Author Topic: Anyone willing to hold my hand to get me back on my feet with a car amp repair?  (Read 3423 times)

0 Members and 1 Guest are viewing this topic.

Offline OldRad

  • Contributor
  • Posts: 20
  • Country: us
Quote
That may be a bit too low.
I thought that might be a possibility also.  So i tried at 12V, 13.4V and 14.2V  same at any of the voltages.
pics attached
and apparently my memory sucks it was a 51khz wavelength.  :)
The other rectifier is 543 forward voltage. so pretty close i think.
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 4915
  • Country: 00

If your supply can limit current,  it is probably safe to crank up the voltage to the nominal level. 
 

Offline OldRad

  • Contributor
  • Posts: 20
  • Country: us
Few more tests done, and im done for the night.  My brain hurts.
So.... I put the rectifiers back in, it still goes into protect.  But now it stays solid in protect and does not loop.
Q135 and Q145 get hot in about 20 seconds. 
Q235 and Q245 stay cool.
Coming off Pin's 9 and 10 There seems to be a dip in the negative portion of the square wave. Almost like something is trying to pull it to ground. Pic attached.
Now on the FET's, the same dip also exists, but it's cleaned up slightly. (i can only attach 2, and this one is close enough to the pins 9 and 10 shot)
Now on the middle pin of the rectifiers, it's a clean square wave.  Pics attached.  Color me confused.
-OR
 

Online oPossum

  • Super Contributor
  • ***
  • Posts: 1177
  • Country: us
  • The other white meat.
Test Q140, R137, R147, R150
 

Offline OldRad

  • Contributor
  • Posts: 20
  • Country: us
Thanks opossum.  I'll do that Wednesday night.
 

Offline OldRad

  • Contributor
  • Posts: 20
  • Country: us
Ok, those resistors are within their 5% tolerances.
I took the transistor out and put it in the Mk-328.  I think it's good, but I don't know what i'm looking at here.
It does show it as an NPN
B=293(not sure what this is) VF=618mV (forward voltage?)
ICE0= .00ma (no clue)
ICEs= .00ma (no clue)

I spent a good deal of today during my downtime trying to study this schematic.  The math is not working for me.
Im sure it's right and im just dumb and dont get it. 
Working from the top of page 22 with Vee2, I can work out the -0.63V reference voltage around R133, 143, 134, 144.  But the +.58 with Vcc2...  That one has me scratching my head.

I really do hate  how much of my skills i have lost.
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 10290
  • Country: my
  • reassessing directives...
But the +.58 with Vcc2...  That one has me scratching my head.
Vcc and Vcc2 should be +33V, Vee and Vee2 should be -33V. you should also can probe Vac waveform on secondary side of the transformer if mosfets and rectifiers working properly. can you confirm that? if Vee2 and Vcc2 not read properly, you should check the diodes rectifier D51-54 (purple colored in my earlier post), those provide power/voltage to the rails. ymmv.
It's extremely difficult to start life.. one features of nature.. physical laws are mathematical theory of great beauty... You may wonder Why? our knowledge shows that nature is so constructed. We simply have to accept it. One could describe the situation by saying that... (Paul Dirac)
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 4915
  • Country: 00
I really do hate  how much of my skills i have lost.

You did seem to pick one b*tch of a problem to refresh them on!   :)
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 10290
  • Country: my
  • reassessing directives...
still better than trying to repair a spectrum analyzer or an oscilloscope.
It's extremely difficult to start life.. one features of nature.. physical laws are mathematical theory of great beauty... You may wonder Why? our knowledge shows that nature is so constructed. We simply have to accept it. One could describe the situation by saying that... (Paul Dirac)
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 4915
  • Country: 00

It is the "obviousness" of the fault that is the issue, more than the type of equipment - after all, we can probably all repair a scope with a blown fuse, on a good day!   :D

The worst problems are the ones where seemingly simple equipment acts up in some inspired way to lead you down the wrong path when troubleshooting.  Especially if we assume that because the device is simple, the problem must also be simple!   :-[

 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 10290
  • Country: my
  • reassessing directives...
It is the "obviousness" of the fault that is the issue, more than the type of equipment - after all, we can probably all repair a scope with a blown fuse, on a good day!   :D
replacing blown fuse in audio amp is a matter of pulling and replacing it from behind, no dismantling needed. i have Advantest SA here that has intermittent fault. when i opened it, it has 5 or 6 pcbs stacked on top of each other and on the side with countless rigid coax cable connections, countless number of power rails and each pcb is more than 2 layers stack technology probaly 4 or 6 layers (how to trace connection?) 3-10X smd components count and complexity, several mcus rams etc. i just dont know where to start, so i put back the cover in sadness, being afraid of making more damage than good. i need to learn SA diagram, there are lot of sections one must be familiar with every rf components, switches, attenuators, amplfiers etc what they do, i say... super difficult on grade 10. i guess will need hundreds or thousands of pages if someone want to hand guide me in a forum.

otoh with audio amp, its 2 layers pcb tech, through holes, super minimal smd count. i managed to create schematics of its major section and understand it in few hours or so less than a day from nothing. i'm not quite interested in audio filter and conditioning section they are seldomly got damaged, so i didnt make its schematics, although i have general picture of what they are, they are only few TH resistors, caps, resistors and caps again and small bjts thats it. audio amp is nothing but (1) step up smps providing higher power rail (usually one +ve and one -ve rail only) (2) audio filters and conditioning at input (seldomly damaged) (3) the audio power amplifier (those big bjts). its quite easy to understand. no, audio amp and a SA are not the same, they like earth and sky difference in complexity, unobviousness-wise and skill/knowledge level required. ymmv.
« Last Edit: April 17, 2019, 05:08:16 pm by Mechatrommer »
It's extremely difficult to start life.. one features of nature.. physical laws are mathematical theory of great beauty... You may wonder Why? our knowledge shows that nature is so constructed. We simply have to accept it. One could describe the situation by saying that... (Paul Dirac)
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 4915
  • Country: 00

Point taken, both are true to a degree:   you can have a simple problem with a complex system (did you plug it in?  :D ) and you can have a non-trivial problem with a simple system (for example, instability of an amp caused by an unfortunate edge case of drifting component, etc.).

The nightmare scenario, of course, is multiple non-trivial problems with a very complex system! -  requires several pots of coffee, with extra sugar...
 

Offline OldRad

  • Contributor
  • Posts: 20
  • Country: us
Vcc and Vcc2 should be +33V, Vee and Vee2 should be -33V. you should also can probe Vac waveform on secondary side of the transformer if mosfets and rectifiers working properly. can you confirm that? if Vee2 and Vcc2 not read properly, you should check the diodes rectifier D51-54 (purple colored in my earlier post), those provide power/voltage to the rails. ymmv.
I'm just trying to do the math on paper to see how they got those to voltages at the test points.  Like i said i can get the - voltage one.  but the + voltage one, I get the same outcome because the components looks the same. 
But you have a good suggestion and i will probe the secondary side tonight and probably rule 1 more thing out.   |O

Quote
You did seem to pick one b*tch of a problem to refresh them on!
For someone who hasn't forgotten as much as i have, it would probably be fixed already.

 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 4915
  • Country: 00
A little while back, I ended up spending most of a day on a flickering desk lamp.  I mean, what could possibly go wrong with a desk lamp?  Changed the tubes...  changed the plug...  changed the switch...  swapped the ballast...  still flickering. 

Turned out to be a wire that was mostly broken internally, making randomly intermittent contact...  while looking perfect from the outside!
 

Offline OldRad

  • Contributor
  • Posts: 20
  • Country: us
Ok  More testing tonight.  +33V and -33V square wave at the secondaries. It was 26Khz. Is this halved because of the rectification?
+33V and -33V DC at the output transistors.
The amp stayed on (but in protect) for about 20 minutes tonight.  The output transistors that got hot the other night stayed cool tonight.  :-//
While probing around, i noticed another part of the board getting warm.  So i flipped it over and started feeling for what was hot.  Best i could tell it was Q243.  But that doesn't even drive the side that got hot the other night.  Grrr.  Anyway, popped it out put it in the mk-328, it read it as an npn.  So do i look at what is driving that? Q241 and Q242????
being frustrated, i started going down the list of resistors and testing those. At least i know for sure what i am supposed to read with those. I got about 100 or so tested tonight.  3 were out of 5% tolerance but within 10%. Maybe something in the circuit throwing the reading off? 
At least i will know this amp inside and out when i'm done.
 

Offline OldRad

  • Contributor
  • Posts: 20
  • Country: us
Quote
So.... I put the rectifiers back in, it still goes into protect.  But now it stays solid in protect and does not loop.
Haha  So i think I'm a dumb***.    :-DD  Due to the space where i was working on this amp, i didn't have enough space to keep the case of the amp near where i was working.   This morning in my awakening stupor, i was downstairs and decided to look at the case again.  Turns out that LED i THOUGHT was the protect LED.... was actually power.   :palm: I was thinking it was the power LED lighting and then going into protect.  Turns out it was protect going to power.  I assume in protect long it enough to build rail voltages and then good.  I wont be able to get it hooked up to a sound source until tonight, but fingers crossed.   
Note to all you young whippersnappers... don't get old. :)
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 4915
  • Country: 00
Sounds like a Type 1 error - null hypothesis rejected for the wrong reason!

My accountant is 75 and sharp as a razor.  He refuses to retire...   says he needs to do numbers to keep the brain circuits working.
 

Offline OldRad

  • Contributor
  • Posts: 20
  • Country: us
Welp, better but not perfect.  It is processing audio out of both channels. However the left channel Q234 and Q245 are getting really hot.  while the right channel is staying cold. The power supply was showing  a 1 amp draw.
Am i correct in thinking that this will  probably be in the chain of transistors leading up to Q235 and Q245?

I also want personally thank Mech and Opossum for getting me this far.  It is very much appreciated. I also appreciate the comments and thought processes from the others who have commented.

Now on to figuring out why these are getting so hot.
 

Online oPossum

  • Super Contributor
  • ***
  • Posts: 1177
  • Country: us
  • The other white meat.
Check the voltage between the bases of Q235 & Q245 and Q135 & Q145. Should be under 1 volt.

 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 4915
  • Country: 00
 
The following users thanked this post: Mechatrommer, OldRad

Offline OldRad

  • Contributor
  • Posts: 20
  • Country: us
Base voltages on all are around .3.  Collectors are 32 +/-  emitters are .05
 

Offline OldRad

  • Contributor
  • Posts: 20
  • Country: us

This may be relevant:

That actually helped a lot. I was not really understanding what was going on in the schematic, nor in the video, until he flipped up the piece of paper to show the other half.  I literally said aha outloud. i love it when the lightbulb above my head turns on.
Thanks for that!
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 10290
  • Country: my
  • reassessing directives...
However the left channel Q234 and Q245 are getting really hot.  while the right channel is staying cold. The power supply was showing  a 1 amp draw.
Am i correct in thinking that this will  probably be in the chain of transistors leading up to Q235 and Q245?
i guess some transistors on earlier stage are off the spec. you may check what opposum suggested and starting from components closest to the power bjt. but if its still unobvious, since you have right channel working flawlessly you may want to feed both channel with the same wave input and check each node on both channel for comparison using oscilloscope, this is where a function generator may help (be warn, only feed with mVpp like 1mVpp or 100mVpp of sine signal because this is how normal audio signal is feed, not Vpp range like 5Vpp, you'll distort output sound badly). if you detect difference in signal shape, that maybe suspect. remove and check with mk328 transistor checker for parameter values. 1A is normal (managable), dont forget not to short bjt/mosfet body to the aluminium enclosure, ensure their silicon (grey) pad are installed. that is some thought i can give and think.

the more you go deeper, the more you need understanding the working of the circuit, and the more you have to work on your own, as different amp have slight different stage/circuit/topology in class AB amplifier. the less precise we can advice, if effort seems not to give any fruit, maybe its time to sit and relax and watch those youtube videos linked earlier. or try recapping about bjt amplifier circuit topology in the net. wish you luck.
« Last Edit: April 19, 2019, 05:44:32 am by Mechatrommer »
It's extremely difficult to start life.. one features of nature.. physical laws are mathematical theory of great beauty... You may wonder Why? our knowledge shows that nature is so constructed. We simply have to accept it. One could describe the situation by saying that... (Paul Dirac)
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 4915
  • Country: 00
You can do worse than binge watch a ton of Alan Wolke (w2aew)'s videos on Youtube,  he is really good at explaining things clearly!

Re. the amp:   If there is an adjustment pot in the amp's bias control (Vbe multiplier),  I would try cleaning it as one of the first steps, as it can cause both intermittent and permanent problems with the bias current.

Pots can go bad by partly or totally losing contact with the track.  If you look at the Vbe multiplier circuit, you can see that the consequence of losing connectivity to the track would be that the Vbe multiplier transistor totally stops conducting... sending ALL the current through the output transistors instead and cranking up the bias current to sky high levels!

« Last Edit: April 19, 2019, 12:39:10 pm by SilverSolder »
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 10290
  • Country: my
  • reassessing directives...
If there is an adjustment pot in the amp's bias control (Vbe multiplier),  I would try cleaning it as one of the first steps, as it can cause both intermittent and permanent problems with the bias current.
its the one suggested by opposum earlier.. no trimpot... fixed bias
Test Q140, R137, R147, R150

btw, probing with oscilloscope may help.. ymmv.
It's extremely difficult to start life.. one features of nature.. physical laws are mathematical theory of great beauty... You may wonder Why? our knowledge shows that nature is so constructed. We simply have to accept it. One could describe the situation by saying that... (Paul Dirac)
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf