating system.

I have seen the guide for making the mt-mini, but I'm unsure of whether the driver it provides exposes a touchscreen interface (with absolute positioning and tracking of individual blobs). I also don't know if I could make a device as large as I want in a compact space, though I guess I could set aside some room for the device if I really needed to.

Electronics are not my forte but I have some familiarity
Got an bug zapper that doesn't have enough strength to it (runs on 2 AA's). Guts seem simple enough, how could I work up/approximate what kind of volts/amps this thing can safely handle?

>>1144916
how about youtubers like https://www.youtube.com/user/bigclivedotcom and
https://www.youtube.com/user/julius256

>>1146920
flip it over. if the circuit's dumb enough increasing the input voltage will increase the output voltage directly.

the limiting factor is the voltage rating of the capacitor. to determine how much voltage you can provide at the battery terminals, you need to know the current voltage stored in that capacitor, which holds the charge until it gets dumped through the bug. that voltage may be too high to measure with a shitty multimeter, and cheaper meters may put too much load on the circuit to measure accurately as well. the only way to get a rough idea then is to measure the turns ratio of the transformer. you could get close enough with an ohmmeter if the resistance isn't too low. report back if you have a multimeter and it works.

>>1146933
the first post is at the character limit
That's why the link to the last thread is in the first reply

>>1146935
The LED and transistor could also be hard to put more voltage through, though changing the dropper resistors to a higher value should do the trick. Bonus points if you can just use pic related. If it's particularly badly designed the wattage ratings of the resistors could be a problem, but I doubt that. If you increase the voltage too too much you might run into trouble with the transformer secondary's insulation, so if you do plan on building a stun gun, I'd build the circuit from scratch.

Is it possible to make a very rough and ready microwave frequency (10-20GHz) antenna without having to resort to striplines and waveguides?

And can I use off the shelf transistors and passives in a heterodyning mixer to divide down said high frequency content to something I can actually work with or do I need very specialized active components and microstrip LC filters?

>>1147007
But microwave antennas based on waveguides can be very simple. Cantenna is probably the best known example. Why do you need to avoid waveguides and striplines?

A typical receiver in the 10-20GHz range starts with a schottky diode mixer. Suitable diodes aren't that difficult to get.

>>1147041
I just wanted to make up a prototype circuit quick with parts I had on hand.

I don't have a bunch of RF diodes and waveguides just lying around and I don't want to have to wait a week and pay out the ass in shipping for a handful of components.

>>1147050
You aren't going to build a working 10GHz receiver with that attitude.
Ignoring that, how are you going to debug or test it?

>>1147069
I don't mean to be like that. It's just a bit inconvenient that's all.

I don't have a reliable means of testing it as even the cheapest spectrum analyzers are too expensive for me. However I do have a reliable 24GHz signal source and I want to preform some measurements of its output power hence why I need a receiver capable of picking up a frequencies over 20GHz and heterodyning them down to a lower frequency that's actually within the bandwidth range of one of my oscilloscopes.

>>1145261
I would think a fuse is unnecessary. The fuse will only protect the source (battery) from over current in case of a short. It won't do much for the sensor because if the fuse blows the sensor will probably be fried. Damage to semiconductors due to over voltage is pretty much instantaneous

>>1147075
You could try using it as a wifi antenna to test it, assuming it's in the right frequency range, then just run some packet-loss tests on pings from one device to another.

>>1146327
>>1146362
Thanks guys, appreciate ya

>>1147007
I doubt you're going to throw together some 1N4001s and all of a sudden get a microwave mixer, dude.

>Switch node power supply radiating like crazy in AM band in radiated emissions testing
>Replace off-brand elec-cap with Panasonic low ESR elec-cap found on random shelf in lab
>Noise completely removed

This is what I deserve for using cheap chinese shit parts in design.

Doing some RF-type work, except, roughly estimating, we're expected to deliver 100+W to a load at 1000Vpp at 1MHz... and yeah, it's supposed to be sinusoidal... and tunable.... in voltage, frequency, AND OH YEAH THE LOAD IS FUCKING VARIABLE BY MORE THAN 10x. End up with one circuit after another, except there's like fucking ridiculous volt spikes across shit. The first one had up to 20kV across some components. I was able to get another down to less than 10kV. This second one actually had 40 fucking goddamn kilovolts across an inductor, but I was able to bring it down with a trade-off in bandwidth and efficiency.

>>1147266
>1 MHz
>RF

Sounds pretty interesting though.

>>1147267
1MHz is AM, still radio
>mfw I try to find shit that works, only for it to A) not work nearly as ideally as a paper said it would B) it only is for 50ohms at like 100V or C) has a trillion parts or is for like 20GHz 30kW

>>1147266

Sounds like you'll wind up with something similar to those big tuners the HF guys use to impedance match their antennas across a wide frequency band.

>>1147268
I'm experimenting with audio solutions before I go into diy radio, but when I do I can tell it'll be a hell of a lot of my theories and simulations being useless.

Do you guys know a more accurate PWM board? I have this, but the output PWM frequency is off by several percent, and I need precise control of the frequency for my project, not only duty cycle.

>>1147273

Either use a micro or a DDS chip, you want something using a crystal clock.

>>1147273
What kind of application does PWM frequency matter more than the duty cycle? Are you sure you don't just want a square-wave generator of some kind?

>>1147280
These: http://www.bluerobotics.com/store/thrusters/besc-30-r1/

They are a bit weird, you control them with the duration of the positive pulse.

"Max Reverse: 1100 μs
Stopped: 1500 μs
Max Forward: 1900 μs
Deadband: 1475-1525 μs"

>>1147075
You don't need a mixer to measure output power. A microwave equivalent of a crystal receiver would do. While it is much simpler and easier to build than a superhet, you still need a diode which works reasonably well at 24GHz.
Another option would be heat-based power detection, where the microwave power heats a thermistor. It still needs to be properly built, though.

Try begging some nearby university lab to demonstrate their stuff.

>>1147284

Just do it with a microcontroller.

>>1147284

AFAICS frequency is completely irrelevant to this thing. It simply latches the duration of the last pulse as it's control value, the frequency just has to be low enough to allow >5 us between pulses.

>>1147299
Correct, frequency is only important if I actually control it with PWM. At the moment I control them by first calculating the pulse width (1 to 2^12) matching the required width in microseconds. And that cannot be done without knowing the precise frequency.

Using the PWM board is convenient as it has a bunch of outputs (I'm running 8 thrusters), and I'm also hooking up a few other devices that use good old duty cycle PWM.

I just need this to be reliable and fairly simple. I'm not sure programming an MCU for it is the best method.

>>1147295
Is right, it's probably the most cost effective method as long as you can get one with an accurate crystal inside, but I'd personally make my own PWM board rather than trying to use some purpose-built board.

But try modifying the current board. If you can draw a diagram of the PWM board, then seperate the frequency generator part of it, you could probably replace it with an equivalent crystal circuit.

>>1144916
If there are pcb design rules for high frequency signals are there rules for regular pcb designs?
Can I fuck my pcb designs up as long as the schematic is correct?

>>1147310

yeah, there's lots and lots of rules, but it's all common sense stuff, like using wide traces when you're carrying a lot of current. a typical PCB software program will include a DRC option (https://en.wikipedia.org/wiki/Design_rule_checking)) so you can check if you're breaking any fundamental rules (thus possibly creating a mini black-hole on you PCB)

>>1147007

Doing a 32 step sequencer. The datasheet (where I just copied the reference design) of 4017 says I can use 9 outputs from the first, but I only want 8 so I moved the input to the first AND gate from Q9 to Q8. Is this correct? Also: what should be done to the Q0 of the last IC in the cascade? Should it be tied high/low or left floating?

Pic related, I recently impulse bought one of these chips. I'd like to make an LED blinker out of it.

I thought it would be easy to assemble some kind of interface circuit to program the device via standard PC serial port. For example, an AVR chip that I am currently using needs just one transistor and a couple of diodes and resistors to interface with a COM port. Then I can use avrdude to transfer the compiled programs into the controller chip.

Is there anything similar for these TI microcontrollers?

>>1147363
>Is this correct?

yes.

>Should it be tied high/low or left floating?

it's an output so if you tied it to either rail, you'd be shorting it out. you could pull it high/low with a resistor, but that accomplishes nothing, so just let it float.

small grumble: dont save schematics in JPG coz they end up fuzzy and hard to read. GIF makes them both smaller and lossless. i lost all respect for NASA when i learned they use JPG.

>>1147397
No, unless you have a bootloader on it, it needs a JTAG programmer
You can probably try one of the cink ones from ebay
Although i recommend getting a msp430 launchpad, it's only 10$ you can use it to get familiar with the architecture and then even use it to program the target chip

>>1147310
Read Bogatin's EMC book

>>1147310
>>1147320
in my experience (only low frequency domain stuff, up to 16MHz micros) the DRC (and ERC for that matter) is a good thing, but you got to recognize that you're the one setting up the rule set.
even if you have the tolerance listing from your manufacturer you can produce a layout that will not break the rules, but still produce enough noise/crosstalk/stray-capacitance to generate glitches you wont be able to figure out without a emc black belt sorcerer to hold your hand.
if you are your own "manufacturer" then you still have to account for how good you are with your /diy/ skill and the tech you're using: can you produce the trace width that you intend to use, can you drill the holes/vias that you are using in your layout, can you etch those sneaky isolations between pads, can you actually solder those snugly fitted smd components etc. etc.

it's pretty much spot on what the other anon said: common sense. and if that's not enough then just post here about particular concerns. we're friendly for the most part ;)

>>1147454
Thanks for your information. Just one more question. Can those equipment (JTAG programmer/Launchpad) be used with Linux or are they generally Windows-only?

>>1147544
i'm pretty sure jtag programmers have drivers for linux available (check your distro for packages whatever package manager you use) or alternatively use a virtual machine and give it access to the required ports.

>>1147397
>µcontroller
>not Atmel
Have fun figuring that one out. TI should have documentation on programming them, and I'm sure there are tutorials out there, but after a quick squizzle on the hivemind I can't find anything. Your best bet is to try the launchpad circuits and see if you can hack your model microcontroller into the space for theirs, program it with that, and then remove it again. It better be cheap enough to warrant such an impulse buy.

>>1147621
>not Atmel
I think you mean Microchip mr. caveman. Get with the times, there is no Atmel anymore.

>>1147627
Oh.

>>1147627
WAT
Is this pretty much a monopoly now besides the some TI stuff?

>>1147689
you can go for arm, lots of stuff to choose from.

Hi /ohm/,
I've always liked old minicomputers, even stuff like the altair 8800, things with nothing more than a faceplate, some switches and a cpu.
I've been meaning to learn the low level stuff for ages, but haven't been able to get my hands on old hardware.
Now I came across this little thing, an OSI 300 computer "trainer". Basically the simplest of simple machines, designed specifically to teach people how to use minicomputers.
Anyway, someone made a small batch of reproduction units at a con a while back and was, for a while, selling them, he has a blog here:
http://randomvariations.com/category/osi-300-trainer/
But his blog's gone dead for a few years and I can't find any way to contact him.
I doubt he'd have many if any units left to sell, but at very least I'd like to get a hold of the Gerbers/schematics. All I can find is a PDF and, whilst fun, I'd rather not have to redo the entire schematic and board myself if I don't have to.
Anyone think they can help? Anyone recognize this board from anywhere?
Thanks.

>>1147795
Come on, anon, it's a one page schematic. Build it on perfboard, if making a PCB is too much work.
I mean, if you're masochistic enough to enter software one bit at a time, little bit of soldering shouldn't be a big deal.

>>1147689
Maybe from hobbyist point of view, but Microchip isn't nowhere near the only player in the 8b field.

Anyone have experience with Deoxit? Worth the price?

>>1147798
I tried learning shift registers on a breadboard once, the connections where so intermittent it was nearly unusable. Now, no matter the project size, I do it soldered. And I don't wanna know what a dead bug 6502 system with a cap and diode for a clock looks like.
Now that I type that it sounds super fun, but still. Start with something that works, then rebuild from scratch.

>>1147795
Ooh, a digital logic trainer! Got a schematic for that? I'd love to build my own and you should too, it's pretty rewarding.

>>1147804
I learnt shift registers on a breadboard with no trouble, and there's nothing particularly complicated. They're essentially arrays of d-flip flops chained together under one clock and an input enable. You can either load them all at once by enabling the parallel inputs when you clock it, or feed a series of bits in the side. An interesting application of them is since each successive bit can be treated as the significant digits of a binary number, by shifting the whole set of data to one side or the other, you can double or halve the number stored easily. I can recommend a textbook on basic digital logic if you're interested, but I never read it because I played too much minecraft to need to learn anything from my lectures.

Digital Fundamentals
Thomas L. Floyd

4 bucks on Amazon for 9th edition, and it's pretty extensive. Covers all from binary to logic gates through latches and shift registers, multiplexing, timing diagrams, Karnaugh maps, ADCs, ROM, PLAs and PALs, all sorts of good stuff it you're willing to sift through 800+ pages.

>>1147825
This better be a good .gif.

>>1147825
>Got a schematic for that?
Yeah, it's here: http://randomvariations.com/wp-content/uploads/2015/08/minitrainer.pdf
Right now I'm just ordering up all the parts and will probably bite the bullet and breadboard it.
Also, time to get a bigger breadboard.
>with no trouble, and there's nothing particularly complicated
It's not that, it's just that the buttons where bouncing like crazy
Still, now I feel like giving it a second shot, proper switches and breadboards this time.
>Digital Fundamentals
>Thomas L. Floyd
Sounds great, I'll pick it up now!
Thanks anon, you've been a real help.

>>1147830
I paid at least $150 for the textbook, that's before I realised the online price difference was that fucking big.

>>1147828
Turns out GIMP is shit at importing PDFs and retaining their quality.

I learnt at my labs using pic related, and they were pretty solid with few exceptions. Pop in a breadboard, slot in your logic gate ICs, and wire them up. They're great because the switches and indicators don't take up any room on the breadboard, and sometimes it really becomes wire spaghetti. Used it with standard 74LS series ICs. I'm not sure what a modern equivalent is like, but searching around it shouldn't be too hard to find something.

>>1147834
>Turns out GIMP is shit at importing PDFs and retaining their quality.

who would have thought that you loose quality when converting a vector format to a lowres raster format

>>1147860
>>1147430
The file has already lost it's resolution in GIMP, the .gif is the same resolution as it is in GIMP.

JUST LOOK AT THE PDF'S NIGGAS
http://randomvariations.com/wp-content/uploads/2015/08/minitrainer.pdf
http://web.archive.org/web/20150826142320/http://www.hansotten.com/uploads/6502/trainer.pdf

>>1147867
Now I just wanna see a png next to the gifs.
AND IF YOU WANNA BITMAP A PDF IMPORT IT INTO INKSCAPE, NOT GIMP

>>1147869
>Completely different schematics
No fun.

Salvage time

big cap is 2200 uF at 400V

humungous choke (tube amp?)

lots of high current semiconductors, also between the pcb and heatsink (as big as the pcb)

Lots of huge non polarized caps.

No idea what the transformer looking things are

nice bunch of 4n7 non polarized caps, some shunts, a relay, probably some nice ICs

I was told to go here to ask about wattage and ohm ratings on audio equipment. How do I know what amp wattage to go for to drive a certain speaker if I know that speakers ohm and watt ratings?

>>1147913

simple version: look how many watts your speakers are (don't believe the gazillion watts you'll find for most car audio) and make sure you amp can deliver less than that.

what speakers are we talking about?

>>1145774
>>1145852
>>1146246
I decided to ditch fritzing and move to Eagle. Here is the new schematic I have been working on. Any comments/suggestions? I tried to be as clear as possible with it.
I decided to use DPDT relays to control what the ammeter will be measuring, and for disconnecting the battery, a SPNC relay. The diode after it is for preventing voltage feedback for regulator. For every relay I decided to add feedback diodes to prevent excessive voltage from the coil. There is capacitors for both of the power supply lines for storing charge between the relay lag. For the USB data cables I will use 2 -way shielded wire.

Eagle questions: How do I add shielded wire symbol to schematic? For ammeter I just drew circle with text in it.

>>1147920
sats from a panasonic system, receiver just died and I don't really want to dig around in printboards to try and fix it

it says 100W DIN(???) 8 ohms on the back, specs for the receiver are 100W 6 ohms according to the manual

I'm guessing higher resistance means less power going in? is it linear? so it'd be 1.33 times less? 75W

I'm also not sure what ohms amps are usually specified in, some say 4. so I should look at twice the wattage?

I also don't know what the relationship between watts and volume is, the volume control on the rec was in minus decibels from maximum, since I never set it to 0dB, 75W is obviously more than I need, but how would I know the wattage to go for that would let me achieve at least the volume I usually set it to before?

>>1147932
does it say anything about watt/dB? that is what you are looking for.

but generally speaking anything above 20 watt is enough to get the cops called on you. the " specs" on most consumer autdio equipment are 500% inflated for marketing purposes

>>1147883
So I spend maybe thrity minutes disassembling that thing and the big cap is out, the choke and the left transformer (no idea on specs yet) are out. The right transformer seems to be a choke too.

from under the PCB I got some 500V 53A mosfets (E53NA50) which is nice. SSTC anyone?

>>1147950
no sorry I can't find anything about that

I did find a formula for calculating how many dB you lose with a different wattage amp, and like you say, even the most pessimistic calculation only loses me about 5-10 dB

I still don't know how the ohms play into it but I guess it's fine so long as the amps range supports whatever the speakers are?

thanks for your help anyway

Does anyone have any experience with the MCP23018? Why is the ACK bit so damn weak? I can't find anything in the datasheet about this.

It's <1/10th of Vcc. With a 1k pull-up it's about 300mV and with a 4.7k it peaks at 350mV.

>>1147926
K2 appears to serve no purpose
-unactivated it connects arduino 5V to PC 5V
-activated, it just adds the fuse (unless I'm missing something)
(similar with K1)

Ammeter refers to a current meter, The way it's drawn it's measuring voltage. "Voltmeter" would be a better term

The 5V regulator does nothing but drive 2 coils in series.

If it's supposed to supply 5V without the PC (don't know if this is right), The regulated 5V shouldnt go through anything in series after the regulator because it will drop voltage. Ground pin 3, move diode to go between pins 2 & 1

>>1147960
I don't understand the problem. I2C acknowledge bit is supposed to be active low.

>>1147978
I know that, but it should be a lot closer to 0mV shouldn't it?

On some occasions, it's detected as a NACK because it's too close to 400mV.

>>1147982
Usually yes, but the datasheet guarantees only that it is under 800mV.
You could try checking if there's an extra pull-up somewhere, for example by pulling the SDA pin downwards with a 1k resistor to ground.

>>1147982
>>1148002
Yeah, you might have contention on the bus. With a larger (weaker) pullup you should have a lower low level, not higher.

>>1148002
>>1148010
Right, that's the strangest part to me. I'm going to disconnect all but the master, pull-ups and the MCP23018 from the bus.

In Eagle there are library parts which have supply/ground pins named internally as VCC/GND. That makes sense to me and I'm used to using those supply symbols/names. Now, there are also parts which have them named VDD/VSS correspondingly. I only noticed this now when I started doing layout and didn't see an air-wire to my supply network.

I got round that issue by doing this god fucking awful zero ohm resistor hack on pic related. Can't connect VCC directly with VDD since the resulting net name has to be either one of them and that messes up either the VCC or VDD air-wires.

Is there a more sane way of achieving uniformity in the names?

>>1147430
Thanks for the .gif tip anon. Being on mobile 4G connection this really helps me keep the data-cap jew silent for some time.

>>1147804
> I tried learning shift registers on a breadboard once, the connections where so intermittent it was nearly unusable
Sounds like TTL (7400 series).
CMOS (4000 series) or HIGH SPEED CMOS(74HC) is a lot more forgiving for breadboarding. They're less sensitive to un-ideal connections, power supply noise, and spaghetti wiring.

>>1147959
"power (Watts) = current (Amps) ^2 * resistance(Ohms)"
Don't put so much current into the speakers that the power is higher than the power rating of the driver. The power rating of the amplifier is just the maximum power it can output, if you put 40Ω drivers on it it wouldn't be outputting nearly that much power. If your amplifier has a maximum voltage value then it's easy to calculate with "power = current * voltage", "voltage = current * resistance", and other derivatives of such, but I'm not sure if they do.

https://en.wikipedia.org/wiki/Gain_(electronics)
Stare at the wikipedia page for electronic gain for a few hours to get the hang of converting decibels to volts if necessary, because the voltage gain (V_out/V_in) multiplied by the voltage in equals the voltage out, if that wasn't obviously tautological. Decibels add instead of multiplying, so putting a -30 dB signal into a 50 dB amp will output 20 dB, and putting 40 dB signal into the same 50 dB amp will make it 90 dB. Of course a 50 dB amplifier will have variable gain/volume/both, and will probably hardly ever go up to eleven, but it's probably a good rule of thumb to keep your amplifier's maximum power output at 10% or so below the maximum power input of the speaker. That way you aren't in any danger of blowing the speakers to pieces with a wrong dial turn, and you won't be trying to change the volume by turning the dial a couple of degrees at a time.

I imagine you can just shove an extra resistor/impedance operator in series with each speaker to add up to the needed resistance to keep your power output in the save zone, and this is effectively putting a permanent volume decrease on it, but I'm no audio specialist and it's not really an ideal practice. Putting an impedance operator on the input would make more sense.

An impedance operator is a capacitor, inductor, and/or resistor. Memristors are for hacks.

>In Figure 1.52A the left-hand side of inductor L is alternately switched between a dc input voltage V in and ground, at some rapid rate, spending equal times connected to each (a “50% duty cycle”). But the defining equation V =L dI/dt requires that the average voltage across an inductor must be zero, otherwise the magnitude of its average current is rising without limit. (This is sometimes called the volt-second balance rule.) From this it follows that the average output voltage is half the input voltage (make sure you understand why).

I don't understand why it's half.
Also, I'm a Comp E student and have only taken two EE classes. Basically the extent of my knowledge is passive circuit analysis/filters, Laplace transforms, and some op-amps. I wanted to read the Art of Electronics to get a more practical perspective, because even though I do fine in the classes, sometimes I only feel like I know how to solve math problems. Should I continue with AoE, or pick something else up?

ok /ohm/ I am wanting to make card reader with arduino. Now, I am guessing a lot of card readers have the same concept or similar concept, so, I guess I cant screw up too hard on picking one.

with
>pic related

I also came across this
>http://www.ebay.com/itm/Credit-Card-Reader-Magnetic-IC-3-Tracks-Magstripe-POS-Encoder-KDR-1918/191831190046?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D2%26asc%3D41670%26meid%3Db877c96d0b9d40ffa914eef6318b6d20%26pid%3D100005%26rk%3D2%26rkt%3D6%26sd%3D201853257563

for 15 bucks or so.

Would the ebay listing I have work, and be compliant or will it be stubborn and not worth my time? I read that to ignore usb card readers when using arduino.

Anyone that have done something like this please let me know.

>>1148320
Honestly I don't see why people recommend AoE to beginners, it's a book written for physics majors and the passage you've called out is a very handwavey-but-rigorous-in-an-energy-conservation-way explanation that would be insightful for physics students but not for anyone else.

I would suggest putting the circuit into a simulator and doing mental calculus on the voltage/current traces until you can explain what's going on.

>>1148417
We recommend it for advanced readers, though.

>>1148320
If you start from presented equation (V =L dI/dt) and assume steady state operation (= average current is constant), you can think it this way:
- for the average current to remain constant, it must mean that the current through the inductor decreases just as much during the switch=gnd phase as it increased during the switch=Vin phase.
- Since the both times are equal (50% duty), abs(dI/dt) must be the same for the both phases
- Thus, the absolute value of voltage across the inductor must be the same in both phases.
- This is possible only when Vout is half of Vin.

An even simpler way to look at it would be that L/C2 forms a low pass filter and if the switching frequency is much higher than the cut-off frequency, then the output voltage will be the averaged input voltage. Since the input voltage is 50% of time 0V and 50% of time Vin, your output voltage will be 50% of Vin.

Or if you won't get it without sufficient amount of math, you can use the steady state assumption (average inductor current is constant) to write the equations for both switch states.

Simulator might help too, like the other anon suggested.

As a general rule, AoE uses very little math and relies more on intuition. If your intuition is not compatible with AoE, then shit might suck and you'd be better off with some more conventionally written book.

>>1146920
Add a capacitor in parallel with the existing one to add more power, I used a big ass capacitor and now everything that comes in contact with it bursts into pieces.
The voltage output is around 2kV, changing the circuit or the voltage output won't do much, since the "zap" comes from the capacitor.

>>1148473
i did that too some a long time ago.
i kept going until i blew the switching transistor.

>>1148475
Anyone got a circuit diagram of this? Sounds like a fun project.

>>1148477
i just modded an electric flyswatter.
there is probably a schem on the web. its a pretty simple step up converter.

>>1148477
Just buy and solder a beefy 2kV rated capacitor in parallel with the existing one (the red thing at the left), that will increase the capacitance and the "zap"
>>1148475
I did it twice and they're still working fine, I guess a BD679 can replace the existing transistor in case of failure.

>>1148508
Pic related, my bug holocauster.

>>1148572
How big is that cap? Does it have to be bipolar?

So I've tried using LT-Spice on macOS and it felt really unintuitive. iCircuit looks a little incomplete and lacking, are there other alternatives for mac, or is learning LT-Spice worth putting in time for?

no answer in the tube thread so ill ask here.
any good diy solutions for variable HV supplies.
gonna use it for tubes so dont need anything in the KV region.
6-800 is about max i need.

the obvious is a stepup up transformer with a variac, then just rectification smoothing + dummy load, fuse etc.

but i had anther idea:

>>1148811
>for a HV supply, could i use a triac to control the secondary of a transformer before rectification and smoothing?
>I'd probably need some fast bipoars as well as a big 'ol back of electrolytics. but it seems doable in my head.
>I only have one variac and I'd like to keep it free.
>besides, a variac AND the big transformer would make it really heavy.

the triac would need to handle 4-500?vac, but that should be doable, right?

>>1148858
you could explore this concept. it's easy to get a high voltage fet that can handle 10 watts. maybe you could salvage a microwave oven transformer and tap it off at a lower voltage, or just rewind it.

i didn't do any math on the numbers shown.

>>1148879
I got an adequate transformer, so no need for that.
mot is too high anyway.

but what exactly am i looking at here?
zero detection then delay before engaging nmos?
why not a triac in the instead?
this feedbackloop regulates or..?

i need to de able to dial voltage, but it doesn't need to be all that exact really.

might be a bit too complicated for a problem with such a simple solution.
I guess i have to be on the lookout for another variac :D

thanks for input.

>>1148886
the fet acts as a discrete linear regulator to minimize the number of fat passives you need to smooth your output. the amps modulate the gate voltage to keep it in the linear region. if you don't need a large amount of power (say <250mA at in an 800-600 range) the fet will work fine with adequate heatsinking. maybe salvage a cooler from a shitty desktop, paste the fet onto the pad, and run its fan off the aux rail.

>>1148890
cool mang.

any name for this configuration so i an read up on it and understand better?

thanks again.

>>1148890
>linear regulator
wait.
so if a draw 100mA regulated to 400v. it would drop the remaining 400v at 100mA and eat 40W?

yeah, i'll go for the variac or triac approach.

>>1148893
hence a salvaged cpu cooler. variacs are obviously the best option, and a triac regulated rail is going to be very dirty without large (expensive) passives. if you can live with that then that works.

>>1148858
The original "obvious setup" describes doing this and rectifying and filtering the output. No filtering should be needed between the triac and the transformer. The alternative as I see it is to rectify the transformer output first, then use DC PWM and then filtering instead. It's probably easier to get the correct rated parts. Twice the caps, but probably still easier.

What you describe in your "other idea" is to use PWM before the current enters the primary. Pic related shows a waveform before and after PWM, and long story short, a sinusoidal wave is the most efficient waveform to put into a transformer. It's got something to do with a differentiated sinusoidal waveform being a sinusoidal waveform, change in magnetic flux, etcetera.

But if it's going to be DC anyways, you might as well PWM it after rectification, it's probably less lossy since the transistors aren't passing through zero all the time, and you could probably do it without filtration caps before the PWM anyway. No need for bipolar caps either.

>>1148935
If you have an oscilloscope (better be a CRT) then you can change what the test load is, and alter the capacitance until your resultant DC voltage waveform is linear enough at that current drain. Don't want any 60Hz flickering tubes, and it's probably bad the tubes too.

>>1148935
What you're proposing seems like it would be more along the lines of Bit Angle Width modulation.
http://www.picbasic.co.uk/forum/showthread.php?t=7393

>>1148962
Incorrect. I'm suggesting he use PWM after the rectifier so that he can use a cheap FET or BJT instead of a triac. This is how most standard DC speed controllers work anyway. I am not suggesting that he uses some sort of digital bit-resolution-based modulation that requires a microcontroller.

The DC PWM setup can be done without any digital components, especially not a microcontroller, just google for a schematic and you'll find plenty of them. In fact it's almost certainly easier than making an AC PWM circuit. I'd personally just buy a cheap Chinese one off AliExpress, they usually use a fairly easy schematic to modify (to suit the needed higher voltages), excluding the mystery unmarked ICs. Might need more heat sinking though.

>>1148972
if you're getting into high voltage pwm there's no reason not to just make a flyback converter instead though

>>1148973
Oh yeah, that's a much better idea. Too bad he said he already had a good transformer.

Any of you have any idea on car batteries?

If so, please help on this

>>1148984

>>1148339
no one? I figured someone must've done a project similar to this.

>>1148893
Considering it is for tube shit and more specifically for tube amplifier shit, why do you care?

>>1148138
Thanks anon, I'll look up which chips I have

I'm not very knowledgeable on electronics, I'm just trying to put a tachometer on my lathe. What should I be searching for if I need something to take 120vac to 12vdc? Searching power supply just pulls up wall warts, but I need a little thing to go inside an enclosure.

>>1144918
Who /pirate radio/ here?

>>1149100
Are you just going to install a car tachometer on your lathe? As long as you expect similar RPMs out of it that's fine, but with a microcontroller and a hall-effect sensor or LED+photodiode you could make a digital one with customisable range and display.

When it comes to power supplies like that, you could just take apart a wall-wart and install its components inside your enclosure with an extension lead or IEC socket sticking out the back. Alternatively you could see about getting a second-hand computer PSU, buying a cheap converter from China, or just making one. If you do make one, then you'd better have an oscilloscope, otherwise making sure you're not getting a shit ton of noise will be problematic. A good idea is to check inside the unit to see if everything is running off the 12V, because if there's just a DC-DC converter and everything inside runs off 5 or 3.3V, then there's no point in getting a 12V supply in the first place. If you take an HD photo of the tachometer PCB I'm sure someone here will know what's going on.

>>1149110
I wish. Does attaching a solenoid to a 3.5mm mono jack so I can listen to music wirelessly with another solenoid attached to my earphones through an opamp count?

>>1149114
Nah, one of these. It's all 120vac inside. Might just go the tearing down a wall wart route.

>>1149117
I meant whether the tachometer ran off 12V or not. If you can't take it apart then don't worry. Just be careful about isolation and grounding when it comes to taking the shell off the wall-wart and putting it inside the metal lathe housing, which is what I assume you plan on doing. Beware of ground loops, if that even applies to a 60Hz inductive load. I'd recommend using a little project box instead if that's what you were planning, just for the added safety and not having to drill (as many) holes in the lathe. I assume you've got the installation of the tachometer's mechanical side down?

>>1149125
>I assume you've got the installation of the tachometer's mechanical side down?
You could say that.

The grey box where the controls are now is all plastic and plenty roomy, so that's where I'll put it all, then just somehow get the wires run to the tach. It's all wired up very nicely(courtesy of the electrical engineer I bought it from), just in a weird spot(courtesy of this being his first lathe), because the dude I got it from had to repair it after the couple he got it from on Craigslist fucked him over with a burnt out motor controller. https://www.amazon.com/gp/product/B00VKAT8A2 is the tachometer.

>>1148138
>>1149043
Shit, it was a 74Hc595.
Still, I'll order a 4000 and maybe put some tiny caps on the buttons to smooth out the input.
I mean, I COULD just learn on a circuit sim or hook it up to an arduino, input onto the arduino, then output to the shift reg, but I WANNA DO IT MANUALLY DAMNIT.

>>1148834
I don't recall it correctly, I think it was between 0.22uF and 1uF, I just asked for the biggest 2kv rated capacitor in the store.
>does it have to be bipolar
The output is around 2kV DC, so I guess using a polarized capacitor would not matter, non polarized capacitors are easier to find and come in a handy rectangular format.

>>1149153
i would think post polarized caps woudn't take the voltage.

>>1149133
>maybe put some tiny caps on the buttons to smooth out the input.
For a clock signal, you'll need a full fleged switch debouncer circuit.

I want to wire a buzzer up to trigger when a door opens in a machine.
The machine has a sensor that gives out 24V when the doors open and the buzzer needs 24V so that's all good.
However, if I wire it directly the buzzer will sound as long as the door is open. I just want it to sound for a few seconds and each time the door opens.

I was thinking of using a voltage divider to step it down to 12V and set up something with a 555 timer so the 555's output triggered a relay that connected the buzzer to the 24V.

I'm having difficulty figuring it out.

>>1149261
Wire the buzzer through a set of NC contacts
add a timing circuit to trigger the relay
24V->Diode->Resistor->capacitor+->12Vcoil+
Cap- and relay- to ground
Pick resistor to match coil resistance, so coil has 12V max
Ballpark for Capacitor: Time = RC. Decrease Cap for shorter time

Diode prevents 12V from being fed back to 24V signal when it turns off. Don't know if it's necessary.

>>1149153
>rectangular format
>not just using the biggest electrolytics you can find

>>1149261
You could probably use a big electrolytic capacitor in series with the buzzer, with a bleed-off resistor in parallel with the cap. The buzzer noise would fade out after an interval of time depending on the capacitance.

>>1149266
>>1149268
Cheers, didn't even think of just using a cap.
Putting a cap between the 24V and buzzer would probably annoy the hell out of me as the buzzer starts getting higher pitched when the capacitor is nearly full.
I just wired the cap to a relay coil and the buzzer through the relay so it cuts off a bit more clean.

I've struggled with soldering since day 1, I've only used dirt cheap irons and tips, is that most likely the issue?

>>1149434
Getting good results with shit irons requires more maintenance, better skills and often extra flux.

>>1149434
Flux-core lead solder is the way to go for a noob. Your iron doesn't matter that much as long as it doesn't take much time to melt when you're touching liquid solder on the iron to solder on the board/component. Making sure you've always got clean, non-oxidised solder on the iron is a good idea, keep some steel wool around to clean the tip. Touching a soldering iron with a little solder on it to a component will heat up that component much faster, the heat transfer rate is proportional to the contact area which liquid metal is pretty good at maximising. If your iron's tip doesn't "wet" with molten solder (sticks like it does to component leads instead of like it runs straight off aluminium) then you either need a new tip or a new iron if it needs replacing.

If you happen to be in Australasia and buy the tiny 15-20g solder-coil tubes from Jaycar, those things don't have flux in them, you have to buy the 200g roll instead. A pot of flux usually isn't needed for standard through-hole components and such.

If you're soldering SMD components, you'll want solder-flux paste and some sort of hold-down mechanism. If you're soldering particularly large/obscured SMD contacts then you'll want a heat blower.

When it comes to helping hands, moveable lighting, PCB vices, extraction fans, buy them when you feel you need them because they're in no way necessary for quality soldering. The light and fan make a nice soldering project anyway.

What have you been working on? Automotive?

>>1149460
>keep some steel wool around to clean the tip.
NO! GAAAAH!
Scraping the plating off the tip means you'll constantly have to work hard to clean it and your thermal transfer will always suck.
Never clean an iron tip with anything other than a wet sponge or brass wool.

>>1149128
aaaaawesome.

>>1149117
Ohai it's a Sieg

I built a small wind turbine, it produces up to 12 volts, I need to build a circuit that regulates the power to charge a cellphone(5volts, 1 amp[max]).
I know I could literally just use a voltage regulator, but if I wanted to improve the efficiency I was thinking maybe a boost-buck. However, IDK how to make a control circuit for that. Do any of you know where I can learn that, or have other ideas how to do this?

>>1149465
>However, IDK how to make a control circuit for that.
You don't have to guess. Buy a boost/buck controller IC, and the datasheet always has a recommended circuit for use.

>>1149465
>I need to build a circuit that regulates the power to charge a cellphone(5volts, 1 amp[max]).

Can you spare a dollar?

http://www.ebay.com/itm/6-24V-12V-24V-to-5V-3A-CAR-USB-Charger-Module-DC-Buck-step-down-Converter-/172398891801

>>1149472
that's buck-only. Like all the supercheapies. He needs a buck/boost.

>>1149462
>scraping
>work hard
Nah, it's just a quick swipe to skim off the oxide layer/burnt flux/solder stalactite. It's less messy than just flicking the iron at the floor, and burns less holes in your shoes.

>Scraping the plating off the tip
Plating? You mean the solid plating on the tip or the solder itself? If it's the solid tip plating then shitty steel wool is hardly too much harder than brass.
>brass wool
I'm pretty sure brass is wetted by solder, is it the idea to have solder stick to the brass? Sounds like a copper solder wick to me.

>>1149465
https://www.amazon.com/dp/B00SM3UALQ
https://www.amazon.com/dp/B00BWF5U0M
20 bucks and you get 5 amps for 4 devices. just snip off the male end and do red to red black to black and short white and green and there you go. just make sure you set it to 5-5.1v with a multimeter first

>>1149480
>If it's the solid tip plating then shitty steel wool is hardly too much harder than brass.
It's a _lot_ harder than brass.
>I'm pretty sure brass is wetted by solder, is it the idea to have solder stick to the brass?
The oxidized/old solder sticks even better. You skim all the crap right off the top of the wetted tip.

>>1149473

$3.84

http://www.ebay.com/itm/DC-DC-Auto-Boost-Buck-step-Up-step-down-Converter-Module-Solar-Voltage-LM2577-/201422600692?

Auto step down step up(SEPIC)
Input: 3-35V
Output: 1.25V-30V
Biggest current 2A
Conversion efficiency 92%

I'm looking for vertical board mount toggle switches, but the sort of shit I want is all either expensive as hell or only available in massive quantities.
How viable is board mounting rectangular shit like this? Do I need to make it a custom part in eagle, would it already be, and would any board makers actually drill/route it properly.

>>1149488
Damn, that's cool.
I need a bigger circuit like that for my solar setup.
Or just a couple kilowatts worth of those, lol.

>>1149488
well, hooray! No, really.

>>1149490
Entirely viable. If your files have the right vias and drills, the vendors will do it.

I don't know if you'd find it in Eagle. I make custom footprints for practically every part I use unless it's a dead standard package like SOIC.

>>1149487
Sounds like a pretty good thing to have, but for some guy soldering with a $20 iron in a briefcase I think I'll be fine for now.

>>1149490
Pretty sure you can just bend the leads of a PCB-mount one up, but they're not very strong without something to support them. Why not just have flexible wires with the chassis as the structural element?

>>1149496
>Sounds like a pretty good thing to have, but for some guy soldering with a $20 iron in a briefcase I think I'll be fine for now.
If you don't want to use brass wool (I can see that, it's a few bucks a ball), wet sponge is even cheaper than steel wool and won't scrape the tip.

>>1149495
Because it's rectangular I'd think it'd need to be routed, and I'm not sure if the cheaper board makers (DirtyPCB) do that sorta shit.
>>1149496
I thought about bending but I want it as flush and consistent as possible.
Wiring is not an option, I'm trying to build a minicomputer out of mostly 70's parts, and fit it in an 80x60x150mm bamboo box. See also: >>1147795

>>1149488
Hot damn its perfect!

>>1149495
At this point I'm confused and an idiot.
I was always under the impression that for cheap PCB's you could only get them cut in basic squares, and have holes drilled at set sizes.
Now I'm looking at this and it looks like even the cheap houses fully route the panel to however you want it.
What I want here would be to etch the board, route it, then plate the holes.
I'm trying to work this out, wouldn't that plate all the edges? So etch, route what you want plated, plate, then route the rest?
I'm surprised a cheap board house would go to all of that trouble.

>>1149518
>wouldn't that plate all the edges
That is possible as well, but normally they drill first, then plate the holes, etch the board and then mill them to separate them from the panel.

>cheap board house would go to all of that trouble
Doing it cheaper while still being attractive to customers requires considerable investments.

>>1149503
Oh, do what I do with wires too thick to fit in perfboard holes. Solder the fat tags on to a snapped-off one of these, which you solder into the board. If you need one longer then just use some thick, solid copper wire as through-hole leads, it shouldn't look too bad at all.

>>1149523
So no plated routed holes?
>>1149529
I wanted the switches flush with the PCB for stability, so the plate they're screwed to wasn't taking the force, but this is actually a pretty decent idea.

>>1149531
I usually prefer the mounting plate taking the force than the PCB, if it's too weak then I'd reinforce it with another plate from behind. But if there's a few standoffs near the switch location then it doesn't matter as much.

How do I tell if an single-ended ADC is unipolar or bipolar?

I have an ADC128S022, but it's not really clear from the datasheet (to me at least):
http://www.ti.com/lit/ds/symlink/adc128s022.pdf

Electronics newbie here, rate my board
Total trash or ok?

>>1149585
>arduino
total trash

>>1149588
Why do people hat the arudino here? Because it's too easy?

>>1149568
page 4, paragraph 6.3
>Analog input voltage: 0 - VA
it's unipolar

>>1149589
because most of the circuit designs and code associated with them are pure trash

>>1149585
Nice array of LEDs.
But What are those four ICs? What does your device do?

>>1149531
Well, plated slots are possible too. PCBs are drilled/milled twice during the normal 2-sided production process. The first drill/mill produces the plated through holes and slots and the second unplated. The second milling also separates the boards from the panel.
People don't usually implicitly define milling patterns for the first pass, but it is certainly possible.

>>1149589
Yeah. Also too popular.

>>1149612
Seem to be quad optos.

So i bought this thing: https://pt.aliexpress.com/item/MT3608-DC-DC-Adjustable-Boost-Module-2A-Boost-Plate-2A-Step-Up-Module-with-MICRO-USB/32611355118.html

To be able to connect a 4 pin fan to by usb, the fan is 12V, so with this the fan will apin at the maximum speed right? And what if i want to regulate the speed with other than the potentiometer in the board? I would rather have another option to do this because that one isn't the best to do so, because i would have to be very carefull to not get over the 12v voltage, i will have to change it's speed quite often.

Replacing a device's on board antenna with an external one but the connectors on the board and on the new antenna are the same, do female to female adapters exist for this?

Pic related, I think they're 2mm SMA connectors.

>>1149612
It's a board to interface the key matrix of an electric typewriter. The ICs are opto-isolators. The LEDs indicate which channels are used, to make the whole think look more interesting

>>1149590
ok, thanks

>>1149631
there is no such thing as a 2mm SMA

those look like they could be SMB or MCX connectors

so how hard is it to interface an FPGA with an ADC/DAC for some basic audio DSP?

>>1149589
It's because so many anons get them without any idea what they're doing, no idea what they want to do, and can't even into LED resistor values or use a soldering iron.
They come here with questions like:
>have this, what do?
>How do I do this complicated thing that is way beyond me

Doesn't look like that applies to you >>1149585

Just ignore the "stop liking what I don't like" shitposters.
I like arduino posts as long as they're projects or specific questions.
You made something with your hands that required thought, planning, knowledge and skills.
10/10 in my book.
>Share your results! /diy/ loves to see problems solved and projects completed!

Pic is blurry but soldering looks solid from what I can see

>>1149676
Easier than the DSP part, unless your "basic" means something totally retarded.

>>1149678
Yeah at lest we're not getting "halp my led lit and now it doesn't, wat do?" posts.

Hey guys i'm making a semi-serious rail-gun for fun and I thought I'd ask if you guys could give me some advice as i'm no absolute expert and i'm rather amateur.

>>1149678
appreciate ya

Hey /ohm/ I bought a cheap rc car from thrift store, and I watched video here on reverse engineering rc car.

Couple of questions, so I have this circuit board
>pic related
in the rc car

in the video I am referring to
https://www.youtube.com/watch?v=aP38w8qssXU&t=649s

he does continuity test, and when he touches the solder points the motor reacts. Mine however doesn't react when I try it, Mu multimeter read 4.5v when I touch the pins on sop, does that mean the motors are bad?

Here is the blog post I am reading.
>http://dduino.blogspot.com/2012/02/control-rc-car-transmitter-with-arduino.html

>>1149705
do you mean a railgun or a coilgun?

shits back and running with no noise.

i couldn't really make sense of it. so i ripped out all the tubes.
it started humming again as soon as there was a pair of powertubes and a preamp tube in place on the left channel.

decided try the set from the other channel even though i've crosschanged 2-3 times now.
and suddenly it was fine again.

I can only conclude dirty-ass pins on my tubes, or maybe a shoddy solderjoint on one of the sockets.

i need to find one of those blue erasers, good for rubbing off corrosion without scraping off coating on the pins.
I'll have to see if i can spot some shitty soldering next time i open it up, but the way its mounted it might not be easy.

thanks for all the comments boys.

>>1149705
Coilguns require some beefy solid state electronics, but railguns are too inefficient for hobbyists to make. Either way you'll need a massive capacitor bank.

Large-scale motors are surprisingly expensive to get one with a decent amount of power for something like an electric vehicle project, despite being far less complex than a combustion engine. With a relatively young/small market with few suppliers (most of them are designed for industrial applications where cost is not a concern) and the closed sourced nature of major automotive manufacturers (who want to keep trade secrets behind closed doors), does anyone build their own? Like actually design their own rotor in CAD to be machined and/or know any suppliers that sell or wind the coils?

>>1149705
The simple fact firearms literally provide more bang for the buck is the reason coil/rail guns don't make sense. Lasers pack more punch if you want to do damage with an electric-powered weapon.

>>1149792
This is why.
Blowing up capacitors going to be a more cost effective means of launching projectiles than buying all the expensive electronicals to deliver that kind of power. If you're going to build up and unleash that kind of electric power, make a lightning gun like the one Winston uses in Overwatch and make Nikola Tesla proud.

>>1149847
They might be much less complex from a machining point of view, with lower tolerances and such, but they have to have many more individual processes done than a combustion engine does. Things like core lamination, coil winding, balancing slot carving, hand-epoxying, etc. We've had 0.25-2 horsepower shop tool motors for decades and they're still sitting around at least $200. If you want to get a hundred hp motor you're probably looking at half-a-dozen kilobux if you're lucky. Don't forget that the EV market is a far less sure one than the internal combustion vehicle market, so the big, mass-producing companies are going to be putting their money into advancing their motor and battery technologies instead of decreasing their current production costs.

>>1149732
I gave a look at this last week by looking at an old RC car I've had since I was like 6 years old.

The IC just sends a logic-level high or low signal to 2 H-bridges that control your steering and drive motors. The remote control is going to contain a complementary IC that has buttons that, at least for mine, short the IC to ground to be encoded in an RF signal. The manufacturer datasheets for these ICs, at least for mine, tells you everything you need to know about their operation including a circuit diagram for a typical usage scenario.

As it only sends/receives high/low signals, see if you can implement speed control with PWM signals from a microcontroller with MOSFETs in place of the control switches.

>>1149854
Seems simple enough to solder some wiring, hook it to a arduino and send high/low signals from the pins?

Also, how can I test the motors? Im scared to hook it directly to a battery and reverse the polarities, and fry it.

>>1149857
Don't. It'd be much smarter to tap into the controller rather than the car. Those motors are hooked up to a circuit that delivers what they're designed for.

Let me look up that IC and I'll draw up what I would try personally. Do you have an Arduino or other microcontroller to dick around with?

>>1149861
arduino leonardo.

>>1149850
>Lasers pack more punch if you want to do damage with an electric-powered weapon.

Lasers are straight-up trash for "doing damage". All you have to do to make something completely immune to most lasers is give it a coat of titanium dioxide.

You may be familiar with it Pic related. Aluminum also works well.


Anyway...coilguns are better suited to larger, heavier, slower projectiles, for reasons intrinsic to their method of operation. They're also much more efficient than railguns. Railguns are better for dumping absurd amounts of power into a smaller, faster projectile. They work far better at high muzzle velocities, but you need a lot more energy to start with.

Pick your poison.

>>1149869
"Where did you find that white paint, Citizen?"

>Also, how can I test the motors? Im scared to hook it directly to a battery and reverse the polarities, and fry it.

hehe. DC motors dont have a polarity. if you reverse the polarity, it just turns in the opposite direction, as it's designed to do. the transistors that power the motor, tho, may not like it, so it's better to isolate (cut) one of the wires from the PCB to the motor before applying a voltage to it.

>>1149857
What I would do is get a 7404 hex-inverter IC for ease of programming the Arduino with the standard PWM signal commands. Hook up like pic related to run it off the remote's battery from the VDD pin the IC in the remote runs on. I'd recommend jerry rigging a 9VDC power jack in place of the 9V battery (or whatever voltage your remote runs on) to save you from wasting 9V batteries while you're testing.

>>1149913
thank you so much for the diagram, I am sorry I should've mention there is no remote with this rc car, My goal was to control it from my pc/android. idk if that changes anything in you diagram though...

>>1149919
oops I think this diagram is for the IC on the RC car, not the actual remote controller.

>>1149920
If you want to control the RC car wirelessly from your computer, I think an inexpensive wifi module exists for arduinos (search amazon).

>>1149944
The second diagram I figured that is all it took. Thanks a bunch!

>>1149869
Can my poison be white paint?

>>1149631
Slightly-illegitimate looking MMCX.

>>1149676
Depends on what you've done up to now.

>>1149585
If anyone is interested in the first working test

>>1150026
nice, you should make a webm of it saying "send nudes" I bet you will hit FP on imgur in no time.

>>1148935
Literally none of those would be any different, because caps charge to max voltage, they do not average.

How do I test my circuit in Fritzing?

>>1150043
You don't, it's just a layout tool. Check the OP

>>1150038
That's semi correct, because it assumes that the load isn't enough to drain the capacitors. The top one and bottom would make an unregulated power supply. Under no load, the voltage would be more or less a smooth, rectified version of what comes out of the transformer. But under load, the caps are discharged partially and do drop the voltage a little smoother than a straight unfiltered PWM setup would do. The DC PWM setup would also require less filter caps since it's got a higher frequency. But they're certainly not ideal.

The centre one would actually drop the primary peak voltage if it's cutting more than half the waveform, and smoothing would probably sort that out, but I think the change in magnetic flux within the transformer coil would cause it to heat up too much, and it might also cause an unwanted voltage spike. Putting caps in between the variac and transformer's primary would solve some of this, but then you're just wasting more power.

A flyback converter really is a far better solution.

>>1150035
kek, will do

>>1150073
send nudes is the new hello world

hey guys, I've asked in other places too, but figured I would try this thread as well, so does anybody know any good literature on on PCB manufacturing that will cover all the modern processes and standards?

>>1150169
IPC and Perfag have the applicable standards. There are several publications from both.

>>1150175
thanks will see

I'm interested in learning how to use an external memory chip with a microcontroller. I might be interested in building some kind of data logger for processing & storing measurement data, such as voltage/current, luminosity or temperature measurements.

From what I understand, EEPROMs have quite limited write cycles. Would a cheap SRAM or Flash chip be suited for this kind of use?

>>1150187
>From what I understand
Check the datasheets. They specify the number of writes you can expect the chip to take, or "endurance". This varies between chips and manufacturers, but EEPROMs usually take at least 100000 writes, often million or more. Flashes generally have lower endurance. Then there are unusual technologies like F-RAM, which have much, much higher endurance.

Normal RAM naturally does not have any limit on writes.

If you're logging shit like temperature, just make sure you don't write your software in retarded way so that it pointlessly hammers the same byte again and again. This can kill EEPROMs quickly. On the other hand, when you do things the right way, you can expect the memory to outlast the practical life of the temp logger.

Need some diynosaur advice on a circuitry issue I'm having.

Recently built variable-speed stir plates, using a 1k pot to vary the speed of a 12v fan motor. Got the pots from China, so I don't expect much out of them in terms of longevity, especially since they are graphite-film.

In a few of them, there is some arcing between the internal contacts, occasionally giving off smoke. I'm guessing it's because of loose graphite between the moving contact and base plate.

Would injecting silicone/dielectric grease into the internals reduce the friction and increase the lifespan, or would it just insulate the contacts, ruining the pot? I have a few I can test it on first, but if y'all have any recommendations, they'd be appreciated.

>>1150197
You have a 1k normal pot in series with a fucking fan? And the pot arcs and occasionally smokes? Yeah, sounds perfectly normal.

Consider using pots with adequate wiper current rating or preferably less retarded control scheme.

>>1150201
You're familiar with 2w computer fans, right? Because professional fan control for PC's use 500ohm pots, and I've used wire-wound pots in similar projects. The pots are rated to 5w, and the fans barely draw 1 on max.

I'm unaccustomed to maintaining graphite pots, again, I never expected much out of them.

>>1150197
>tfw the only pot I had laying around was 10k and now the fan goes from full bast to off within 1mm of turning the knob

>>1150208
>10k
1k max rating, parallel with 500ohm res, runs for 3/4 turn, but thanks for asking and not being a dense child about it

>>1150213
what

>>1150204
How can it arc at 12V?

/ohm/,

Is this is a good beginner multimeter?

UNI-T UT890D

I need something cheap and cannot afford $100 multimeters.

>>1150187
>>1150195
>SRAM
>RAM
You'll lose that data when you power-off, you'll almost certainly want non-volatile memory. Whether you go with EEPROM or Flash memory depends on your scale, but I'd think that the mass-production off Flash chips for USB drives would answer that question for you. Of course, that scale of SMD soldering isn't going to be trivial, so if you're staying in the hobby market EEPROM might well be the better choice.

>>1150197
You should be using a PWM speed controller, running any current through one of those carbon pots is always a shitty idea.

>>1150253
>uni-t
seems like a decent cheap brand

>seperate ampere and voltage jacks
good

>manual ranging
major pain, there's others in that price class that do automatic ranging

>>1150272
>there's others in that price class that do automatic ranging
Give me suggestions please

>>1150272
>>1150279
Could you please take a look at this page (it has a good selection of cheapass chinese multimeters) and tell me which one should I pick? cheaper the better of course...
http://www.gear
best.com/multimeter-_gear/

I need a multimeter to diagnose faults in my electronic devices. I'd like to repair few (broken) things (monitor, headphone amplifier etc) and need something to help me diagnose what's wrong so I can order the parts.

Thanks!

>>1150282
if its low voltage then it doesn't really matter, so if it runs on batteries thats fine. if it plugs in and you are measuring resistance/conductivity/capacitance/diodes when its not plugged in then thats fine. if its plugged in and your measuring voltage on the low voltage side of an isolated supply/regulator then thats fine too.
if your measuring line voltages or stuff plugged in that might be at line voltage then you want a meter with a proper cat rating made by a proper company. just because it says cat rating on it means fuck all, chinks don't give a shit they write anything on it. so do other people too actually. all the cat rating means is that when you try to measure something too high voltage for whatever reason it doesn't explode and kill you. for that matter you should practice not holding onto your meter when measuring just in case it does blow up you don't want your hand to come off.

other than that have a look at features, backlight, autorange (some people hate autorange! decide if you are one of those people) capacitance, transistor (some people think having pokey holes for transistors on a meter is a safety issue, and they are right you shouldn't use it near live and not extra low voltages), conductivity beeper (very useful if you don't have two sets of eyes) maybe frequency counter but thats kind of specialised i guess.

i like a nice set of silicon leads too but as long as they are rated then whatever. check them every time you use them for cracks/damage/wear etc.

if you are going into industry the standard is probably fluke but you will pay for it.

if you want to dive headfirst into the rabbit hole check somewhere like eevblog forums but as always be aware of fanboyism. just like anything people will defend manufacturers/brands while curiously having absolutely no investment to protect.

oh, yeah, you won't get very far with a meter. fine for checking fixed voltages/supplies but you can't look at signals.a scope is much more versatile.

I'm trying to simulate an AC clipper across an inductor in Multisim and Multisim is shitting itself.

I have zeners back to back across the inductor (which should cause it to short on itself) and instead Multisim thinks the current should instantly transition to negative infinity and crash the simulation. I've tried putting a small resistance in series, I've changed breakdown current and voltage, etc, and nothing works. I've had some issues before at high voltage/high amperage

>>1150317
This bullshit is actually happening at 100V, even with a 1Mohm resistor in series with the zeners. Fuck Multisim.

>>1150279
http://www.eevblog.com/forum/testgear/multimeter-spreadsheet/
Have fun.

Is this real life or just a fantasy?

>>1150342
>try to reduce 20mp phone picture by taking a screenshot
>.png
Fml

>>1150294
>>1150337
Thanks anons!

I guess I'm leaning towards UNI-T 136C at this point. It has autorange, temp probe and it seems that can even handle capacitance in 100uF range.

If I stick with this hobby, I'll save money later on to buy something in $150 range.

Correct me if I'm wrong but it's probably pointless to spend money on a midrange one when these cheapass have 95% same features but are not as precise.

>>1150282
The Uni-t 61 series, in particular the UT-61e at $50, seems to be very popular in the entry level, but still professional market. I've heard good things about uni-t from several places, but I never used one of their meters personally.

>>1150330
>>1150317
I got my way through the clipper issue. Unfortunately it won't be as beneficial as I'd hoped.

>>1150245
Eh, I guess "spark" is a better description. Just a tiny blue/yellow flash, worst when it's moving, sometimes lingering and flickering on what I assumed was either dust or uneven graphite.

>>1150375
It's probably tiny particles of graphite burning from the heat. I imagine the pot is getting hot, and this will affect its accuracy. If the little flashes happen often enough, which it probably is, the resistance of the pot will increase over time, so you'll need to find a better potentiometer.

Just bought components to make an arduino uno, what are so good beginner projects?

>study EE for 2 years
>drop out cause I'm retarded
>go study CS for the next 6 years
>want to get back into electronics
So, I picked up a book, busted out my arduino, salvaged some parts and ordered a whole bunch more from china.
So far so good, only out €250 or so for a heap of stuff.
But now I have an irresistible urge to set up a whole fucking electronics lab with a multimeter, scope, power supply, function gen, soldering station, logic analyzer and all that jazz.
This is gonna cost me an arm and a leg, isn't it?

You can easily build a simple adjustable PSU, logic thing, function gen, etc from Aliexpress trash. The components are pretty good. You just need some wire, solder and cases.

>>1150523
>>1150522
Logic analyzer: https://www.aliexpress.com/item/IEZ-USB-FX2LP-CY7C68013A-USB-core-board-development-board-USB-logic-analyzer-I2C-serial-and-SPI/32464847698.html

Signal generator: https://www.aliexpress.com/item/AD9850-DDS-Signal-Generator-Module-0-40MHz-Test-Equipment/32601850164.html

>>1149847
I know of ways to build high powered electric motors. The biggest limitation is lack of software. Without an easy way to rapidly design and simulate possible motor designs it's difficult. There is a fully featured 3d motor design and simulation program but it is $15,000 per license. It's not even complicated software they just charge extreme amounts because of obscurity. An open source one would be fairly easy to make but it's near impossible to find anyone willing to work on anything open source and electric motors.

I don't even know where those sort of programs could be pirated either. Even if I wanted to buy it there's no way to evaluate the software to see if it's worthwhile for smaller businesses to buy it.

>>1150522
>multimeter
You really should have a multimeter. Even a shitty one will do and won't cost much.
>soldering station
A matter of convenience. Buy if you have disposable money.
>power supply
Your next project.
>scope
Rather useful and you're bound to get one sooner or later, but you can certainly start without it.
>function gen
Your second or maybe Nth project. Usefulness depends entirely on what you're doing. It isn't usually that much needed.
>logic analyzer
Scope largely does its job as well. Get a scope first, consider this later if needed.

>irresistible urge
Uh, it might go away after a while. If not, congrats on your new money sink. You might want to consider getting a ham license too, if you still have some spare cash.

Thinking of making myself a wireless headset + microphone how would i go about implementing this.

What components would i need

>>1150530
Wired headset plus this:
https://www.aliexpress.com/item/Ugreen-3-5mm-Bluetooth-Audio-Receiver-Universal-Car-Speaker-Music-Receiver-Adapter-with-Microphone-for-Headphone/32624733221.html
Or any other Bluetooth dongle with microphone functionality. Take it out of it's housing, cram it's board into one side of your shitty plastic headset, put a bigger LiPo in the other side, and bingo-bango you've got yourself a wireless headset.

I'd have done it ages ago if the plastic on those things weren't so fragile, I'm half looking for a sturdy pair of headphones with good sound quality and half looking how to make my own. With a detachable, replaceable mic of course. Good luck.

>>1150529
>logic analyzer too expensive

this has become a moot point ever since the Saleae and it's clones exist... $8 for 8 channels 24MHz

>>1150531
>strapping a lipo to your ear
after I've seen how those can blow up I'd be very wary of such a thing

>>1150535
wrong pic, here you go

>>1150540
I just find this to be a very hard story to believe.

> all that soot but no blistering from the heat
> "melted plastic" on her neck, but no blistering
> ear is perfectly clean except for around the bottom edge
> long hair over the burn area is not burnt
> lines in the "soot" perpendicular to jawline, almost like she was dragging black makeup from her ear to her jaw
> damage to her lip?
> no signs of first aid (area cleanup, bandaging) despite lots and lots of cabin crew having first aid tickets

Don't get me wrong, I have seen LiPo's go off and I know the carnage they can cause, and for that reason I call bullshit. They shoot flame for fucks sake, she would be burnt as fuck.

>>1150556
>I just find this to be a very hard story to believe.

Yeah, it's some kind of false-flag set-up by the ATSB to discourage passengers using their own headphones on-board.

https://au.news.yahoo.com/technology/a/34652641/headphones-explode-mid-flight-burning-womans-face-and-hair/#page1

>>1150342
>Croatian engineering

Should relays make a clicking sound? If not, what is wrong?

>>1150583
yes

>>1150583
Mechanical relays consist of a small coil and a spring loaded metal arm. When a tiny current activates the coil, the arm snaps back, completing the bigger current circuit. The snap/click sound is how you know the relay is still good.
F.ex if your car battery is fully charged but the engine won't turn with key? If you hear a click, the starter is shot. If you hear nothing, the starter relay (3$ fix) is shot.

>>1150583
Unless it's a solid state relay, yes they should click

Possibly stupid question comming trough.
What is the effect of bandwith-selection on oscilloscopes?
I want to detect a ~3mV signal at 21MHz. The only Way I found to cut down the noise to an acceptable level was selecting "Bandwidth 20MHz" Will this fuck with me? I only want to see the signal at all. Voltages don't need to be accurate.

>>1150634
The amplitudes won't be accurate, there will be frequency dependent phase shift and the combined result of these both is waveform distortion.
Pic related, 20MHz (shitty) square wave with and without scope's 20MHz filter. Note that the filter will also fuck up square waves of significantly lower frequency.

>>1150634
Frequency or Time sets the length of the timeslice on the screen (x axis)
This,
>________---------------_______
at a lower frequency or higher time , will look like this
>--_--_--_--_--_--_--_--_--_--_--

>>1150634
you should probably FFT my man, this is pretty low signal and noise will easily drown it
or you could make a 21 MHz bandpass

>>1150654
>you should probably FFT
Thanks, I'll look into that.

hey im fairly new to all this, only project so far was an electro etcher, and wanted someone to tell me if i screwed something up in my second project.

i have a case that will fit ~30 18650 batteries, a friend salvaged for me. the plan is to attach them all in parallel to make a massive usb power bank. will this work using these parts?

http://www.ebay.com/itm/302076444202
http://www.ebay.ca/itm/192139776951
http://www.ebay.com/itm/162336765007
http://www.ebay.com/itm/400891160595

also where can i buy leds without having to buy 100?

>>1150728
just buy 100, you'll find use for them for sure and they're just a few bucks from aliexpress

>>1150729
enjoy you're house fire

Hi guys. I've got a question about digital logic being controlled by analog circuitry.

This circuit will have the digital input typically be high, with the switch open and cap charged to 5V. When the switch closes, the digital input is pulled low, and the cap instantaneously discharges.

Now, when the switch opens again, the cap begins to charge again, and the voltage at the digital input increases. The chip has low defined as <1.35V, and high as >3.15V. It's the intermediate region that I'm concerned about. The RC should have a time constant of a few seconds, so it'll be in the uncertain region for a while. Will there be a smooth transition point that the manufacturer doesn't spec because it varies chip to chip? Will it jitter, rapidly switching from 0 to 1 in that region?

Thanks in advance. If it matters, the chip is a CD74HC4053E.

>>1150763
Alternatively, I may have just thought of a simpler way to do it. Are there IC's that, when input a high signal, will output high for a certain amount of time (ideally programmed with an external RC and not a clock?) Or the same question for low signals.

>>1150763
>>1150764
i wouldn't try it on that chip. it doesn't specify a schmitt trigger input and its datasheet recommends a maximum 1us transition time.

you can use a schmitt trigger inverter or buffer instead. their transition voltages aren't precise so if you need very precise timing you'll need a comparator.

>>1150728
>~30 18650 batteries

Your solar cell charger
>Rated voltage: 5V
>Rated power: 0.95W
Under ideal conditions the panel will supply less than 200ma
200ma / 30 batteries = 6.7ma per cell
Your battery bank will never charge
also, the voltmeter will want some of that too.

>>1150779
ya the solar panel was never meant to be the primary means of charging the pack, mostly just something to be left outside for a few hours to hopefully turn on a phone in an emergency. how do i figure out how long it would take to charge it in theory? should i just ditch it?

do the other parts look like they will work? i'm not really into electronics so i think i could easily miss something obvious.

>>1150787
>325 watt hours / (0.95W * 0.95 (eff) * 0.4 (sun))
>37 days

i assume the sunlight factor is being generous. that efficiency is if you're using a good buck converter with MPPT. if you're using one of the shitty linear charge regulators it'll be 2+ months. self discharge should be less than ~20mA, idk what protection circuitry 18650s have inside.

>>1150795
wow ok so i might as well not use the solar panel at all then, thanks anon.

>>1150808
it'll at least counter the self discharge, but that might not be worth it for the added complexity.

>>1150728
Even if you do ditch all the solar and go to charging by ""USB"", I'm not sure how the charging circuitry will handle it. If you want a fairly quick charge of 1A per cell, that's 30A at 3.7V, or just over 100 W. I'm pretty sure a standard $0.50 charge circuit can't output this kind of power. I'm also not sure what happens if one cell happens to be a little defective and charges less/more quickly. If this happens and a significant fraction of the 30A goes into a single cell, you're going to have quite the firework display on your hands. And the more cells, the higher the chance that one will be defective, especially if they're from different sources, and even more so if they're "salvaged". To mitigate danger, I'd run a few thermocouples through your battery briefcase, hooked up to an auto-power-off once they hit 45°C.

Making a single power bank is a bit of a waste, you could run 10 USBs off that thing, all pulling 2A. Personally I'd make it a laptop UPS with 12V and 5V outputs. Making multiple 1-5 18650 power banks would be fairly useful, but it's too much of a competitive market to sell them in, even without safety regulations. If you do want to charge it quickly, I'd look into what they use to charge E-bikes.

>>1144916
Hi I want to replicate my lenovo thinkpad piezo speaker bios startup sound, how do I find out which notes are played?
thinkpad t420 / thinkpad t61
I could put an oscilloscope to it and record the waveform maybe but that means i have to connect the scope to the speaker while the laptop is running and I'm not comfortable doing that
pls help thanks

>>1150910
Just record it and plug it into audacity to analyse the frequencies. Like pic related.

>>1150915
yes thanks this is what i just did from advice on irc
it got me this:
1200hz, set to square wave:
http://onlinetonegenerator.com/

it sounds exactly like a Thinkpad now :)

>>1150921

>>1150921
Jesus, that sounds awful.

Took apart an old telephone casing to salvage this lcd screen.

Im relatively new to electronics, is there a component i can buy to connect this to my arduino or breadboard?

>>1151006
>ribbon cables
Well if they've got detachable connectors on one side or the other then you'll just have to buy some sockets, but since it doesn't look like it I don't think so. I'll assume you don't have a logic analyser or scope, so you'll have to look up the datasheet and figure out which I/Os you'll be needing. Typically LCDs are so cheap that it's not worth salvaging them, at least if they're a 7-segment display. That's entirely up to you, of course. If it's a dot matrix display and/or has a backlight, you might want to keep it, but it will be some effort to get it going.

For a little depressing comparison, you can get 10 7-segment LED separate digits for $1.50, and a 4-digit board LCD for under 70 cents. On Aliexpress, where your dollar is their lack of corporate responsibility.

Also welcome to the bump limit.

>>1151007
>Also welcome to the bump limit.

>>1151071

new thread

I know this is a question I should know the answer to already, but I'm pretty dumb.
Let's say I have the primary side of a transformer hooked up to the wall, and the secondary side an open circuit? Should be little to no current on the primary side, right?
How about if I short the secondary? Blow a fuse?

>>1151666
reactive current will flow in the transformer according to its primary winding inductance. no power will be dissipated due to that alone. however, transformers also have a winding resistance so as that reactive current flows a relatively small amount of power will be lost as heat.

if you short the secondary the transformer action will work and as much current as possible will flow, limited only by the leakage inductance and resistance. you'll blow a fuse, trigger your breaker, or melt your transformer. if the windings aren't isolated you might melt them together and then it'll be an even shorter short across your mains.

if you blow a primary fuse it's open circuit.

don't worry transformers are so poorly explained to most people that no questions are stupid.

>>1151815
Thanks anon. I only learned about them in circuits 2 two years ago and I'm more of a microcontroller guy than a power guy, so I have a healthy fear of messing around with mains power