/diy/ - Do It Yourself

It would be helpful to know what you're working on and what you're trying to do with it, because (no offense to you intended) there may be a better way to do what you're trying to do and it would ultimately benefit you to know what that is, if that's the case.

Lots of words to say little but do you get what I'm saying?

>>572661

It's a 70's console unit. A while back I was trying to fix (what ending up being) a blown power transistor for the L channel. I systematically went Transistor to transitor looking for the bad one, starting with Q1 (transistors are marked Q1,Q2 etc, on this set). I eventally found the bad transistor, but I screwed up the FM amp ones in the process. What's happenig now, is that any FM station will come in loud for 2 seconds then fade out instantly, then resume after 30 seconds or so, and repeat. I KNOW that didnt happen before I messed with the transistors, so I'm pretty sure I just didnt solder one back to the board correctly.

So, tl;dr I want to check the FM amp transistors to make sure they're making contact with the circuit.

I've been an electronics tech for over 30 years (no I'm not kidding, and please leave off with the crap about my age, ok?), spent a fair amount of that time fixing broken shit, and here's what I recommend:

You really can't test transistors (or any other component for that matter) in-circuit. The exception here is if you have an oscilliscope and a schematic, understand how things are supposed to work, and can follow the signal through the circuit in question until you find the stage that's the problem; I'll assume you don't have an o-scope so we'll skip that.

Any suspected component needs to be removed from the circuit in order to test it properly and definitively.

In the case of a bipolar transistor, don't use the diode junction test on your meter, it's not sensitive enough. Use the resistance setting. Check base-emitter and base-collector, both forward biased and reverse biased. Obviously there should be some conductance in the forward direction, but there should be NONE at all in the reverse direction. Then check emitter-collector in both directions, there should be infinite resistance in both cases.

Checking diodes should be self-explanatory.

Other things you could look for:
o Power supply voltages that aren't what they should be
o Cold solder joints (re-solder anything even remotely questionable)
o Cracked traces, especially around through-hole pads where you've been working
o Bad electrolytic capacitors (bulged out, electrolyte residue on one or both legs, etc)

>>572664

Thank you for that, I replace the caps when I first got it, so that's all good. and dont worry about the age, I visit AudioKarma pretty often so it's all good man.

The more I look at the solder joints, the more I think that I need to get my gun out.
Now what happens when the little metal 'O' that surround the hole in the board comes off? I'm pretty sure the solder contacts that which sends it to the circuit...

Secondly, there are voltages at each end of the transistor in my original picture, how do I test those?

>>572666
>The more I look at the solder joints, the more I think that I need to get my gun out.
..OK, here's something that needs to be addressed immediately:
You need to get a pencil-type soldering iron, 30-40 watts maximum, for working on PCB's.
If your "gun" is what I think it is, it's better suited to soldering rain gutters than electronics. It's too big and too powerful and will ruin PCB's and some components.
You don't need an expensive one, $20 will go a long way.

>Now what happens when the little metal 'O' that surround the hole in the board comes off?
"Through-hole".
Is it a double-sided PCB (has traces on both sides)? If the plated-through-hole delaminates from the PCB, then it *breaks* the connection between the top and bottom traces. If this is happening, it's because what I said above: Your "gun" is too powerful, too hot, it's ruining the PCB. Run some wire-wrap wire through and solder it on both sides, or solder the component in that hole on both sides, to restore the connection, and get a proper soldering iron so you stop delaminating the PCB.

>>572666
>Secondly, there are voltages at each end of the transistor in my original picture, how do I test those?
Don't bother with that, use the methods I discussed above to identify components that are failing.

Of course based on your other posts I now suspect you have broken traces and broken plated-through-holes that need to be repaired.

>>572673
Sorry, gun was just the term I was using, but I actally use whats in the picture. It hasn't let me down let, and its only 30W.

I'm going to visual check each transistor solder spot and see how it looks. I'll let you know in a few mins.

>>572675
Yeah, that iron is OK. Not sure what your soldering skills are like, just remember that it's a 30+ year old PCB, and the longer you keep the iron on a pad, the greater the chance it'll delaminate from the PCB.

Do you have a desoldering pump? Similar to this pic? If not and you're using desoldering braid, consider getting one like this, it works much better than braid.

>>572687

i've learned that the hard way, on other projects. I currently use braid, but i think I'll get something similar to what you show. seems like it will do the trick better.

I think i found the bad solder joint. Im going to resolder it here soon

>>572687
>desoldering pump
these can cause solder to spatter. not an issue for old through-hole electronics; but on anything modern & surface mount, it can cause mysterious problems.

>>572725
Well since the OP is working on 40 year old electronics it shouldn't be a problem.

>>572656
Hey OP you still working on this?

>>572694
god i love old boards with bendy traces

>>572808
That's long before the days of PCBs being laid out on a computer, everything was hand-drawn then shrunk down to 1:1.

>>572805

yeah. i lost my soldering gun actually. i found it about 5 minutes ago, but I now I can;t find my desoldering braid...

OP here,

so I fixed that one, but I'm still getting drop outs.
basically FM stations come in and fade out instantly and the signal intensity needle doesnt even register on FM, but works fine on AM.

I attached the schematic. the entire upper row is FM circuitry. the intensity dial is M3 which is upper 2/3 right area (look for the big arrow and follow it down.

due to the size I cant post the full thing, but it's here: http://i.imgur.com/xKKcsFs.png

>>572855
Do you have a small audio amplifier you can use to trace signals with? Connect it to the negative side of C59 (which is the mono FM audio signal). Does it still fade out? If it does then the stereo demux is the problem.

You might just need access to an oscilliscope to find this problem.

>>572855
>>572855
Let's back up a little, here. Earlier you said:
>I eventally found the bad transistor, but I screwed up the FM amp ones in the process
Which one(s) and how do you figure you "screwed them up"?

>>572861
Well I know I dont have an o-scope.. but what kind of amplifier thing would I need?

>>572863

I desoldered one end of each transistor to test it (bringing it out of circuit). By desoldering/resoldering with amateur skills, I think I either messed up the trace around the hole or just did a crappy resolder job.
I started at Transistor 1 (Q1) and worked my way down the list, so I desoldered at least one lead on most of the FM transistors. I didnt touch any of the other alighnment settings,

>>572940
http://www.zeitnitz.de/Christian/scope_en

Maybe something's wrong with the AGC? Look at stuff connected to line 13.

>>572971
>http://www.zeitnitz.de/Christian/scope_en
That's not going to work at all for RF, that'll only work for low frequency signals, and trying to push it beyond 20kHz, even if it works, would distort the signal so much that it would be essentially useless. I'm sure the OP appreciates the suggestion though.

>>572940
>I desoldered one end of each transistor
OK, as per the instructions I posted above, you need to *completely* remove transistors when you test them, or you're "testing" everything else it's connected to as well, making your ohmmeter readings meaningless.

>>572940
>but what kind of amplifier thing would I need?
Just an audio amplifier.

Tell you what: You can temporarily remove C66, and solder a wire into the hole that does NOT go to IC1, and connect the other end to the positive side of C59. That bypasses the stereo decoder entirely. Also, when the tuned-in station "fades out", does that co-incide precisely with the Stereo indicator light coming on?

>>572971

Thanks for the help, but as >>572998 said, it probably won't help in my situation

>>572999
I'm just going to buy some new transistors while I'm at it, and get some more desoldering braid at radioshack.
I'll give the bypass a shot when I get back from the store.
Yes, both the station will come in and the stereo light will glow and then both will fade out simultaneously..

>>572999

Ok, I bypassed the MPX by going from C59+ to C66 (out) but I'm still getting the dropouts...

>>573162
OK then you can rule out the stereo decoder being the problem.

You say your 'tuning' meter needle doesn't ever move, or does it move then drop at the same time as the output?

>>573171

The tuning meter, as in a little needle dial that goes up when a station is a stronger signal and goes down when there's no signal. And it doesnt work at all on FM mode, stereo or mono, but works fine on AM mode

>>573172
Did you go through and check all the possibly affected transistors using the method I described above (>>572664)?

Check all the different power supply voltages against what the schematic lists them as. If any one of them is missing then there's the problem.

There is a multi-pole switch (S201). Check it carefully with an ohmmeter to be sure that it's actually making contact on all poles when it's in the "FM" position.

>>573174

Some transistors that I removed and checked had low hFe values, they read 151 instead of 200 (for a 2N2222). Is this an issue?

i'll check the voltages and S201 and I'll let you know

>>573179
Listen to me: Forget that your meter has a "diode junction test" and a "transistor test" function, they are crap and will not show you marginal transistors. Check them the way I described to you and no other way. Use clipleads if you must to keep your fingers off the ohmmeter probes so the stray resistance doesn't give you a false reading.

>>573179
Also you're not arbitrarily replacing the original transistors with just any old transistors are you? If you replace a transistor you MUST use an equivalent part or the radio is NOT likely to work.

>>573181

Right right, sorry. I'll do the Ohms testing.
And I didn't replace them, just removed them to test and put them back in. I have equivalent transistors on hand. 2N2222 are proper replacements for the ones in the unit (GE part number GE-20), so that's fine.

>>573174
I checked all voltages from 1-5 (power) and 6-7 (FM) and got ranges +/- 1v.

>>573200

Also since you've come this far, here is a link to the full service manual.. https://www.mediafire.com/?kcf1zrgubvas1ou

bump for advice

>>573476
Have you done everything else I've suggested you do?

>>573481

yes, I checked power sources as per >>573200
I put deoxit in S201 ( which is one big slidey switch) as per >>573174
i checked Q1 initially and got conductance in fwd direction but not reverse as per >>572664
I will continue to look for cold solder joints

worst case, I live without FM radio on this thing. AM works, Tape In/out works, Phono works, speaker works...

Your problem sounds like a capacitor is in backwards,like it's getting charged and being unable to conduct sound, then discharging and being able to conduct a little bit

>>573491

Good suggestion. I have found one so far C60 (which is within the MPX circuit). in my defense, the - side wasn't clearly defined on the board. HOWEVER. I am still getting dropouts so I will look for more.

>>573484
>I put deoxit in S201 ( which is one big slidey switch)
that's not actually what I suggested you do, I suggested actually checking it with an ohmmeter. FYI contact cleaner is largely useless anyway and won't fix a worn-out switch.

>>573503
You've already ruled out the stereo decoder as possibly being the problem though.

Unless you know you removed and re-installed a component you can rule out anything being "soldered in backwards" because this is a production unit that did at one point in time work properly, not something that never worked ever.

>>573545
Here is a picture of the switch mechanism, where do I put the leads to check it?

>>573549
Well, when I first got the unit I replaced all the electr. capacitors, which is why I'm rechecking them, I replaced about 20 or so, so I want to make sure I didnt foolishly put one in backwards

>>573550
You have a full schematic, from that you should be able to figure out how to check the switch.

Chances are if you put an electrolytic cap in backwards you would have known by now because they get extremely upset and tend to explode when you connect them backwards.

>>573617

I am looking at the schematic and the switch leads, but honestly i have no idea how to test it..

the ONLY cap that was backwards was a 2.2uF, really tiny guy. I lucked out on that.

I'm still trying everything I can but I'm running out of time

>>573862
Come on guy, you need to apply yourself here. A slide switch, which this in essence is regardless of being multi-pole, is easier to figure out than, say, a rotary switch in this regard. Take a look at how the switch works mechanically-speaking, look at how it's footprint is laid out on the PCB, then look at the schematic; how do you *think* it works?

>the ONLY cap that was backwards was a 2.2uF, really tiny guy
Which cap was that? It may have not blown up when the radio was energized with it installed backwards, but that doesn't mean something else didn't get damaged when it was.

>>573882
I understand that the switch makes contact with 2 at a time, either the top and middle, or the bottom and middle, which turns it on of off. I get that. What I don't get is why there are THREE sets of contacts for the FM, for example, The schematic lists only only one. (picture, (left side)
Bear with me here, I'm trying to learn as fast as I can!

the cap was C60, located in the MPX area (picture, right side). but like you determined earlier, the MPX was bypassed completely, and it still didn't have an effect.

>>573886
>What I don't get is why there are THREE sets of contacts
You don't need to worry about that, just verify that there is continuity through the switch on all poles.

>>573934

Well, somehow its working perfectly again... I asked some folks at audiokarma, and they said check R72 which is variable resistor for the stereo sensitivity. i was getting 115ohms when it shouldve been 350.
and at that point I found one lone trace that was lacking solder and I guess that made it work. I feel like an idiot.
Thanks for all your help man, for real