/diy/ - Do It Yourself

alright, third time's the charm.

>>1221700
Yeah, dot-matrix, not segment, I should have been more specific. Maybe OLEDs would be nicer to look at, like one or two of these:
https://www.aliexpress.com/item/1-3-OLED-module-white-color-SPI-128X64-1-3-inch-OLED-7pin-LCD-LED-Display/32818944937.html
A little mini-scope or a bar graph of current would be quite nice to have, as would an Ah accumulator, or even just a resettable peak current readout.
>What kind of aesthetics can you get out of a dot matrix anyways
Fast animation might be out of the question, but at 128x64 in a 30mm diagonal screen, the possibilities for aesthetics in 16x48 character boxes are plentiful. You like blackletter numerals? Nixie style thin and modern? You really want dem Slav runes? You got 'em. There are even modules such as Pic related with the top (or bottom) 1/4 in yellow and the rest in blue, so you can have brighter indicators for ranges or icons or whatever.
A full RGB screen is almost certainly overkill for a power supply. I could see it being nice for programmability, if one wants to glitch the input power to a chip or assembly or whatever, but that's another DAC or two and associated circuitry, so nah.

>>1221721
I wonder if the guy who posted that as the OP to his "extended battery phone" thread will find out that I suggested it for the /ohm/ pic? He was kind of upset that we laughed at his design. That is part of a plastic fork, by the way.

>>1221727
>128x64
I was actually planning on running the thing without a microcontroller and just using some simple 7-seg decoders or something similar, but you're not giving me much of a choice. What would taking the digital straight from the ADC and putting it into the pins of a dot matrix look like? I could probably throw the signal into a few decoders and maybe demultiplexers back and forth until I get something that looks neat, but that's more along the lines of an art project.

I agree that a bar graph would be a really good way of displaying voltage and current, like on the bottom of a good multimeter's screen, but I'd like a high resolution so I can see subtle (300-100mV, 30-10mA) changes, meaning at least 100 pixels/LEDs. A single line, wide dot matrix would be perfect for this, but what kind of encoders do they have in them? Is it typically raw digital video data in, or text data?

For the numbers themselves I think I'll look for the most aesthetic 4 digit 7seg display I can find, and see if I can pull the parallel straight out of the ADC and into the 7-seg decoder without any digital processing at all, and while it would have to be fixed decimal-place, I think I'm fine with losing a little precision on the low end.

And now a good idea: take the output voltage and use it to drive a frequency proportional to the voltage to a small speaker/piezo element, so I can HEAR the difference in the voltage! Current too! GENIUS! (I'd use the (buffered) output voltage as the threshold voltage for an opamp relaxation oscillator, and do the same with current except amplify the voltage across a shunt and drive a lower frequency)

>>1221743
On this note, is it better to measure current by amplifying the voltage across the shunt, or just using a very low reference voltage on the ADC?

How do i get caller-id on a phone line? The line is a sub network at my workplace, the phones there can show the id. The line is at constant 48V. I had to divide the signal to 5V to be able to measure it safely with a scope but i couldnt see any changes when it was ringing. The signal might be lost due to the divider, how can i get the signal on top of the 48V without probing it? My scope is rated only to 10V

>>1221774
You could measure it through a capacitor, and therefore only pick up the AC/fluctuating parts of the 48V with your scope, but I can't see why it would change when ringing or calling. The phone calls themselves almost certainly won't be transmitted over the 48V, that's just power to the phones and routers and whatever else you have on the POE network. You can probably trace caller id by looking at what IP addresses are being uses because those are probably IP phones.

TL;DR talk to your sysadmin/network engineer.

>>1221776
It is an old school telephone line, no ethernet.
Will try the caps

>>1221727
Those OLEDs are pretty nice looking, and not even twice the price of an equivalently sized greyscale LCD. If I use one with the two different colours then I could have current on blue and voltage on yellow, which would be neat. But getting that to work without a microcontroller will likely prove difficult.

Oddly enough, I can't find any 1*100+ LED rows/dot matrices for an easy bar graph, just 10 LED VU meter ones. I could just go with a binary readout with a line of 14-16 LEDs, one representing each ADC bit. It's not a linear bar graph, but it's certainly more compact and more detailed than any linear bar graph I'm likely to acquire. If I do this and a 4-digit 7-segment display it will probably be easiest, which combined with the audible voltage and current shifts should give me plenty of data.

>>1221774
>>1221776
>>1221778
Okay, so, it is going to depend on how your PBX (personal branch eXtension or whatever the hell that originally stood for) at your office functions.
If it's a 'sub network'; as in an internal line, the PBX system there is transmitting the info it received when it got the call, prior to routing it to you.
It most likely IS using the 48v (or whatever carrier voltage) to transmit all the info, including the voice once you answer.


How it does that though, is gonna depend on the model. OLD telephones used to use 55v iirc and 95v for the ring... not sure what model you guys have there.
Do you know the model and/or brand?

>>1221647
You can make a serial programmer for aorund $10. Look up the JDM programmer. I built that, then used it to program a PIC with a USB interface so I could make a USB programmer called the USBPICProg.

redpill me on digikey vs sparkfun

>>1221813
Digikey if you're not a little bitch boy

>>1221748
Whatever has the better signal to noise ratio will lead to the best result.

>>1221813
digikey is overpriced but thats what you pay for low minimum quantities and good packaging/ship times.

sparkfun is massively overpriced and just takes advantage of people too lazy or new to order and solder a $5 pcb.

>>1221836
What's a cheap alternative?

>>1221846
Chinamen of eBay etc.

>>1221810
The station is a KAPSCH Meridian and the phones are the same brand.

Ill try your AC coupling tip after the weekend.

>>1221851
with digikey you pay for 100% legit merchandise (mfr factory->digikey->you)

ebay if it's high demand you run into the possibility of fakes. got a bunch of fake voltage regulators from china before.

>>1221721
Can I use skinny cheap wire from dollar tree with a red crimp connector or do I have to buy better thicker wire?

Wanting to build a case + controller for my lg g3 (something like a handheld retropie)

anyone have a few starting points I could use? i'm not an engineer so getting a 3d printed case may be difficult/expensive.

>>1221927
The POTS adds (!) the ringing (AC) voltage to the line (DC) voltage. Be careful with your 10V scope.
Look for calling line identification presentation (CLIP) to see what type of signal to expect.

>>1222035
This.

Careful with scope, and careful in general. It won't kill you but it's downright uncomfortable to be shocked by a ring signal. :P
(This is more a concern when you've got a dozen lines on like, a 66 block... but still... don't zap yourself and don't zap your equipment.)

Also, maybe try (if you can find one) of those old 'pass-through' caller ID screens to see if that works?
Like, at least that might actually tell you if your PBX is doing something 'weird' as opposed to using the standard CLIP methods.

>>1222043
>don't zap yourself
unless you're an electrophile like me. I had my first close encounter with electricity at age three to four when I was poking around in the wall sockets (220V) with a wet toothpick. Since then I know that only one side is live.. My experimental zapper has about 40Vpp, adjustable frequency, funny physiological effects.

>>1222074
That's... all sorts of amazing. XD

>>1221743
There's an I2C controller running a memory read/write interface on board those little things, so probably not very useful to you.
Have you considered the ICL710x integrated ADC/7-seg decoder chip series? It couldn't get much easier to build a voltage or current meter than that.
The voltage theremin idea is a good one, especially for looking for side channels on microcontrollers. Heh heh heh..

>>1221813
Mouser, if you have a big enough order to be worth an $8 handling fee.

>>1221998
If 22-16awg, sure. If not, splice it to a larger wire or add a few more stripped ends before you crimp it.

>>1222074
Bretty gud you kinky little monkey!

>>1222002
you don't have to be an engineer to learn CAD (eg solidworks) and design a case. pretty much all you have to bear in mind is minimum wall thickness (50mil, pref 100+) and producibility (no overhangs).

designing the electronics behind the controller is a no-go for you unless you salvage shit out of another controller and reverse engineer it a bit. probably not something you should bother fucking with as you're unlikely to succeed.

>>1221829
Which I can only assume depends on the amplifier. I'll still probably go the amplifier route because it gives me more freedom when it comes to adjusting the gain. I just realised that since I'll be using standard ADC(s) and standard 7-segment displays without any digital processing in between that my maximum resolution will be 100V and 10A, meaning I'll need to be stepping down the actual maximum of 30V to be 3/10th of my voltage reference, and will lose some resolution in that manner.

>>1222112
ADC+7seg decoder? That's pretty neat as long as I can get them in high enough resolution. Looking at the datasheet for the ICL7107 it says 3.5 digits, which I assume means 12 (or 13) bits, and therefore has a maximum count of 4096 (or 8192). Since 30V and 3A are my maximum outputs anyway, this suits me fine. I wasn't completely sold on these until I saw that they come in PDIP form factor. I can get 2 from AliExpress for $1.50, but are ADCs something that you shouldn't trust from AliExpress? Thanks for the great IC suggestion btw.

>>1222211
3.5 digits means 2000 counts. You would be wanting 3.75 digits, which are rather more expensive, or 4.5 digits, which are cheaper new than new 3.75s. Perhaps you want the MAX1499 or MAX1447 which require a few less passives, and add a peak/hold function, but cost a bit more, aren't AliExpress items, and don't come in PDIP.
The Intersil 3.5-digit display DACs are pretty commodified and not in particularly high demand, so I don't think there would be any real problem with counterfeiting, if you're willing to deal with the limited count.

>>1222234
>if you're willing to deal with the limited count.
What happens if you put more than the reference voltage into one of those things? Do they just flash the digits at you or blow up? I might have to add a resistor in series with the voltage adjust to stop that happening.

>>1222263
Read the datasheet. Obvs you never feed them a voltage outside of the power rails, but in general they just display some overload pattern when outside of their reference range.

>>1222322
As I'll be dividing the voltage down and amplifying the current up for 2A and 20V to equal the maximum range of 2V (or 200mV?), I won't be putting any more than 3V (or 300mV?) through the ADC since the power supply can only output 30V and 3A, so no worries about overshooting the rails there since it says the maximum values for the analogue in are the voltage rails.

Now when it comes to only having a maximum range of 20V and 2A, I might add another range or two to get a 200V range and 20A range, and maybe also a 2V and 200mA range too. All it takes is a multi-contact (rotary?) switch and a few different resistors for the amplifier/divider.

In the meanwhile, I think I'll build a quick aural inductance meter that uses a variable frequency generator and a few capacitors to find resonance based off the signal amplitude. Because I just discovered the elegance of opamp relaxation oscillators, that's why.

How would you do it to have a container recognize paper?

I mean, What would you use to know a container has anything that isn't paper inside it?

Why do opamp relaxation oscillators have this startup time area? I'm trying to get audio frequencies out of them, but with 1s startup at 20Hz I can't see it as ideal, and this is both with 1 10n cap and 10µ cap with appropriate R values. Also I'm not quite getting the frequency I'm calculating (21Hz instead of 16.8Hz in this case, and 10.6kHz instead of 30.3kHz), which I assume is due to the opamp not being rail-to-rail or some other imperfection with it. Now I understand that this primitive oscillator has its limits and is expected to display unideal characteristics, but I just want to be able to plan around them, both for spice and IRL.

>>1222339
You could even add a comparator and do auto-ranging, were you so inclined.

>>1222411
I could see R5 influencing the virtual ground. Any change when the virtual ground is driven by 10k resistors instead of 100k?

>>1222411
I suggest adding a low-impedance buffer for the virtual ground, or better yer, for the purposes of analysis and simplicity, why not add a proper voltage source for each rail?

>>1222424
>>1222427
Making R3 and R5 1MΩs, making R1 and R2 100mΩ, and making C1 and C2 10F made negligible difference, I don't think that's the root of my problem. The voltage rails are essentially stationary with respect to ground, and the capacitor potential barely fluctuates by a µV before increasing. Changing the voltage dividing capacitors to 1pF didn't help either so the issue isn't with those charging up (not that it would have been), so I'll just try to build one on a breadboard and see what happens.

>>1222429
I think it's actually a spice thing as when I set the voltage sources to start at 0V and move to 2.5 each afterwards then I get oscillation from the get go, problem solved. While up at 20kHz I do have to use a 100Ω resistor instead of a 200Ω resistor to get the right frequency, since the expected square wave output is surprisingly sinusoidal, I think this is just a product of the op amp's ideal frequency range being reached. Now I just have to figure out how to drive a speaker with this (amplified) signal through an electrolytic capacitor without biasing it the wrong way.

>>1222439
Ok, tested, it works well. It turns out that if you make an elektrosluch you can use it to listen to what waveforms you're making as opposed to using an oscilloscope. Just by putting an inductor to ground on the output of my frequency generator, I can hear it as I adjust the potentiometer. For 3/4 around the pot it sounds like the rail-hitting meme bass often accompanied by a highly compressed "laughing while crying" emoji, so now I know what a square-wave sounds like. The bass turns to mosquito whine really quickly as the potentiometer is turned down, so I'll definitely want a (reverse?) logarithmic potentiometer or a 25-turn potentiometer if I want to use this as any kind of variable frequency generator.

Oddly enough I couldn't get a 2-transistor oscillator to work the same way, but this is better anyways.

You see these kids making these so called "one million volt" tasers all over youtube. Anyone have a schematic for a police grade one. Those things pump out around 50k volts and Im really curious how they give them that much punch.

>>1222035
>>1222043

I check the voltages every time with multimeter. One time some technician measured 300 volts in a phone or internet connector, since then im not trusting this place....

I was more concerned with the isolation, since the phone line can be at any potential compared to the local ground. But i solved that by connecting the scope to a laptop on batteries.

>>1222472
It's all about the waveform. A civilian stun gun will simply use a very high voltage to make the subject convulse in pain, while police TASER models have a waveform designed much like a defibrillator, but with the purpose of causing muscle spasms instead. I think. You can probably find patents for this sort of thing online.

>>1222473
>connecting the scope to a laptop on batteries
Never heard of the Kapton tape trick? Just wrap a little around the ground pin, and your scope's ground is now floating with respect to the building's ground.

>>1222476
Been there done that.
Its easier like this

>>1222474
Oh very interesting. Thanks anon. Ill search around.

Where do I get simple stuff 22 gauge wire, female pin headers, etc in my city.
Places like home depot are for heavy duty high voltage things when Im looking for hobby grade electronics. Like if I wanted to buy a bread board where would I go. I buy everything off online but when I run out of wire I dont want to have to wait a week to continue my project.

>>1222472
/diy/ only has two board-specific rules. One of them is "no weapons".

>>1222119
i'm confused as to why I can't use something like an adafruit feather to build a simple controller

>>1222500
>in my city
where would that be?

stuff is so cheap off ebay just buy large amounts

I have an atmega8 and an arduino uno. It is worth making an arduino out of the atmega 8. I dont want to solder to the arduino so the atmega8 could make everything more compact.

>>1222473
>laptop on batteries
And now you can't touch it "since the phone line can be at any potential compared to the local ground."

>>1222500
try radioshack
oh wait..

>>1222548
fix'd

>>1222514
because you can shove any shit controller in your case, but packaging the switches in a robust manner is what you'll be hard pressed to do with no engineering background. and then you have to figure out how to get your meme controller to talk to your phone. good luck with that, i'd literally rather slice my wrists than touch android. i'm just telling you up front that if you want to actually build a whole controller from scratch without ever even having 3d printed something then you're guaranteed to fail.

if on the other hand you just wanted to print a bracket to connect your phone to some generic usb game controller then that may be easy.

Does it make sense to build your own circuit transformers? The cheapest in my local store are like 5usd

>>1222473
Phone lines are usually -50v
Special circuits are 110 or 220 up to rarely 300 but you shouldn't have any of those circuits easy for you to access.

Pbx systems are about the same as normal pots lines 50v, but I could see some upping that so the station could get power from the line for features.

Phones also ring to 110v ac

>>1221774
The caller is sent as a burst of modulated bits. It sounds like a modem or fax for a second and it comes through before the first ring.

Are you proving the phone across the pair or with one side to ground? It's a differential signal, you won't get much looking at ground.

>>1222581
Proving=probing

Anyone do EE-telephony sorts of work? My boss started taking the piss at me claiming I leave early and shit. I work in telephony (SIP phones, virtual modems, etc) and I'm just getting all sorts of shit. Stuff to do that would impress them would be nice (and I had an idea by integrating visual content) but otherwise I'm not sure if I can keep my job if he keeps thinking I'm useless.

>>1222573
Depends on your amps I'd say. If you're dealing with millis you won't saturate a core, but if you're dealing with amps it's really easy to knock out a transformer.

>>1222581
>before the first ring
or after the first short ring or after the first ring unless you pick up too fast..it's a mess
en.wikipedia.org/wiki/FSK_standards_for_use_in_Caller_ID_and_remote_metering

>>1222548
Do you have metal buttons on your keyboard?

>>1222581
>>1222625
Thanks

Its differential i feel, only 2 wires coming out of the wall.

>>1221721
I've noticed that every personal-use FM transmitter has a shitload of interference/static. What causes this, and could it be prevented by making my own FM transmitter?

>>1222671
transmitter has no interference, receiver has because of noisy environment and transmitter is too weak

>>1222674
So it's all about signal strength? I'm comparing against commercial FM stations which don't seem to have any interference, with the car in the same place.

I guess that means I could build my own that has a stronger signal than what's available off the shelf

>>1222671
What's the legal situation? Does Ali sell FM transmitters?

>>1222677
I'm in Ontario, Canada, so whatever regulations are set by the CRTC. I'm looking them up now, I'll report back

>>1222678
Currently, the only FM broadcasting transmitters that Industry Canada exempts from authorization are those that meet the requirement of Broadcasting Equipment Technical Standards 1 (BETS-1).
[blah blah]
which corresponds to a transmitter power output of less than 1 microwatt (μW)

>>1222680
I think I found the spec sheet of one I own
https://fccid.io/IKQFMT4/Test-Report/Test-Report-1264885

RF Power level <50nW

which means it could be 20 times as strong legally where I am?

>>1222456
>rail-hitting meme bass often accompanied by a highly compressed "laughing while crying" emoji
This is relevant to my interests.

>>1222500
Look up electronic component distributors or maybe the local drone shop.

>>1222526
>making an arduino
No, it's never worth making an arduino. Just get an external AVR programmer and enjoy life.

>>1222570
>he doesn't know about the bluetooth HID profiles

>>1222607
Tried that, got canned anyway.
The boss is clearly gunning for you if he's making up false accusations. Start worrying about impressing your next employer.

>>1222642
>Its differential i feel
Of course it is. They use a literal 1200 baud simplex modem to send or receive the data. Place a telephone isolation transformer across the line (after the first ring, using a relay) in series with a capacitor to avoid drawing off-hook current. Modulated data will then be available on the secondary. Use a PLL (e.g. LM565 or CD4046), modem chip, or ADC+software to demodulate. Send to UART RX pin and enjoy.

>>1222690
Looks right but I'm no communications lawyer.

>>1222690
Yes, you're the lucky one. I'm limited to CE cert, 50 nanowatt ERP. Technically these things are quite good for 10 bucks, stable PLL, all channels, stereo.. all that's missing is a mW of output and a better antenna which would be enough for a kilometer or so. Wikipedia has something:
en.wikipedia.org/wiki/FM_transmitter_(personal_device)
USA 10nW EU+UK 50nW Canada 1µW.

>>1222717
Does anyone have experience with 1µW transmitters? I'm assuming all the ones I've tried are 10 or 50 nW for sale in the states.

>>1222716
>it's never worth making an arduino
why not?
>get an external AVR programmer
I dont see how this is related.

>>1222788
Arduino is almost the diametric opposite of doing it yourself.
>I dont see how this is related.
Arduino hardware is literally a USB-to-async-serial interface and a mediocre bootloader on (usually) an ATmega with (some) IOs broken out to a space-wasting pin socket form factor. There are other, more practical tools to burn an Arduino or other .hex file into the device.

>>1222717
For all you fm listeners on the thread, you know how when you drive further away from a transmitter, the signal to noise ratio decreases but the volume stays the same? I can only imagine that the receiver is dynamically changing the volume or gain of the signal in response to its amplitude, but how is this done? Does it have something to do with the carrier wave?

>>1222831
>FM, aka frequency modulation
Receivers preprocess the received signal with rough tuning, automatic gain control, and frequency downconversion, which is then passed to a phase-locked loop, which tracks the input signal's frequency. Once the PLL is locked, the filtered error output of the PLL will be proportional to the input signal at the FM transmitter, largely insensitive to the received signal strength. Once a PLL locks to a signal, it takes a lot of S/N degradation, usually a few to several times more noise than signal, to knock it out of lock.
This is also how manually-tuned FM radios stayed tuned in and accurate, even when adjusted slightly off-channel.

>>1222847
Oh that makes sense, neat.

>>1222831
>but how is this done?
The AGC circuit does this.
Automatic Gain Control
It uses a feedback loop in the 10.7mhz IF circuit to keep the gain somewhat stable.

>>1222861
>keep the gain somewhat stable
meant to say keep the LEVEL somewhat stable

What are some essential parts to keep around for prototyping, apart from the obvious resistors and capacitors?
Apart from an assortment of caps and resistors, right now I have some 4017s, hc595s, and 555s.
What else should I get for a decent assortment of parts?
Also, what's a decent cheap bench power supply?

>>1222871
>cheap bench power supply
I'm putting one of those LM2596-based constant current/constant voltage LED drivers/battery chargers in a housing (eventually) and it only cost me $8, though it is limited to maximum 30V, maximum 3A, and minimum 1.2V. But it's also common to use a computer PSU with an added breakout board with binding posts/banana sockets on it, which is a good option if you want well-regulated ±12V, +5V, and +3.3V power rails.

>>1222878
I'm not really interested in something ultra-cheap like that, just looking for a decent, fairly inexpensive (around $50 or so). I've heard that buying a used HP/Agilent or Power Designs supply for around that is cheap and will last forever, just want to know if there's any options with new models so I can avoid paying $40 for shipping.

>>1222881
The AliExpress search for variable power supply comes up with a $50 one plus shipping, though I have heard about cheap power supplies exploding. If you're looking for used then there isn't really any parametric search you can use, just browse eBay or whatever until you see something that suits you. That's what I've been doing with oscilloscopes and I still haven't found anything decent in my price range, but that's mainly due to shipping costs.

Hey /diy/

Long time dabbler in electronics here. Mostly arduino stuff but a few other things.

Just invested in my first benchtop power supply.

It is 13.8 volt fixed regulated power supply which can deliver up to 20 amps.

I'd like to put together a box that can sit on top of it and vary the voltage, ideally with a display for volts and amps and one that can handle 20a.

Any advice you guys can give? Point me in the right direction?

>>1222871
When it comes to bench supplies, $40 IS ultra-cheap. You can get a dual output bench supply for $50 and up, but always check the reviews before you chink it.
Linear power supplies are necessarily heavy because of the transformers and the heat sinks. The only way to get around shipping charges for them is picking them up.
Surplus stores are a good option. Does your local university have a surplus store? You might be surprised what they just want to have rid of and how little they want for it.
Moar below

>>1222895
>13.8V
Sounds more like a ham radio supply, actually.
>20A
That's a lot to ask. Are you sure that 3A or 5A wouldn't suffice? Will you need higher voltages as well?

>>1222881
>>1222895
Personally, I just use a 12V 10A switcher or so, bring it out to a barrier strip at my workstation, run my bench lights off of it, and use those little dollar-each step-up/down boards to get whatever voltages I happen to need at that moment.
Building your own bench power supply isn't that hard, really. Consider one of these https://www.aliexpress.com/item/180W-36V-5A-Single-Output-Switching-power-supply-for-LED-Strip-light-AC-to-DC/32267285465.html followed by Pic related if you really need all the comforts of a lab supply.

>>1222717
Pic related is all that I can find for low power fm broadcasting in New Zealand, and it keeps talking about a general user licence, so I assume it isn't the law that applies to personal fm broadcasters such as the ones you use when you don't have an aux input on your car's stereo. Are there any key words I should be looking for, or do these transmitters fall into the "unwanted emissions" table?

>>1222896

It's pretty huge yeah, the price was right though. Thinking about it I probably will never draw 20a out of it. I just want to vary the voltage for low voltage circuits.

I've never had a benchtop power supply so I'm obviously in the dark about what I need.

Can I make up a distribution box with multiple binding posts that I can plug many of pic related into?

>>1222895
>>1222901
People make stuff like that for ATX (computer) supplies, which incidentally might suit your needs better.

>>1222903

Is there a simple circuit I can make?

>>1222901
>Can I make up a distribution box with multiple binding posts that I can plug many of pic related into?
You sure can. Pic related contains some other possible interconnection ideas depending on your needs for permanence, convenience, and power draw.

>>1222905
Google 'diy atx supply' for some ideas.

>>1222897
>Are there any key words I should be looking for
unlicensed

>>1222913
Ok, turns out "the Ministry of Business, Innovation and Employment’s General User Radio License" refers to something that isn't a licence you need to get, and you can broadcast up to 1 watt without any licensing at all so long as you broadcast your station identification once an hour and stick to the allocated frequencies, which is pretty neat. But this definitely isn't what a pocket-sized local FM transmitter does.

A local online store notes one of these products with a "maximum allowable power" to be 50nW, so I could assume that that's what the law states, but since nothing is ever custom made for this small country and we always get consumer electronics designed for the European and American markets, I don't think I can bet on that.

i saw that i have an electret mic lying around here, are these things useful for anything beyond clap detection?

>>1222831
In FM transmissions (FM as in frequency modulation, not only the VHF broadcast band) the information is transferred by modulating the _frequency_ of the carrier alone, not the amplitude of the signal. Changes in amplitude are removed by the limiter (look at the waveforms in the diagram) by simply clipping the signal. This works as long as the received signal is slightly stronger than the noise generated by the RF front end (the technical term is 'capture ratio'). In principle there is no need for the automatic gain control (AGC) found in AM receivers (AM as in amplitude modulation, not only the Medium Wave broadcast band).

>>1222861
>The AGC circuit does this.
This is a misconception. FM receivers may have such a function (called 'delayed AGC') but for totally different reasons, mainly to prevent overload of the first RF stage or to drive the field strength indicator.

That didn't taste too good, but the result was worth it for sure, pic related.

>>1222980
Make a computer mic that isn't made of chintzy plastic? Harvest the electret to use for sketchy electromagnetism experiments?

>>1222986
So the "loudness" of the music you hear is completely independent to the amplitude of the broadcast waveform? I need to read into this, any particularly good sources on early and simple radio?

>>1222989
Yes, 1µV at the antenna input is enough for (almost) noise-free reception. That's the main advantage of frequency modulation, any 'static' known from AM (medium wave, short wave..) is simply clipped off. If you have a good FM receiver that allows you to disable the mute function (or has none) and you tune it to an unused frequency (or disconnect the antenna) you would hear a bone dry white noise that is as loud as any station. This is because a good receiver is designed such that the internal noise of the RF tuner section already 'saturates' the last limiter stage.

As to sources I would recommend old (pre-internet) ham radio literature like the ARRL handbook (www.arrl.org/handbook). This once was a complete compendium of all things related to radio. I don't know if it still is.

>>1223006
Neato!

Is it possible to transmit a 60 MHz clock with a 8-bit data bus synchronized to it using 20-cm jumper cables?

I've had a RasPi3 for a while and would like to attach a small little piezo buzzer to it as an alarm.

I've been looking at active buzzers on chink sites and came across a bunch. They all seem to have 2 pins (+'ve and ground), require 5V and look like pic related.

How do I add this to RasPi? I ask because I thought it would be easy to just hook ground to one of the GND pins and hook +'ve to one of the GPIO pins. But GPIO pins, correct me if I'm wrong, output 3.3V.

RasPi does have 5V pin but if I hooked it to that, it would buzz continuously.

Not sure what to do... maybe I should get some other kind of buzzer?

ideas? Advice?

thanks!

>>1223151

the gpio pins are limited to 16 ma and are 3.3 volts, so use one to drive a transistor which will switch the 5v power to the buzzer.

>>1223151
The buzzers I look up often have a variety of input voltages, one was 3-30V, another was 1-13V, so look around a little harder. And when it comes to powering them you could always take the signal from one of the 3.3V logic pins and feed it into the gate/base of a transistor that activates and puts 5V across the buzzer, but you will probably lose a little voltage on the way through.

>>1221846
chon electronics.

>>1223169
>>1223177
thanks.

>>1223151

the vast majority of piezo buzzers have 2 wires and need a pulse waveform at approx the resonant frequency of the element.
those that work directly from a DC voltage always have 3 wires (like pic) and a small circuit inside to generate the pulses. these are a minority so check the specs well.

also, it's odd that you'd be looking to buy one of these coz you probably have a dozen useless gadgets at home from which you could pull one.

>>1223151

the vast majority of piezo buzzers have 2 wires and need a pulse waveform at approx the resonant frequency of the element.
those that work directly from a DC voltage always have 3 wires (like pic) and a small circuit inside to generate the pulses. these are a minority so check the specs well.

also, it's odd that you'd be looking to buy one of these coz you probably have a dozen useless gadgets at home from which you could pull one.

>>1222927
https://gazette.govt.nz/notice/id/2017-go415

Low (MHz) 87.5000
High (MHz) 108.0000
Maximum Power dBW –50.0 (= -20dBm = 10µW)
Special condition 4 (= audio only)

That would be 200 times the usual 50nW or about 14 times the range.

Similar situation for WLAN ('WiFi'):
Low (MHz) 2400.0000
High (MHz) 2483.5000
Maximum Power dBW 6.0 (= 36dBm = 4W)
Special condition 13 (= digital modulation)

This in contrast to EU limit 20dBm (100mW) and US limit 23dBm (200mW)

Image: Want to go on air? Ali has it all.

>>1223006
Radio doesn't change much... Look at a book from the '30s (e.g. Terman) and you'll see that they figured everything out by then, except they only had vacuum tubes to work with.

>>1223035
Remember IDE cables?

>>1223236
except antennas.
not even including computer modeled things like fractal antennas, there's still the occasional oddity someone came up with in their garage

>>1223218
I have a bunch of junk around but don't have any active buzzers. I have no idea how to generate 5kHz signal to drive a passive one. that's why I'd like to get a simple active one for $1 and be done with it.

>>1221721
I wired up a USB>12 volt adapter to power my Nitecore i2 universal battery charger. The USB cable got really hot so I added another one. Used a penny as a heatsink, bending the clamp into shape was a pain in the ass. Now I can charge almost any battery using my 21 watt USB solar panel.

>>1223290
I added vents to the case later. Doesn't get hot though. Next I'm making a 19.5v adapter to power my laptop from my USB powerbank. But i'll wire that right into the battery pack instead of going through the USB ports.

>>1223218
That's exactly backwards. The 3-terminal buzzers require a driver circuit. The two-terminal devices you are referring to are speakers.

>>1223323
>That's exactly backwards

then why are you agreeing with me, dumbass? cant into reading comprehension?

>>1223336
Take your autie meds.

>>1223338
>>1223336

>>1223227
Holy shit thanks so much. I was actually wondering what kind of range you could get from one of those car FM transmitters, but I do plan on going full ham eventually, and that's a really good resource.

>>1223290
Penny heatsink? Are pennies still pure copper? My country's copper coins are just copper plated steel, I'll have to use an old piece of pipe bent into shape. Nice little mod.

>>1223323
>>1223336
All of the buzzers at my local shop have two wires, and the driverless piezo element only has two wires too. I've got no clue why one would need 3 wires.

Is this the proper place to ask how to fix my oven?

The bottom wire won't get hot, like in pic related. The top coil still works and everything else on the stove still works I'm pretty sure. Do I have to replace the coil?

I'm currently using Altium on windows and i'm very happy with it. Windows sucks balls however so I'm going back to Xubuntu and Altium doesn't work on wine. What's the closest linux pcb editor to Altium?

>>1223359
The coil is probably fine, you might have a bad relay. If so you can either replace the control board or buy a soldering iron and replace the bad relay yourself. You can get a good iron for $40 and the relay for $5, or spend about $80 (pic related) on a board.

>>1223371
Ok, I'll have to check to see what is what.

>>1223354
>going full ham
Do it, get a CEPT license. You will meet friendly and helpful people wherever you are on the globe.
>no clue why one would need 3 wires
The third wire is the feedback electrode for the oscillator, 1 transistor, 3 resistors.

>>1223371
>The coil is probably fine, you might have a bad relay. If so you can either replace the control board or buy a soldering iron and replace the bad relay yourself. You can get a good iron for $40 and the relay for $5, or spend about $80 (pic related) on a board.

>I have no idea of the brand and model of range you have but I know exactly what's wrong, how to fix it, and how much it will cost.

This place gets worse every day.

>>1223411
It's an oven. There's not a whole lot of variation.

>>1223411
>you might
Kill yourself.

>>1223412
True. If everything else works, then it's almost certainly a single relay, a transistor to power the relay if they need it (physical contact relay?), or the element itself.

Unscrew the element from the oven (after turning it off at the wall) and tape the circuitry of a torch to it to see if it works. If the element is fine, the light will probably go on, assuming it's a thirsty incandescent lamp. The thing should be between 3Ω and 12Ω, or 12-50Ω if you're on 220V (which Yank ovens might be?), you can do the maths. You could also borrow/buy a multimeter, which is probably a better idea. And if you're feeling gutsy you could always put the element on a few bricks outside and tie the free ends of a mains socket to either terminal and see if it heats up.

>>1223371
Maybe Autodesk EAGLE will do the things you need to do, and there is a native Linux version, and it used to be and maybe still is cheaper than Altium. KiCAD is libré and isn't bad and does have length-targeting and pair-matching tools, but it's still "in progress" and wonky af and doesn't have built-in autorouting and is slow at laying down traces in some modes.

>>1223412
>>1223413
>>1223439

Help me out here. Where is the relay - or the control board?

>>1223467

there are no relays on older units: the big power knobs do all the work. relays are only found in digital push-button wifi-enabled stuff.

so, only 2 things can go wrong: the knob goes bad (dirty or crumbling electrical contacts), or the element breaks. about 60% of the time its the element. the way to test it is to open the back and see if there's 220V across it when you turn the oven on. if there it, and it isnt heating, it's the element, otherwise it's the knob. replace the bad one.

>>1223467

I can't see either.

I don't think they're there.

You could try testing the elements for the resistance they specify, I guess.

Probably best to unplug the thing first.

>>1223471
Agreed, there's a fault somewhere and it's probably in the element or knob, but it might be in the connections from the element to the main board, like a loose wire or corroded contact.

>>1223218
>small circuit inside
beep

>>1223761

yeah, that guy may be right, but the vast majority of the two types that I have only have two wires and beep wonderfully off of 5v dc.

I suppose in the early days you had to send a series of pulses, but it's 2017 man, that's like having to add something to your fucking paint.

>>1223779
Because there's a small circuit inside. Neither a piezo disc nor a magnetic transducer beep on there own.

Electronic persons, can i join outputs of L293D (in my shops they didnt have powerfull L293) to one motor instead 2x motor? Like picrel?

can I solder to a bidirection logic converter with a heat gun. My soldering iron broke and its all I got. Kinda worried that a heat gun will break the board though.

>>1223858
cant you insulate what you dont want to heat?

>>1223848
sure, why not. but pay attention to heatsinking in the datasheet

>>1221721
>have bought shittons of microcontrollers and all kind of shit
>from raspberry over arduinos over esp8266s to attiny13As
>can't think of anything cool to build
>tfw

>>1223863
>sure, why not

some types of power transistors cannot be paralleled in a simple manner as discussed in pic related.


however it appears that L293s can be paralleled based on a quick search, but I'd google a bit to be sure.

>>1223893
Different L293s count as discrete components while structures within a single L293 do not.

>>1223893
>>1223863
I just connected them like that and looks like works, and its cold with small radiator on top, it drives liek 0,5A / 11V.

>>1223903
>structures within a single L293 do not.

interesting point. and they would be matched very well.

>>1223903
>Different L293s count as discrete components while structures within a single L293 do not.
Oh i dont know, they look like different modules that sharing only VCCs and GNDs, separate PWM, and even enable pins. Just like sharing supply only.

>>1223860
when the smd parts are on the opposite side of the solder joints.

>>1223890
Perhaps you should aim lower and think harder.
My next """project""" is going to be a stupid little PWM driver with touch switch for an LED lamp. The motivation: using up some of these stupid little PICs I have laying around.

>>1223962
Put the board level in a stable vise while heating. Surface tension will hold the underside parts on just as it does in production.
Or, work from below.

>>1223929
Because they are on the same bit of silicon made in the same process together as one part makes them very similar, matched. Problem with paralleling power transistors is one has lower resistance and takes all the current and self destructs, same die, process makes them very close so less of an imbalance plus being in the same package they are connected thermally too, which I don't know why is a good thing because a transistor heating up would increase its resistance and balance it in my mind but apparently I'm stupid or something.
Most drivers combined say in the datasheet it's ok to parallel them because they are well matched because they expect it because why else would you want a single package with multiple drivers in it that's going to be a nightmare to cool rather than individuals you can space out a bit and cool easily.

>>1223909
>0,5A / 11V
That would be within the specified 0.6 A per channel. I would not hesitate to draw twice the current from a single, paralleled L293D (1.2 A per channel) because the transistors are created in the same process and can be considered matched pairs. The main heat transfer goes through the middle pins (4,5,12,13) to the board which can be assisted by a heat sink on top.

>>1221721
my pops is an EE and has loads of components.. what are some cool projects i could build? (he takes them from work and it counts as r&d costs so im not wasting his money desu)

>>1223968
>being in the same package they are connected thermally too, which I don't know why is a good thing because a transistor heating up would increase its resistance
Vbe is the property of interest and has a negative temperature coefficient. Mounting on a common heatsink wouldn't harmonize the temperature and thus Vbe quickly enough to prevent one from running away.
>rather than individuals you can space out a bit and cool easily.
Sometimes you just want the convenience and board space more than you want the ultra low temperatures.

Anyone got the /ohm/ roll sheet handy for >>1223987

>>1223890
Make an ATTiny flash the ESP, make the ESP flash the ATTiny, see which dies first.

>>1223354
Only pennies made before 1982 are pure copper. They're worth like 7 cents per penny lmao. Some people order tons of pennies from the bank and use a machine to sort them, return the new ones and hodl the copper ones "until the US gets rid of pennies and allows people to melt/scrap them" or until they feel like secretly casting some ingots to sell

I have access to some windowed EPROM ICs, is there much they're good for? The largest is 2 Meg, the smallest is 16k.

>>1224002
Not without an EPROM programmer. Even then, not really, unless you feel like building a microcoded TTL CPU.

Has anybody taken an electret out of an electret mic and messed about with it around magnetic and electric fields? Just wondering if the force is noticeable to human hands. If not, I'll buy a few and fuck about with them and see.

Also making a pen with a microphone in the end sounds fun, a microcontroller would be necessary, right? Would extra non-volatile memory be a good idea too? I can already imagine a tiny SMD board crammed into a pen full of bodge wires.

>>1224171
>Also making a pen with a microphone in the end sounds fun, a microcontroller would be necessary, right?

WTF, dude, a micro-controller to handle a microphone? these are 2 completely separate worlds. one is made of tiny analog voltages, the other of number-crunching and decision-making digital calculating machines. there's no relation at all between the two.

>>1224177
Microcontrollers have ADCs, and he wants to use it for recording audio. Mic + opamp + stm32 or other powerful enough controller + microsd card would work.

>>1224296
And all that in a pen. That's something.

>>1224343
>And all that in a pen. That's something.

James Bond could do it.

>>1224350
But there was no James Bond.

>>1224343
>doesn't know about chip scale packaging
>anon's phone probably has half a dozen or more of them

dumbass here
i would like to get into electronics but I am slightly confused

volt is the pressure for the electrons to move
ohm is how sticky the path for the electrons is
amperes is how speedy they are

which means 5v through a 10 ohm resistor gives 0.5 amp right? just want to fix my microwave desu

>>1224387
Chip happens..

>>1224410
>amperes is how speedy they are
in a way, but more like charge transport per time
they need not be that speedy if enough flow in parallel

>>1224411
Does it ever.

>>1224414
like this?

>>1224426
yes, also inside the same conductor/wire. they can move quite slow. if you have a current of 1 ampere and a wire with a cross section of 1 mm^2 the speed is only about ten inches per hour. stickyness and pressure (or tension) are reasonable metaphors.

>>1221721
>>1221743
I'm a bit upset yeah.
I still haven't made any progress with it though.
Are the sub $400 3& printers on amazon good? I need one

>>1224431
3D*

>>1224410

>>1224426
>like
5v in serie with a 10ohm would give .5mA. in serie volt split and amp are the same, in parallele volts are the same and current split.

>>1224410
>amperes is how speedy they are
amperes is the amount of flow.. e.g. water current.

the speed is constant and irrelevant.

>>1224431
>3D printers on amazon good
There's a general for that where you would get more knowledgeable answers.
>tfw just ordered me a wanhao duplicator 7

>>1224296
Exactly, there's not point in having a microphone unless I can record the audio, though maybe putting a small analogue radio transmitter inside the pen would be easier.

>>1224453
>.5mA
You mean .5A or 500mA, right?

I fucking hate this bullshit. Is it the ads?

>>1224603
Ok, so looking at this schematic for a short-range (3-5m) FM transmitter, how does it work? The website states:
>Please use only 74F13 because 74S13 and the other ones does not work in 88-108MHz range.
So what's up with bodging 4 Schmitt trigger NAND gates together, and where is that 100nF capacitor going? Is there any legitimacy in this circuit?

Also:
>electrect

>>1223359
The control relay for broil (infrared from top) might be stuck on so that you can't bake. Google up a manual to figure out what's going on.

>>1224618
>>1223467
>>1223471

>>1224620
So, broken mode switch that is stuck in broil, stripped mode knob that doesn't make a clicking sound when you go to bake mode or failed lower heating element (bake) that never turns on?

>>1224436
Is volt going to fuck ampere?

>>1224611

i'd guess there's a close to zero chance that circuit is gonna work. the problem is the input voltage is in the millivolts, and the schmitt triggers need about a volt of difference between a high an low input to trigger.

but there are dozens of 2-tansistor circuits out there that will work. but a much better idea is to get an old MP3 FM transmitter. you can get 'em for around $2 at the thrift store, $10 new. some even have digital display of the selected frequency.

in any case, whether home-made or made-in-china the sound quality tends to be kinda shitty. very muddy.

>>1224646
It's a crazy abuse of a fast TTL chip but I'd lay even odds that it works. It almost certainly doesn't run in digital mode, much like all those CD40xx half-analog circuits we see from time to time. The mic input is probably pin 5.
They might be muddy because FM as broadcast has its high frequencies preemphasized. The 100nF input cap could be made smaller to fake this and the 22k might have to follow suit.

>>1224646
I was just looking at particularly compact designs for fitting into a pen, but those 2-transistor ones probably have both better sound quality and are more compact than a bent-legged DIP14 for cramming into the plastic barrel. Those MP3 FM transmitters (the car ones, right?) are neat, but are a little superfluous for what I want. If there was a neat chip prebuilt for recording audio with USB output I'd go with that, but for now transmitting the audio through radio and receiving it with a TRRS dongle that my phone records from is good enough for me, and it's all in analogue, which is more fun.

>>1224656
Ah, didn't see that pin 5 wasn't there. I think the 22k resistor to 5V is to provide the required input bias to the electret's amplifying FET, a larger resistor would probably mean less output amplitude, though I'm not sure how much, if any, voltage is being dropped across the resistor. But still, how do schmitt triggers fit into an analogue circuit?

>>1224656
>preemphasis
maybe it does phase modulation

I posted in previous thread about my subwoofer having taking a shit, literally. Two capacitors that has leaked.

I replaced them the other day and the sub now starts at least, didn't before. But I won't get any audio.
So something else got fried most likely.

How would I go on with finding the next faulty part?

I talked with the company that makes these subwoofers and they simply told me to buy a whole new amp for 200usd.

>>1224742
Do you have the circuit diagram?

>>1224744
Not that I can find right now.

The circuit board was for all of the settings such as volume etc. I'd have to contact them but I doubt that they will hand it out.

>>1224749
Did you try google <model name> diagram or <model name> manual?

>>1224755
First thing I did and nothing but frequency response and exploded blueprints.

>Pic related
If the voltage after the transformer and rectifier is ~10V, why are the capacitors rated 35V? For voltage spikes? If so, then why is the capacitor after the regulator even higher (50V)?

From Practical Electronics for Inventors.

>>1224771
>why are the capacitors rated 35V?
Because the author had enough of them in his junk box.
Are you recycling old posts?

>>1224771
>If the voltage after the transformer and rectifier is ~10V
The capacitors charge to 12.6V * sqrt(2) = 18V. (Minus diode losses.) Allow +10% mains voltage tolerance and use (the typical) 80% derating, and you'll end up with 12.6V * sqrt(2) * 1.1 / 0.8 = 24.5V. Then there's the transformer's load regulation (no load = higher voltage), but this depends on the used transformer. Something like 10% would be quite normal, so now you have around 27V. 35V is the closest standard value.
The lowest voltage 100n through hole capacitors tend to be rated for 50V. If you want the tiniest possible capacitor, 6.3V is the lowest reasonable standard value (5V with that 80% derating). Too bad many small cheap ceramic capacitors seriously suck when you run them close to their maximum voltage.

Also, see >>1224797. Never underestimate the significance of the author's junk box. I think one Linear Technology's application note even commented this, saying that often the suggested non-critical components in their example circuits are what the author of the circuit happened to have.

>>1224771
Derating. Electrolytic cap lifetimes double for every 10degC drop in temperature. I believe a similar formula for applied voltage applies but I'm too lazy to look it up.
For a true 7805, you don't strictly need any decoupling caps at all, just input stiffening caps to supply current for demand spikes and keep output ripple where you want it. For other regulator designs, you do need output caps, and the ESR of the cap is a critical component of the feedback loop.

>>1224797
>Are you recycling old posts?
No just a newfag i guess

>>1224841
>>1224884
Hmm I don't really follow completely, perhaps I'll go back to the chapter on capacitors...
Thanks anyway.

>>1224634

>>1224884
Has anyone here put a low voltage capacitor up to their power supply and increased the voltage until the cap went short-circuit? If so, how high above their rated voltage is ok? What about with reverse bias for an electrolytic or tantalum?

>>1225025

over-voltaged caps dont generally short-circuit. the usual progression is: chemicals start to boil, they deform, the rubber stopper starts to leak, smoke comes out, they explode. the voltage this happens at is very variable, but it's much lower when you reverse polarity. if you wanna be sure to blow one, connect directly across 120Vac.

>tries to measure resistance
>shitty DMM is showing over-count
>see lead is in 10A current mode
Well that could have been much worse. But I'm considering getting the cheapest DC clamp-meter off AliExpress just to avoid this shit.

>>1225025
Design the circuit so the caps are the next-rating above what the max voltage that node will see is.

'Lytics pop/leak when reversed. Tantalums are like little flashbangs when reverse-biased.

>>1225086
kek
>be me, checking out brake controller on my truck
>switch leads to 10-amp current terminal
>clip into brake lead and ground on trailer connector
>sweep through range on controller
>everythingsworkinggreat.jpg
>move clips over to turn signal terminal
>turn on turn signal for that side
>doesn't work
>look at meter
>fuckme.exe
>switch back to voltage terminal on meter
>try turn signal again, nothing
>check fuse block
>find "Turn signal, Trailer" fuse
>surpriseitsblown.mp3
>replace fuse
>everything works as expected

>>1225094
That was lucky. I hear you can get multimeter leads with blade fuse holders halfway down them, which you fill with a fuse of about half the rating of your multimeter's. An automotive blade fuse is much cheaper to replace than a Fluke one, which is why I've heard of mechanics using these with their $300 meters.

I should make myself one of those, especially since my black alligator-clip lead broke and I've been using the red alligator lead as the ground clip wile I probe about with the black pointy probe, but I'd certainly like to find a clip with stronger spring-action and less-malleable teeth. I also replaced the blown fuse with a piece of wire because I couldn't desolder both sides of the fuse at once, so I really should do something about it...

Can someone help with my power supply?

Got it cheap a few days ago its 13.8v fixed, 20a regulated.

I put a led spot on it, drawing a current of 1.6a and it dropped its voltage right down to 5.6v.

I opened it up and it looks pretty typical, no blown caps, no scorches.

Mains runs in, to a fuse, then a switch, then to a bigggg transformer with three secondary windings.

Two of the secondary windings are the same thickness and they go through 8 big diodes each to some caps and then they seem to run to four 2n3055 transistors on some big heatsinks out the back of the case.

The smaller winding on the transformer goes to a seperate board with its own smaller rectifier circuit, theres a whole bunch of resistors and zeners, couple small transistors, going to a to220 package transistor labeled d313 which seems to go to another 2n3055 which looks like it is connected to and controls the other four. So i have reasoned that this small board is some sort of regulator circuit reading from the binding posts how much current is being drawn and letting the appropriate current from the back 2n3055s through to the binding posts.

So my best guess at this point is that the 2n3055 on the regulator circuit needs replacing but thats mostly an estimate. Im still learning so i dont know for sure.

Can anyone help? Am i on the right track? Is there anything i should be testing?

Im figuring its the first 2n3055 and not one of the back ones because i think if it was the back ones id at least get the 1.6a my light needed from it

>>1225121
Posting pictures would sure help, but chances are it's a problem with the 2n3055.

Dumb question but I need to set digital pin 11 on an Arduino shield to LOW for >100ms to reset the board's factory settings. Will this sketch work? I'm new to this and don't want to brick the thing.

#define resetPin 11

void setup()
{
pinMode(resetPin, OUTPUT);
}

void loop()
{
digitalWrite(resetPin, LOW);
delay(1000);
}

It's bike light guy again, sorry. I've been messing about with the switching voltage/current regulator I bought and collecting a bunch of data, when I realised that instead of putting the signal into a triangle-wave + comparator PWM dimmer, I could just turn the current down instead with a potentiometer soldered where the current-adjusting trimpot now sits. There's nothing wrong with this idea and it's probably more efficient anyways, right?

I'm loving this circuit more and more, and wondering if you could just use the LM2596 itself to drive some really heavy-duty power FETs and make a 20A+ power supply. There's a limit on the current at 130mA, but that only applies to the constant current mode, and I can go lower when in constant voltage mode, so I hope I can go below the 1.25V constant voltage limit when in constant current (not yet tested), but either way I guess it isn't ideal for low voltages and currents as a power supply. What's the smallest voltage division that conventional power supplies can output anyways?

Thinking about making a variable/programmable load for testing batteries and power supplies, thoughts on my circuit? What do you think would be a reasonable voltage/current range to shoot for? I'm thinking having the ability to test loads ranging from 1v to 60v would be nice at around 20 amps max, maybe 30. Idk how easy that voltage range would be to manage without needing to switch between different load resistors. Also would like to be able to set and control for current, voltage, and power limits.
Mechanically, I was thinking of making the load resistor connect through binding posts or otherwise make it a seperate, external piece, so that I could make it water-cooled for higher power loads (by water cooled, I meant stick a power resistor in a cup of water and let it boil off).

Thoughts?

>>1225025
www.google.com/search?q=exploding+capacitor
Including instructions. As the sticky says, "Don't be afraid to get your hands dirty."

>>1225025
Depends on capacitor and overvoltage duration. Many ceramics can take several times their rated voltage, if you apply it only for a minute or less.

>>1225137
>60V, 20A
>1.2kW
g-good luck...

From what I hear, FETs themselves are used as variable load resistors, which is neat because they pretty much come with heat-sink mounting hardware, but for dissipating that much you'll need a heck of a lot of FETs or one solid italian salami resistor. Now I'm wondering if there's a good thing to put all that energy towards, like a water splitter or reverse-osmosis machine, or maybe you'd fancy refining some bauxite?

Since a variable load is typically controlled by voltage or current, you could pretty much just use a beefy buck/boost converter that responds to feedback of the power source, and put that power to good use. If you're clever, you could get it to charge your devices and stuff preferentially over mains, and burn only the left-over power. A water tank is a great place for excess energy to go too, and you don't need a massive heat-sink. Or just pump all that power into energy storage if you're feeling ambitious.

>>1225121
One of the others could also be blown or degraded. Can you check the voltages across the emitter and base on all of them?

>>1225135
Don't apologize, I've been missing the high-brow conversation. What's with all the fish, then?
>There's nothing wrong with this idea and it's probably more efficient anyways, right?
Indeed, you could do just that. The lamp itself might shift color slightly at different I(f) but nothing as bad as an incandescent.
>What's the smallest voltage division that conventional power supplies can output anyways?
1.25V is a fairly common but not universal reference voltage in power supplies and one generally can't go below that reference voltage. Pic related in >>1222896 takes the clever tack of supplying a negative voltage source from a large-ish negative supply using two diode drops to ground as regulators, thus allowing an LM317 a continuous output range that includes zero volts. It looks like you could use the same sort of trickery with the 2596.
You probably could make a high-current power supply using the 2596 as a controller, but it wouldn't be the optimal device. You would still need an external current limiting arrangement similar to the one on that board. dc-dc controllers designed for that purpose include such things as FET drivers with bootstrapping (higher Vgs for quicker rise time and less FET dissipation), current-mode control, soft-start, synchronous conversion (replaces Schottky diode with another, controlled FET for super efficiency), and usually much higher switching frequencies. The TPS40304A, for example, has a Vout(min) of 0.6V, a 25A maximum design output, a 600kHz switching frequency, and synchronous operation, if you can work with a fine-pitch VSON package. The LM27402 has similar specs with a 30A Iout(max) and high-side current sensing, in a TSSOP package which might be easier to reliably reflow.

>>1225137
A buddy of mine used an old waffle iron as a dummy load.

>>1221721
Quick question.

Trying to do simple AC ripple suppression for my mini speaker but I'm having a little trouble getting bits to fit.

Can I use the smaller inductor and it'll do the same thing (not same strength) as the larger round ones?

>>1225156
>pulled components
>really mobilizes the memories
You'd have to know where the noise is coming in from first. Are you talking about 50/60Hz line frequency coming in over the power supply? Neither of those looks like they will do much for low-frequency ripple but either could clean up a switching power supply's output a bit, in combination with an appropriate capacitor. To clean up ac line ripple, a larger capacitor and a linear voltage regulator will help more.
If not through the power supply, you might just try shielding your amplifier and everything on the input side with a metal can of some sort, grounded but prevented from shorting against your circuitry. Are you using shielded wire on the inputs to your power amp?

>>1225158
the left over AC noise from converting to DC I think. I'm not an electrician and this is my first attempt so I'm not 100% but I'm pretty sure since plugging the speaker into a power bank produces clean sound but a phone charger makes it buzz.

>>1225160
I'm also going with a chain of smaller capacitors to be able to squeeze them into the little places around the speakers (once I've wrapped the exposed ends obviously)

>>1225160
Could it be possible that you're overdrawing the power supply? In any case, phone chargers are by no means optimized for delivering ripple-free power. Perhaps a higher-current charger can supply the current you need more cleanly.
You'd probably need 1000uF or more to take the hum out.

Dumb question:
Can I weld with jumper cables and a 20S2P 10,4Ah 20C LiPo battery?

>>1225161
a cold joint and a short?

>>1225151
>Don't apologize
The apology was kind of a joke.

>high-brow conversation
Now you crack me up.

>What's with all the fish, then?
Don't get me started. I tried throwing those sardines into a pasta sauce, but since the sardines were whole I thought I'd try chopping them up with a fork. After stirring the pot a bit they turned into an unappetising grey mush. To mask the colour and smell I added a teaspoon or two of paprika. The first meal was edible, slightly on the tasty side of the fence, but nothing to writ home about. By the day I got the second serving out of the fridge to heat up on the stove and eat (only actually cooking once every 4 days is handy), the smell was overpowering, so now there's a horrid concoction of sardine tomato juice in my apartment complex's garbage bin wrapped in 3 plastic bags. I swear it's going to ferment and build up pressure before exploding and sending fish sauce everywhere. I ate pasta and a carrot that night. It tastes pretty good if you make the sauce with canned tuna, but obviously sardines are a different kettle of fish. That's the joke.

Good to know that the current limiting will work, but I've never heard about colour change. By the looks of things it shouldn't be too noticeable, especially since it won't be strobing between dim and bright.

>>1225161
You need soldering flux, friend. A temperature controlled iron probably helps also.

>>1225124

Bought a couple 2n3055 today, I couldn't post the pics til I got home but here they are.

The black metal part in this one has a 2n3055 that I just replaced. No luck.

>>1225151
I can do that, yes. What am I looking for? Just any difference between them?

I tested the resistance on the pot on a working model and it appears that the working one is 6k whereas mine is lower at about 5k. Not sure what it does in the context of the circuit but it couldn't hurt to replace it right?

I put my meter across the three secondary windings of the transformer just now. The two big ones come out at 18v and the little one comes out as 42v.

I'll have to compare it on the working model tomorrow but that seems like its probably ok.

>>1225190
The little one's 42? Volts AC before rectification or volts DC after rectification? Sounds a bit high unless you're running POE hardware.

>>1225203

AC

This is straight off the big transformer

>>1225184
Yeah I noticed that myself a few minutes after and fixed the short before plugging it into anything.

>>1225170
The ones I'm using are 50v 10uF but it was my only option that was small enough...
Seems to work well enough though.

>>1225185
I know, this is my first attempt at anything DIY though so I'm still getting my stuff together, just wanted to have a fiddle to see if I could accomplish anything a little early though.

>>1225181
Pls answer.

i always used OrCAD Capture to simulate circuits. although it sometimes is very frustrating i think it is a very good tool.

is there any particular reason to use LTspice? it feels so unhandy

>>1225314

orcad is for drawing relatively nice schematics. spice-type programs are simulators; the drawing part is secondary, which is why they tend to produce such maddeningly bad schematics.

also, the clientele is different. simulators tend to attract egg-heads rather than builders.

>>1225314

orcad was a schematic drawing program first, so it produces relatively nice schematics. spice-type programs are simulators first; the schematic part is secondary, which is why they tend to produce such maddeningly bad schematics.

also, the clientele is different. simulators tend to attract abstract-minded egg-heads rather than builders.

>>1225142
I'm not really to concerned about being able to dissipate that energy, a couple fat resistors in a pool of water should be able to sink a far amount of energy, and I'm probably not going to be testing anything that can put out 1.2kW for extended periods of time.

>>1225181
Unless you build a nice controlled power converter for it, no, the short circuit current will be too high to do sensible welding.

>>1225190
20% tolerance on pot resistance is normal. Don't just go replacing all this stuff before you know what's wrong, or you may end up blowing eprfectly good new parts.

>>1225188
Yes. First you need to look at the immediately adjacent componetns and sketch out a schematic. Then measure the voltages on the device and plot the voltages on the schematic. Any voltage that's way off of the others, assuming that they are all connected identically, or makes no sense for the component in question, is probably due to the broken part.

>>1225314
The library of Linear Technology parts built-in is one reason. Personally, I use KiCad and ngspice when I need to simulate something.

>>1225390
>a couple fat resistors in a pool of water should be able to sink a far amount of energy,

for a one-shot deal, anything is ok. but like that guy said, proper variable loads use power fets and huge heat sinks, and whatever air flow is needed.

it all depends on your situation. is it a one-off test, or do you want to be able to provide a variable load? variable loads are cool when you need something where you can turn a knob and it sinks 0 amps minimum, and maybe 50 or whatever amps max.

>>1225408
Plannning on building a variable load for testing batteries and such, though I wasn't looking for some sort of metrology grade device. something that I can use to control energy discharging.

Why FETs, as opposed to larger BJT's? Aren't BJT's easier to control the exact current, as the input-output current relationship is linear?

Also I figured that using power resistors for the load would be easier and cheaper, as I could swap out the particular load for my applications, and could also use the variable load for fancy heating projects later.

Also not quite sure what you mean by "one shot deal".

Anybody got ideas for fun projects to do with the scanner bar out of a printer-scanner? I recently salvaged an old printer for parts and thought it might be fun to play around with.

>>1225448
>why FETs
BJTs are more prone to thermal runaway.
Power resistors are easier but not cheaper, unless you go with the waffle iron.
>Also not quite sure what you mean by "one shot deal".
I think anon meant a current sink built as a jig for just one project, not intended to be reused as a general tool.

>>1225448
>input-output current relationship is linear
did you ever touch a data sheet?

>>1225448
You could try to go full analogue with some sort of comparator regulated PWM switch to a massive resistor, which just has to be 10 of ohms of nichrome in a water bath if you really want it to be. Not sure how easy it would be to regulate, you'd probably need some way to set what the current is regardless of the input voltage, which might be tough. There's probably a transistor based circuit out there on google, if you know the right keywords.

>>1223294
So I added a 3.5mm barrel adapter so that I could charge my car battery booster (which can power my laptop for an hour) with my solar panel. Had to adjust the converter to output 15 v, cool thing is that's still within acceptable range for my nitecore charger since car 12v outlets reach up to 15v while running.

>>1223294
So I added a 3.5mm barrel adapter so that I could charge my car battery booster (which can power my laptop for an hour) with my solar panel. Had to adjust the converter to output 15 v, cool thing is that's still within acceptable range for my nitecore charger since car 12v outlets reach up to 15v while running, so I don't have to adjust the voltage when switching devices.

>>1225481
Turns out my inverter runs fine off this car battery booster. Just had to get an ec5 cigarette lighter adapter. It only kept my laptop running for about an hour under an average load I didn't measure. But going dc-ac-dc is pretty inefficient. I ordered another voltage converter and will be going dc-dc soon, as well as adding an ec5 plug to my Anker 26,800mah power bank so I can use 12v devices (my laptop adapter in particular.)

Where can I buy flexible OLEDs? Are they even on the market yet?

>>1225470
I was thinking of doing just about this, but with a microcontroller between the comparator and the fet, using interrupts so that it's about as quick as it would be without it. This way I could have nice digital control over the thing, so I could do things like run pre-programmed load sequences and data collection. Possibly set the analog setpoint/reference point with a DAC. I do like the idea of using an analog comparator circuit to switch the fet instead of having to program and tune loops in a microcontroller though, I'm not exactly a control systems expert and I'm not sure if the control tuning would change based on the voltage of the test supply, initial thoughts tells me it would.

QQ, how long does a simple "if pin goes high set an output pin high" interrupt take? The actual interrupt process is pretty quick, right?

How do y'all like the idea of a half-can solar panel? With this and the sardine can I can tell I'll have a sleek futuristic bike on my hands!

Joking aside, I'm about to measure the open circuit voltage and short circuit current of the solar panel, plus currents and voltages with various load resistances (hence the Mtn Dew rheostat) to get a general idea of the kind of charging power I can be expecting. Since the panel just sits across the batteries with only a protection diode between, either this thing outputs little to no current over 4.2V, or it's power output to begin with is pitiful. If the last case is true then I'll drop it, but hopefully I can charge my battery when riding. Any idea of the kind of voltage/current curve I can expect?

>>1225590
the voltage drop off is dependent on the diode you used. what diode did you use?

>>1225573
Or a microcontroller with a built-in comparator, even? A lot of the smaller ones have them.
>Possibly set the analog setpoint/reference point with a DAC
Some micros even have that built in, though some of those don't have very many bits to the internal DAC.
>how long does a simple "if pin goes high set an output pin high" interrupt take? The actual interrupt process is pretty quick, right?
Depends on the processor, of course. A slow PIC might take over 5uS. A fast AVR8 could get it done in less than a microsecond. I think I've seem some micros that can route the comparator output directly to the timer output flip-flop, which would switch virtually instantaneously.

>>1225590
>Since the panel just sits across the batteries with only a protection diode between
Insolation can vary. I suspect the battery bank's charging circuit would be more tolerant of overvolting than Li batteries would!

>>1225601
No, the diode is part of the original PCB that connects the panel positive to the 18650 positives, and it bypasses the charging circuitry.

>>1225314
LTSpice is free. And since it's an advertising tool, it has a good library of LT's parts. Which is nice if you like to use their stuff.

>>1225336
Yeah, it was, but everyone I know uses it solely for simulation and has something else for the "real" stuff. And when their local sales representative offers courses, they're always advertised as simulation courses.
Also, it was PSpice before it became OrCAD Capture.

>>1225451
>BJTs are more prone to thermal runaway.
Unfortunately the modern-day mosfets aren't immune to that either. Almost all of them are optimized for switching and if you use them for high power linear stuff, you need to be careful.

>>1225687
tfw I spend at least an hour a day just messing about on LTspice
This shit is the new minecraft classic, when will we get spice: campaign mode?

>>1225690
Every simulator is an endless time sink almost by definition.

>>1225692
But I think I enjoy it too much. I was going to sleep a couple of nights ago when I got an idea for a type of switching voltage regulator that didn't use inductors or transformers, so I turned on the lights and plotted it out in spice. While I can't find complete Schmitt trigger comparators to try it properly, I don't think it would work. I went back to sleep a quarter hour later

The idea was that you'd charge a capacitor through a resistor until it was at the correct voltage when the comparator would stop it, and then it would discharge until the comparator turned on again to boost the voltage up again. Turns out the comparator has negative feedback and stays half on, meaning the transistor stays in the linear region and hogs all the power without regulating properly. Would this work with a Schmitt trigger?

>>1225694

sounds like you're trying to recreate an LM3909.

>>1225690
tearsofpride.jpg

>>1225694
You could add some hysteresis of your own with one resistor from op-amp output to op-amp + input. Ripple would be at a level equal to the hysteresis band and a frequency that varies by load current. You decide if you want to subject your load to that.

>>1222607
>>1222716
So, turns out my boss wasn't as much of an ass as I thought. When he said my job was learning, he wasn't speaking metaphorically. I was literally supposed to be reading and shit the past two months, not trying to find actual work to do.

It turned into a daylong back-and-forth, but afterwards it turned out there were a lot of misunderstandings.

Ask questions if there's disagreements I guess.

>>1225694
Literally there's like a 400+ or something on switched-capacitor voltage/power regulators. They can reach nearly 100% efficiency 'in theory', but the guy (pretty sure he has a PhD or two) just goes into total analysis, down to the microcircuitry level. I'll see if I can find it later.

>>1225721
>there's like a 400+
I am inferring that this is the amount of videos on some guy's youtube channel?

Technically if you have 0 resistance everywhere you can't charge up a capacitor to below the input voltage, so can never reach 100% efficiency, or am I missing something?

Also are there electrostatic relays that trigger on voltage not current (with capacitor plates and dielectrics/electrets)? Because they would be really handy for a few select uses. They'd use a little power when they switch, but to keep them switched you don't have to keep pumping current in like with an electromagnetic relay.

>>1225727
There are latching relays with two coils. I don't know about electrostatic relays, but I would imagine they'd have the minor problem of being sensitive to the voltage across/current through their contacts.
You could have 100% efficiency in the ideal world, but your ripple would be outrageous.

>>1225732
>ripple would be outrageous
Well yes, if you consider a 42% duty cycle 12V square wave to be just really badly regulated 5V, I'm not stopping you.

>>1225747
what's the worst that could happen, right?

Hey I am wondering if my multimeter is broken or if I am missing something. I just got it today along with the 9Vs.

The continuity is okay but when I test voltage it jumps from 0 V with differing decimals to the 50.0Vs then ~5.00 V.

>>1225883
It's current set to AC mode hence the ~. Try DC mode.

>>1225889
That did it. Thanks!

>>1225883
>differing decimals to the 50.0Vs then ~5.00 V
damn noisy environment you have there

is this a good trade to get into

>>1225747
Did you forget there's a cap on the other side? So really it's just 12V for an instant, which then declines smoothly as the load requires down to -2V, at which point infinite dV/dt brings it instantly back up to 12V. Wallah, regualted 5V with ±7V ripple!

So some dipshit at Arduino LLC decided to put the Arduino Zero SPI comms on the ICSP header, which has male pins.

Now I need a half dozen female-to-female dupont jumper cables to connect my SD cardbreakout to it, and of course of the thousands of dupont cables I have, exactly 0 are FtF.

How would you guys go about rigging up FtF cables? Chop MtF ones in half and solder two female halves together? I'm new to this and need to have my prototype functioning today, so I'll take any crazy ideas you guys have.

>>1225992
>not having a dupont crimping kit
Soldering two female halves together would work, or, if speed isn't too critical, you could simply use a solderless breadboard for busing.

>>1224431
How long do you think the battery life would be if you get the thing to not explode on 18650s?

Does anyone have a tl;dr on proper winding for a three-phase motor? I have a 1/3 hp single phase AC motor I want to redo for three-phase.

>>1226008
what kind of single phase AC motor?

>>1226002
I do have a good pair of wire strippers/crimpers that I'm moderately sure can handle crimping these things, but none of these arduino starter kits I bought have female headers I can crimp on.

I'm wiring up something to go on a drone, so I'm trying to avoid using a mini breadboard due to weight; didn't realize that would impose a speed limitation too. if I have to resort to that, what kind of speeds would I be able to acheive? Is there a max baud rate on them? I need to print ~1400 bytes worth of characters from my sensor at least twice a second.

>>1226010
Sorry, its a half-horse, 123D0101. It's an inductive motor, I believe reversing.

>>1226011
A solderless breadboard is pretty capacitive. I haven't mathed it out but I wouldn't risk using them for anything much faster than 10MHz. 5MHz should be entirely worry-free.
If you just spliced two halves together for each signal, you'd be fine for SPI or async serial at any speed.
An SN-01BM and https://www.aliexpress.com/item/1300PCS-2-54mm-1P-2P-3P-4P-5P-6P-7P-8P-9P-10P-Dupont-Housing-Plug/32593491588.html are really, really nice to have around. They have a male terminal assortment too if you're into that sort of thing.

>>1226016
>They have a male terminal assortment too if you're into that sort of thing

Sorry I don't swing that way.

Hmm, could I use wire nuts to connect the two female cable halves? Do they make them for wire gagues that small?

>plug speaker into power source
Nothing bad happens
>Connect phone to speaker with 3.5mm
Nothing bad happens, can listen to music great
>Connect phone to charger/power source
Loud annoying buzzing sound comes from the speakers

Can someone please help? This was my first attempt at a diy projects and I'm going to have to start fresh because of this issue. I can't figure this out on my own, I've never been good at studying, I learn best when advised by others.

>>1226019
I don't think so. I'd just splice and solder together with heat shrink over them.

>>1226020
Did you try Pic related yet?

>>1226020
most likely the pwm noise, from the supply, is getting onto the usb shield and capacitive coupling adding noise onto the output. seen it happen on plenty of laptops, sometimes even the hdmi port would have troubles

>>1226031
No, but thank you.
Is that just a capacitor? But inline with the ground 3.5mm line only?

I'll give it a try for my next build.

>>1226033
Yeah, I figured since my better phone chargers make it happen less but the cheaper ones are simply intolerable.

>>1226054
Yes. The idea is to lift any dc ground loops that might be present. In pro audio that problem occurs a lot, and they often have ground lift switches on inputs for just this reason. The capacitor should be fairly large, but nonpolar.
Also, if your inputs are connected to headphone jacks and not line level outs, be sure you're loading them enough, say, with 62 ohm resistors from each input to ground, on the jack side of the capacitors.

if you have a common collector amplifier and want to drive a small impedance, as far as i understand the output impedance of the amplifier has to be about the same as the impedance you want to drive

but what if you have a push-pull amplifier? i assume the same criteria applies. is it just about choosing the right collector current? i dont really understand how to design those properly

>>1226063
>same criteria applies
No. Look up push-pull class AB.

>>1226060
Thanks for that.

The strangest part is that it even buzzes when I touch the phone.

Could I use something like this? I get all my parts from scrapped/junk parts like ATX power supplies or blown wall warts, so my choices are kinda limited. If I order online it could take over a month to arrive.

>>1226072
i looked up tons of sources regarding class AB and they are all about the general idea of the amplifier, which is clear. but impedances below 100 ohm are still a problem

>>1226005
84Wh or 20,000mah, so about 10 days or 35 hours of screen on time doing web browsing and gps or something.
Why would it explode?

>>1226250
Nice.

>why would it explode
I have no clue how electrocity works, to me it's just all pixies floating around and from what I've heard 18650s are filled with a lot of very angry pixies. I guess the voltage is the same but for some reason I was expecting 18650 to push like 3 amps through just because it can and didn't see resistors or fuses or other pixie-wrangling devices to prevent this.

Yes I am retarded.

>>1224431
Would love to do something like this for a small handheld emulation machine.

It's a shame even the SD820 struggles with PSP emulation.

>>1226250
>Why would it explode?
Because the cells themselves don't have the protection circuitry of a normal LiPO. If you charge them too quickly then up they go. Same if you discharge them too quickly. If you discharge them too low they're permanently damaged, and if you charge them too high they're also permanently damaged if they don't catch fire. Since you have 6 cells, presumably in parallel, you'll have to give "6 cells worth of current" to charge them at a reasonable speed, which takes particular circuitry. Chances are you can fit one of these chips in the gap beneath the camera, but if anything in your construction goes wrong and two cells get shorted, then say goodbye to some combination of your hand, ear, and thigh.

Can you listen to AM radio without filtering off the carrier frequency if you amplify the signal in the receiver with a 20Hz-20kHz op-amp?

>>1226282

yeah, you can listen to AM radio using just any old high-gain amplifier. in fact, it's a major headache when you design a microphone amplifier, for example, and it starts playing music by itself. old tape recorders did this too: if you pressed play without a tape in it, you could hear AM radio in super lo-fi quality. and no way to tune to a diff station.

>>1226275
It will not short due to jumper wire layout, at least the final design.

I fully intend on using the phone charging circuit and voltage regulator. I know it will be slow, but meh.
Also the Panasonic cells I intend to use do have cell protection. They have a little board on the end

Anyone used a Protoshield before with an Arduino? I need to access the MOSI and MISO pins from my M0's ICSP header, but the SPI part of the protoshield documentation says

>On the ICSP connector only 5V, GND and RST are wired to the respective pins on the header. MOSI and MISO are present only on the connector pads.

Where the fuck are the "connector pads"?

When a transformer specifies "240V in, 12V out" is that 240Vrms and 12Vpeak? That factor of 1.414 makes a difference. I'm thinking of simply buying a cheap mains transformer and using it as a flyback transformer.

Also if the voltage gets too high on the end, since I want to keep it AC, can I just put two Zener diodes facing opposite directions in series across the output? I think the shorting will help stop the ability of the transformer to act like a current source, though I'm not sure. I'm only putting a square wave in, how high could the voltage get over the coils ratio*Vin?

Also just gonna leave this here...

>>1226302
Followup to this, this is part of the Arduino.cc R3 Protoshield schematic. Do these circles with the X's mean the pins go nowhere on the protoboard?

Surely Arduino would not go full retard and not connect the MOSI and MISO pins from the ICSP header...?

>>1225014
t-that looks kinda l-lewd

>>1225014
Is AMP a POWER BOTTOM?

>>1225405

This was very helpful, i tested the 2n3055s on their own and again under load.

I found one that didnt match the others so i replaced it.

Power supply now pushes out a nice even 21 volts.

While this was enough to power my light and not blow it, it still isnt the 13.8 it should be.

Something seems funny on the regulator circuit and i am suspecting the d313 transistor. Im gonna buy one and just replace that cause they're cheap but what else could i be looking for?

>>1225727
>so can never reach 100% efficiency
Switched capacitor converters can reach 100% efficiency (well, close to it) as long as you charge the capacitors fully. This limits the output voltage to some multiple of the input voltage, like -1, +1.5, +2, etc. The key to near-perfect efficiency is to keep the capacitor voltages nearly constant (fully charged).
This "charge the capacitor only partially through a resistor" idea has the same efficiency as a fixed-ratio converter followed by a linear regulator. Your resistor burns the excess energy.

Here's one charge pump which uses a variation of your idea (it varies the resistance of the capacitor charging mosfets to attain regulation): http://www.ti.com/product/tps60151

>>1226335
I would make sure that any voltage references in the circuit are doing the right thing. It could be a zener or other voltage reference failed open, or some element of the voltage feedback loop is behaving badly. You might need to reverse-engineer the regulator part of the circuit (a useful skill to develop in any case) and probe around to check theories.

>>1226082
The push-pull stage is basically two emitter followers. They have much lower output impedance than input impedance. Also, their gain is about 1, so you need something else like a common collector amplifier for driving them if you want higher gain. Which, in turn, can be dimensioned assuming the push-pull stage's input impedance is output impedance / hfe.

>>1226308
> is that 240Vrms and 12Vpeak?
No, it's 240Vrms in, 12Vrms out. (Or 240Vp-p in, 12Vp-p out.)
>That factor of 1.414 makes a difference. I'm thinking of simply buying a cheap mains transformer and using it as a flyback transformer.
Sounds like you have misunderstood something. If nothing else, at least what a flyback converter is.

i need to step down 220v house current to 5v so i can plug my arduino to vaccuum cleaner and have it light up

is there an easier way?

>>1226358

disregard that i found what im looking for

http://www.circuitbasics.com/setting-up-a-5v-relay-on-the-arduino/

hey electronics general

I am looking for digital timer to use on my small solar wifi project , i got one of these (12v version) although its a bit too chunky. I want something to automatically cycle 12h/12h

much appreciated

>>1226366
why not just use the sun m8

>>1226366
if youre already coding onto a controller could you not use another for a timer?

>>1226369

my dodgy project bleeds too much power and doesn't start up idk. however it works fine if i let it charge like 1-2h in the morning.Im sure there is work around but i am not so smart.

would be nice to cycle it 9am-6pm but 0900 to 2100 is fine.

>>1226344
>same efficiency as a fixed-ratio converter followed by a linear regulator
Just what I was thinking.

>>1226352
Not flyback converter, transformer. A flyback transformer simply refers to a transformer where the output waveform is not intended to be the same as the input waveform, while the converter is a DC converter. Unfortunately, I can't find any information on the voltage output of a square wave fed into a transformer unless it's got an AC-blocking diode on the secondary, especially in regards to the turns ratio compared to the voltage ratio. I tested it in spice and it told me I'd get an exact voltage ratio with in-phase square waves on both sides, but I'm not sure if that's correct. Pic related. So is this not to be classified as a flyback transformer at all? I thought output waveforms had derivatives or something in there, because inductors I guess.

>>1226375
Photodiode and FET controlling the power? Or a half-LDR voltage divider to trigger the FET. You'd barely be wasting any power if you do it right, and it wouldn't activate on dark days.

>>1226375
What do you mean by "too chunky"? Form factor? Doesn't keep time? Draws too much power of its own? What exactly needs to fit better?
Draw us some pictures?

>>1226383
>I tested it in spice and it told me I'd get an exact voltage ratio with in-phase square waves on both sides
That's exactly what a transformer does, unless the core (saturation, resonance) or windings (self-capacitance, insulation withstand voltage, mutual capacitance, etc.) give it reason to do otherwise. Note also that your voltage source drives both sides of the square wave with unchanging source impedance, which a single-ended switch and freewheel diode would not do. Add a diode and see if waveforms look different.

>>1226388
If I set the impedance of the primary side to 5Ω I get barely 12V output and only in little spikes, and if I add a diode it goes to 0.6V. Should I be trying to amplify a square wave instead, because this turned to shit faster than those fucking sardines.

>>1226082
No you didn't and no they aren't.
>>1226347
>two emitter followers
a.k.a. common collector
>common collector amplifier for .. gain
make that common emitter

Hi friendos,

I am wanting to build an amplifier and i have 0 electronics knowledge. Here is what I got:

An old Topping Amplifier box with all RCA/Speaker terminals and shit that no longer works.

I have bought a Sure-Hifi Amplifier board: https://www.parts-express.com/sure-electronics-aa-ab32189-2x100w-tda7498-class-d-amplifier-board--320-303

What do I need to turn that Amp Board into a fully functioning amplifier with the stuff I have in my possession? I've got a solder but thats probably all I have.

I noticed that the board doesnt have a knob.. so will I be able to use my existing toppings knob components to control the volume?

Thanks for any help you can give me!

>>1226393
Try modeling the FET to ground with the diode, in the traditional flyback style. I haven't tested but I think on the secondary you'll get a pulse that keeps growing, which may flatten off if you saturate your transformer, followed by a sharp, deep drop as the FET cuts off and the diode opens.
I think you wanted a sine wave on the far side, didn't you? Then you'd need to drive a clean-ish sine wave on the primary, probably with a bridge-tied load to reach the desired amplitude, or you'd need to make clean dc on the secondary in the traditional flyback style and use some high-voltage transistors to shape your wave there.

>>1226393
>>1226402
>includes Pic related
>doesn't mention Pic related
Also, that Pic related is one way to get a clean-ish, with heavy emphasis on ish sine wave in case your transformer has a split "primary". You could set up the FETs as a 3-bit DAC that doesn't even have to be strictly linear. You could trigger them with a 4-bit binary counter a 4-to-16 multiplexer, and some OR gates.

>>1226398
>HURR DURR I KNOW MORE THAN YOU
never change /ohm/

In case I'm not around to bake the new bread, someone ordered Pic related a while ago.

>>1226406
THIS has gotta be a joke

>>1226412
yes

thats the joke

>>1226402
That doesn't look terribly ideal for not wasting power through those resistors. I was thinking of just amplifying the triangle wave from between C3 and R4 on the relaxation oscillator I'll be using, pic related. I can make the wave more triangular by increasing R3 with respect to R5. I just have to figure out a quick and dirty way to use a FET to amplify the signal linearly by a factor of 2 or so, which might be tough since the triangle wave signal needs a pretty high input impedance, so I'll probably use another of my opamps as a buffer.

Just testing things I think I have to send the triangle wave into the transformer with respect to ground and not either of the +/-2.5V rails, which might prove difficult. But with big enough capacitors it should be fine. But since my inductances aren't giving me any voltage ratio other than 1, I think my new spice command isn't quite right.

>>1226401
ATTENTION TO THIS DUDE! HE NEEDS HELP AND HE NEEDS IT NOWWWWW

>>1226372

If my design was in top nick then yeah sure, that could help it work more efficiently. which it is not. Its a frankenstein of various cheap china bucks / boosters. Works great however me having to manually turn it on and off. If i was to drill into the roof i will not be able to.

>>1226383
>Photodiode
>FET

that's actually pretty good. I could use that at least cut off power in the evening and save same of the battery charge for the morning start up. thanks ill look more into it. This is promising holy shit i think you've done it. thanks.

>>1226384
yeah the form factor is way too large its a fat piece of shit i want it the size of a casio watch. i rather not butcher that timer that i have, it was expensive..they are made for 120/240v (mine is 12v) its probably 3times the volume and twice the cost of my cheap project.If it was smaller id make the exception and would take it over the photo/fet suggestion.

Rather not further embarrass myself anonymously with pictures. Thanks though =]

>>1226426
I love carrots, heres some soup I made with carrots

>>1226398
>make that common emitter
Right, stupid brain fart.

>>1225992
AVR ICSP is SPI, that's why they share the header

>>1226401
You need a voltage source between 15-36V, connect that between the VCC and GND terminals. Plug in audio to the two RCA jacks, and speakers into the output terminals. There is no analogue volume control.

>>1226428
I bought 1.5kg of them because it was cheaper than a single 500g bag. Now my fridge's vege-bin is full, but they last for ages so I'm good. Beans are my second staple vegetable, can never have enough beans. Made a stir-fry tonight with just beans and carrots, along with everything else in my pantry. Ginger, chilli, honey, garlic, cornflour, and some chorizo. Kinda weird but damn tasty. Better than those fucking sardines for sure.

>>1226458
well done sir, i like weird recipes that are familiar only to yourself and probably weird to others

>>1226454
I mean how do I mod it with analogue volume control and into a commercial like product?

Anyone got any experience making their own terminal connectors. I have access to a 3d printer if that helps.

>>1221721
I have a question about the book recommendations.

Are those books all meant to be read as a sequence or do they all cover the same things with varying degrees of rigour?

I ask because buying all of those books will get expensive (I like physical copies) and many have bad reviews on Amazon and aren't really clear about what it is they're supposed to be teaching.

I try to make new control board for my coffemaker and connect it to existing high voltage board (pic related). Basically I need to close the relay using connector on the right. How do I do it? Shorting 2.5V and 4.14V to GND desn't work.

>>1226727
top side. The big square on first picture is old uC.

>>1226420
>not wasting power
You could also tune the Rds(on) via Vgs or FET selection. It was an idea, anyway.
>amplify the signal linearly by a factor of 2
Easily enough done as long as your opamps are rail-to-rail all around. Also, linearity doesn't strictly matter if the distortion is at the rail side, as it'll make your wave more siney.
>send the triangle wave into the transformer with respect to ground
Right. You'll end up bridging it whether you create a high-power virtual ground or a complementary signal. Linear amplifiers are not particularly power-efficient as it is. Now you get to make two.

>>1226458
>mexican soup
>ginger
So close.

>>1226727
2.5 is probably a divided ac line sync to enable zero-voltage switching. To protect the pump you would ideally wait for that to cross zero volts then switch.
4.14 is definitely the switch input. R9 is pulling it to ground. Try shorting 4.14 to 5V.

>>1226713
Not actually my own recommendations, just reposting the standard pasta.

NEW BREAD

>>1226748
>>1226748
>>1226748