Badcaps.net Forum
Go Back   Badcaps Forums > Electronics Theory and Troubleshooting > Audio Equipment Discussion
Register FAQ Calendar Search Today's Posts Mark Forums Read

 
Thread Tools Display Modes
Old 08-11-2017, 05:01 PM   #81
Agent24
I see dead caps
 
Agent24's Avatar
 
Join Date: Oct 2007
City & State: Hiding inside a plated-through hole
My Country: New Zealand
Line Voltage: 230VAC 50Hz
I'm a: Hobbyist Tech
Posts: 4,359
Default Re: Crest CPX1500

Good to see you made some progress with the missing connection to the rectifier. That would explain why Channel B was behaving so badly compared to A!

I am trying to understand the opamp\feedback circuits here, (If someone else could lend a hand, that would be great!!) I may well be getting things wrong at this point!


As far as I understand it, though:

U100-A is set as a buffer for the input signal to the rest of the amp. That's easy enough.

U100-B seems to be set as an inverting amplifier (?) for the input signal, and also taking feedback from the Speaker+ output line (before L100). It seems to drive Q104 and Q105, which in turn drive the output drivers and output stage.

U102-B is used as a comparator, seems like it has some job to do with the GCL function. If its output goes high, then it should make U102-A also produce a high output and turn the DDT LED on, to indicate that the GCL limiter is active.


One half of U101 (Pins 1, 2 and 3) takes input from the buffered output of U100-A and feeds it back to U100-A's input. I am not sure what the purpose of this is, perhaps to lower distortion?

The other half of U101 makes even less sense to me:
The output from U102-B goes through to Pin 5 of U101. This seems to be comparing the output from U102-B's comparison, with the original input signal at Pin 6, out of R102.
If U101's Pin 6 (inverting input), ie: the audio input signal were ever to become higher in voltage than the input to Pin 5, then U101's output on Pin 7 would go LOW, and short the 15V+IA rail to 15V-IA. I am really not sure what purpose this serves, but I am probably misunderstanding the circuit significantly??


I don't know why the amp seems to oscillate (perhaps) and go into short condition now, when you have it powered without U101 and U201, or worse, with them connected.

I am confused though, earlier you said you had signal at the speaker from Channel A, with U101 removed. What has changed between then and now going into a restart loop, except for fixing the -Ve rectifier connection to channel B?

What if you run with Channel B disconnected again? Does Channel A work again without tripping the lightbulb?
__________________
"Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
-David VanHorn
Agent24 is offline   Reply With Quote
Old 08-12-2017, 05:29 AM   #82
5inc
Senior Member
 
Join Date: Dec 2016
City & State: Barcelona
My Country: Spain
I'm a: Knowledge Seeker
Posts: 120
Default Re: Crest CPX1500

ok that's a chunk of info I'll have to digest.

What has changed from the point I had signal at output A and a red ddt light?
I had tried replacing U301 which hadn't been changed yet, and believed it made no difference.
At that point is when I started checking every voltage I could compare and ended up seeing there was no -80v rail. Re-connected the bridge rectifier and begun getting these peaks.

All lights are off, and every 3 seconds channelA green+channelB's red/yellow/green lights blink. At the time those go off both red lights blink again. Lightbulb also blinks.

Signal on output A isn't affected by peaks. Output B gets some nasty thumps.


Disconnecting channel B makes the amp start without peaks (appart from the fan going a bit crazy at given moments), so the main problem must still reside there.

And about what you said, as soon as U101 is fitted in place it makes the amp go into short, so the voltage at pin6 must be higher than on pin 5.

But when i check voltages on u101 on the board with the IC out i get 15v on pin5 and 0v on pin6. Also when I probe pin6 i get a notable ground hum.
The same happens on U201 (i didn't get the hum because it belongs to the other channel)
5inc is offline   Reply With Quote
Old 08-16-2017, 07:54 AM   #83
5inc
Senior Member
 
Join Date: Dec 2016
City & State: Barcelona
My Country: Spain
I'm a: Knowledge Seeker
Posts: 120
Default Re: Crest CPX1500

After shorting the amp because of a probe-slip and restoring it to it's blinky restart loop point, I thougt I found something.

Disconnecting v+/0v/v- going to channel B makes the amp stay still, and from there i'm seeing there is no DC coming out of bridge rectifier D200 with black probe on chassy but since v+/0v/-v leads are disconnected this must be normal(AC is going in, allthough 53v on one side, 62v on the other). If I measure across the bridge rectifier itself for voltage I get +53V/0v/-53v DC.

D100 carries +82v/0v/-82v DC on it's output, and it's input gets 55v/0/55v AC, with it's channel loaded on it.

I was thinking D200 was bad, so I swapped both and still get the same results... should they both be ok... or both malfunctioning?

Appart from that, I still can not find the damn short that is making the amp trip, I even tried killing power input to u200 and u202 by taking r294,r295, r298 and r299 out... AND I disconnected one leg of R113, it still blinks.

if I also disconnect R208,R212, R117, R218 and R219, that will isolate the IC half of the board from the transistor part, but do you recomend trying that, or do you think something could go really wrong if I do so?

I ask because once i took cr128 out and that made the fuses go...
5inc is offline   Reply With Quote
Old 08-28-2017, 05:14 AM   #84
5inc
Senior Member
 
Join Date: Dec 2016
City & State: Barcelona
My Country: Spain
I'm a: Knowledge Seeker
Posts: 120
Default Re: Crest CPX1500

After a while going nuts about what was going on, yesterday, I gave the channel B board a good clean up with alcohol, and what do you think? it stopped blinking.

Channel B red DDT led still remains on, and checking voltages I found -25v between R234,CR214, and the point where it joins Q203, CR221 and C216. That includes the "-15IB" line that feeds U200, follows through "-15PB" and around R214, R215 it returns to -16v again.

I don't understand how the voltage is higher (or lower?) only in that area, if after all the components on the line the voltages concur with the channel beside. If there is no -28v line, what component could build up such voltage?

One by one I took out Q203, CR224, CR214, R235(this one made the amp run into short), and replaced CR202,CR221,CR222. There was allways -25v there. Any ideas what else I could take out to isolate that leak?

I also have -15v on pin1 of U102 that runs to red DDT led of channel A, which isn't lighting up. Unless I toggle the GCL enable defeat switch, then the led lights and voltage rises to +1.9v.

The upper side of the situation is, I now have both output A and B signal. They both ignore startup delay time.

Any ideas what I could look for, or should I start fitting components back on?
5inc is offline   Reply With Quote
Old 09-02-2017, 08:13 PM   #85
Agent24
I see dead caps
 
Agent24's Avatar
 
Join Date: Oct 2007
City & State: Hiding inside a plated-through hole
My Country: New Zealand
Line Voltage: 230VAC 50Hz
I'm a: Hobbyist Tech
Posts: 4,359
Default Re: Crest CPX1500

Quote:
Originally Posted by 5inc View Post
D100 carries +82v/0v/-82v DC on it's output, and it's input gets 55v/0/55v AC, with it's channel loaded on it.

I was thinking D200 was bad, so I swapped both and still get the same results... should they both be ok... or both malfunctioning?
Both are probably OK. If you get +/-82v out of both bridges themselves. If they were shorted the primary fuse would keep blowing.

Quote:
Originally Posted by 5inc View Post
Appart from that, I still can not find the damn short that is making the amp trip, I even tried killing power input to u200 and u202 by taking r294,r295, r298 and r299 out... AND I disconnected one leg of R113, it still blinks.
Disconnecting power only is not the best idea because with an input signal but no power supply, some ICs can be damaged.

Can you remember what exactly you were probing and where your probe slipped, that caused the fuse to blow again?

Quote:
Originally Posted by 5inc View Post
if I also disconnect R208,R212, R117, R218 and R219, that will isolate the IC half of the board from the transistor part, but do you recomend trying that, or do you think something could go really wrong if I do so?

I ask because once i took cr128 out and that made the fuses go...
I don't know what exactly will happen.
Agent24 is offline   Reply With Quote
Old 09-02-2017, 08:16 PM   #86
Agent24
I see dead caps
 
Agent24's Avatar
 
Join Date: Oct 2007
City & State: Hiding inside a plated-through hole
My Country: New Zealand
Line Voltage: 230VAC 50Hz
I'm a: Hobbyist Tech
Posts: 4,359
Default Re: Crest CPX1500

Quote:
Originally Posted by 5inc View Post
After a while going nuts about what was going on, yesterday, I gave the channel B board a good clean up with alcohol, and what do you think? it stopped blinking.

Channel B red DDT led still remains on, and checking voltages I found -25v between R234,CR214, and the point where it joins Q203, CR221 and C216. That includes the "-15IB" line that feeds U200, follows through "-15PB" and around R214, R215 it returns to -16v again.
What are you measuring with respect to, ie: where is the negative lead of your meter? Ground\Chassis?

Quote:
Originally Posted by 5inc View Post
I don't understand how the voltage is higher (or lower?) only in that area, if after all the components on the line the voltages concur with the channel beside. If there is no -28v line, what component could build up such voltage?

One by one I took out Q203, CR224, CR214, R235(this one made the amp run into short), and replaced CR202,CR221,CR222. There was allways -25v there. Any ideas what else I could take out to isolate that leak?
Taking out R235 will mess up what Q205 (and probably Q204 as well) are doing, and these need to be working otherwise the output transistors won't be driven\biased etc properly.

Quote:
Originally Posted by 5inc View Post
I also have -15v on pin1 of U102 that runs to red DDT led of channel A, which isn't lighting up. Unless I toggle the GCL enable defeat switch, then the led lights and voltage rises to +1.9v.
I believe that in normal operation the LED should be OFF anyway. If the LED comes on during operation, it means that the amplifier is limiting due to clipping. So if you can enable GCL and it will turn the LED on, even without an input signal, then something is still faulty.

Quote:
Originally Posted by 5inc View Post
The upper side of the situation is, I now have both output A and B signal. They both ignore startup delay time.
Are you talking about output at the speaker terminals?

Quote:
Originally Posted by 5inc View Post
Any ideas what I could look for, or should I start fitting components back on?
Not really. I'm still a bit confused as to what the current state of the amplifier is, and whether or not the main amplifier section and speaker output is working in both channels or not.
Agent24 is offline   Reply With Quote
Old 09-03-2017, 04:37 AM   #87
5inc
Senior Member
 
Join Date: Dec 2016
City & State: Barcelona
My Country: Spain
I'm a: Knowledge Seeker
Posts: 120
Default Re: Crest CPX1500

Quote:
Originally Posted by Agent24 View Post
Both are probably OK. If you get +/-82v out of both bridges themselves. If they were shorted the primary fuse would keep blowing.
So, bridges are ok then, thanks.


Quote:
Originally Posted by Agent24 View Post
Disconnecting power only is not the best idea because with an input signal but no power supply, some ICs can be damaged.
I don't know if I succesfully disconnected power only, but at some point I did take out R295, so that may have damaged U200, and also I did something similar on u202 so those are next on list.


Quote:
Originally Posted by Agent24 View Post
Can you remember what exactly you were probing and where your probe slipped, that caused the fuse to blow again?
I can't remember exactly where the probe slipped, but it made CR224 short out and blew the fuses.


Quote:
Originally Posted by Agent24 View Post
What are you measuring with respect to, ie: where is the negative lead of your meter? Ground\Chassis?
I have done all voltage checks with negative lead on chassy, except for when I was checking bridge rectifier with channel B disconnected, I measured it across then.


Quote:
Originally Posted by Agent24 View Post
Taking out R235 will mess up what Q205 (and probably Q204 as well) are doing, and these need to be working otherwise the output transistors won't be driven\biased etc properly.
Note taken: leave R235 alone, it's what keeps Q205 stable. I still wonder what the -25v is doing there though. From what you said and what I saw, it might be coming from pin7 of u202, but I'm reading -14v there. I'll make a little drawing on where I read -25v.


Quote:
Originally Posted by Agent24 View Post
I believe that in normal operation the LED should be OFF anyway. If the LED comes on during operation, it means that the amplifier is limiting due to clipping. So if you can enable GCL and it will turn the LED on, even without an input signal, then something is still faulty.
Yes, in normal operation red DDT leds should be off. Channel A's comes on when I enable CGL, channel B's allways remains on.


Quote:
Originally Posted by Agent24 View Post
Are you talking about output at the speaker terminals?
Yes, when I say I have output I mean there is signal at the speaker terminals, and it sounds pretty good (I hooked it up to a speaker). U101 and U201 are DISCONNECTED



Quote:
Originally Posted by Agent24 View Post
Not really. I'm still a bit confused as to what the current state of the amplifier is, and whether or not the main amplifier section and speaker output is working in both channels or not.
Of course, I'm also confused as to what the current state of the amp is.

It is obvious there is one fault somewhere in channel B that builds up an excessive negative voltage on the negative power input of U200.

There is a second fault that disables U101. When placed, speaker terminal only sends hiss.

There is a third fault (or maybe the first) that makes U201 trip. This last one, could have to do with the fact pins 4 and 5 are recieving this -25v. What happens when placed is the amp jumps into a restart loop.


I have been comparing voltages, but since neither channel is working fully, comparing ain't 100% reliable.

And I can't find any voltage references in the shematic appart from the +-15v,+28v and +-v from bridges so I can't tell what is good or bad.

I hope that helps, thanks
5inc is offline   Reply With Quote
Old 09-16-2017, 11:04 PM   #88
Agent24
I see dead caps
 
Agent24's Avatar
 
Join Date: Oct 2007
City & State: Hiding inside a plated-through hole
My Country: New Zealand
Line Voltage: 230VAC 50Hz
I'm a: Hobbyist Tech
Posts: 4,359
Default Re: Crest CPX1500

You need to figure out why you are reading -25v at the -Ve rail to U200, because it can only handle up to +/-18v and may be damaged again now.

Can you draw on the schematic and upload a picture where you find this -25v? I am a bit confused by your description.

I have no idea why there would show -25v at 15V-PB. The voltage should be clamped to -15v by CR224 anyway. What voltage do you get if you measure across CR224?

You don't get -25v at 15V-PA in channel A, do you?

Is U401 still OK? It's not shorted from input to output, is it?
Agent24 is offline   Reply With Quote
Old 09-18-2017, 08:32 AM   #89
5inc
Senior Member
 
Join Date: Dec 2016
City & State: Barcelona
My Country: Spain
I'm a: Knowledge Seeker
Posts: 120
Default Re: Crest CPX1500

Here you go, I made a quick coloured drawing over the schematic stating where voltage is most critical. If you want me to go further, just ask.

I DON'T get -25 on channel A thank god, and I can't read short on U401 in any direction...
Attached Images
File Type: png voltages crest-25v.png (199.9 KB, 5 views)
5inc is offline   Reply With Quote
Old 09-18-2017, 01:24 PM   #90
Agent24
I see dead caps
 
Agent24's Avatar
 
Join Date: Oct 2007
City & State: Hiding inside a plated-through hole
My Country: New Zealand
Line Voltage: 230VAC 50Hz
I'm a: Hobbyist Tech
Posts: 4,359
Default Re: Crest CPX1500

Is CR224 open circuit?
Agent24 is offline   Reply With Quote
Old 09-20-2017, 10:56 AM   #91
5inc
Senior Member
 
Join Date: Dec 2016
City & State: Barcelona
My Country: Spain
I'm a: Knowledge Seeker
Posts: 120
Default Re: Crest CPX1500

Why, yes. It is open circuit!

I replaced it, and that solved the first and third fault yay!

But the second fault still remains, something is still making U101 stop the signal through. The same happens with U201.

Signal tracing time?
5inc is offline   Reply With Quote
Old 09-20-2017, 12:36 PM   #92
5inc
Senior Member
 
Join Date: Dec 2016
City & State: Barcelona
My Country: Spain
I'm a: Knowledge Seeker
Posts: 120
Default Re: Crest CPX1500

I first changed U200 in case the -25v killed it, but no luck. I haven't changed U101/201 because I'm down to the last pair, will need to order more if not, but they don't read short.

I found something funny, while I probe r203/204 with the signal tracer the amp jumps into short..

I tried on the other channel but the layout is different, there aint no r103/104.

Will keep on looking.

Last edited by 5inc; 09-20-2017 at 12:50 PM..
5inc is offline   Reply With Quote
Old 09-20-2017, 01:45 PM   #93
5inc
Senior Member
 
Join Date: Dec 2016
City & State: Barcelona
My Country: Spain
I'm a: Knowledge Seeker
Posts: 120
Default Re: Crest CPX1500

I picked up clean signal on pins 2,3 of U201, and a nasty one on pin 7. but no short through inputs or output.

I tried to see what happened if I took out U201 and what happened was Q203 blew up.

I am now stuck in the debate of wether using or not the last pair of CA3080 i have. Assuming U201 died with that -25v, U101 could still remain unharmed. what do you think?
5inc is offline   Reply With Quote
Old 09-20-2017, 04:18 PM   #94
Agent24
I see dead caps
 
Agent24's Avatar
 
Join Date: Oct 2007
City & State: Hiding inside a plated-through hole
My Country: New Zealand
Line Voltage: 230VAC 50Hz
I'm a: Hobbyist Tech
Posts: 4,359
Default Re: Crest CPX1500

The thing is it may not be that U201 or U101 are at fault, Perhaps there is something wrong with whatever input they are receiving.

The problem is that I don't know exactly what you should expect to find, or how the circuit works in detail.


R203 and R204 seem to be part of the circuit to do with switching between stereo and bridge operation. It makes sense they will only be present in one channel, so don't worry about not being able to find an R103 or R104 in the other channel.

When you probe something which causes a short condition, I think your signal tracer is probably upsetting the operation of the amp and causing some of the power transistors to saturate (turn full ON). This may just be due to the very sensitive opamp inputs you are probing around with.
What are you using for a signal tracer? Do you have a DC blocking capacitor on the input of it?


Pin 7 of U201 is connected to the 15v rail, so it makes sense you are not going to pick up a nice audio signal there. Although I still do not understand why on earth they have connected an output directly to a power rail.


I have no idea how taking U201 out caused Q203 to blow up, but again, it may have caused some instability. Have you checked all the surrounding diodes etc that connect to Q203? Maybe one of them was faulty in some way as well.
Agent24 is offline   Reply With Quote
Old 09-20-2017, 04:22 PM   #95
Agent24
I see dead caps
 
Agent24's Avatar
 
Join Date: Oct 2007
City & State: Hiding inside a plated-through hole
My Country: New Zealand
Line Voltage: 230VAC 50Hz
I'm a: Hobbyist Tech
Posts: 4,359
Default Re: Crest CPX1500

Wait. Are U201 and U101 CA3080s?
I thought you said they were NJM4560?

Those two ICs are very very different.
Agent24 is offline   Reply With Quote
Old 09-20-2017, 05:12 PM   #96
Agent24
I see dead caps
 
Agent24's Avatar
 
Join Date: Oct 2007
City & State: Hiding inside a plated-through hole
My Country: New Zealand
Line Voltage: 230VAC 50Hz
I'm a: Hobbyist Tech
Posts: 4,359
Default Re: Crest CPX1500

If U201/101 are CA3080s, the circuit makes a lot more sense. Pin 7 is then the +Ve power supply and Pin 6 is the output.

Pin 5 is a bias adjust (?) and I think it's being used in conjunction with the output from U102-B to disable or enable the operation of U101, which seems connected in a negative feedback arrangement to reduce input to U100-A when the GCL switch is enabled and there is clipping detected.


Now, with the GCL switch set to defeat, Pin 5 of U101 should be pulled toward the negative rail and I believe U101 will not do anything to the input signal.

However, if the GCL switch is set to enabled, then Pin 5 voltage becomes dependent on whatever U102 is doing. If U102-B senses clipping, then it drives its output high which will cause Pin 5 of U101 to rise, and U101 will begin reducing the level of the input signal.

A high output from U102-B will also feed into and cause U102-A to produce a high output and light the DDT LED.


So, what voltage do you get at U101 (and U202) Pin 5? If it is always high, then it could be limiting the input signal even though there is no reason to, and so you hear nothing from the speaker.


Quote:
Originally Posted by 5inc View Post
and about U101... I'm a bit confused with this one. I've been seeing different things that seem weird, but I'm not sure. Appart from the fact that I don't know nothing about feedback. And appart from the fact DDT led finally goes off, I have no other constance that that is ok (so I'm not certain yet).
Then, between pins 3-6 I have a 40 ohm short. It took me 2 pieces to realize the short wasn't on the IC but on the track.
Between pins 2-3 I have 100ohm (as expected), but pins 1-2 show 140ohms. Also on the track.
The ICs out of circuit read fine.
I also made one test with U101 out of circuit and signal arrived to the output! But as soon as I put it back on I lost all signal again to return to a constant pinkish noise.
I am interested about these low ohm readings you measured earlier. Have they changed after cleaning the board? Have you cleaned that part of the board yet?
Agent24 is offline   Reply With Quote
Old 09-25-2017, 04:04 AM   #97
5inc
Senior Member
 
Join Date: Dec 2016
City & State: Barcelona
My Country: Spain
I'm a: Knowledge Seeker
Posts: 120
Default Re: Crest CPX1500

This is the tracer I built. Blocking cap included.
http://musicfromouterspace.com/analo...naltracer.html

Pin5 has voltage, something around -15v both on U101/201(it is where the -25v was), also with CGL enabled and in defeat.

So as you said it could be limiting the input signal even though there is no reason to. Do you have any suggestions?

I just measured again the resistivity between pins of U101/201 and they are still the same
U101
1-2: 141ohm
2-3: 97ohm
3-6: 43ohm
U201
1-2: 160ohm
2-3: 97ohm
3-6: 70ohm
5inc is offline   Reply With Quote
Old 09-26-2017, 02:10 PM   #98
Agent24
I see dead caps
 
Agent24's Avatar
 
Join Date: Oct 2007
City & State: Hiding inside a plated-through hole
My Country: New Zealand
Line Voltage: 230VAC 50Hz
I'm a: Hobbyist Tech
Posts: 4,359
Default Re: Crest CPX1500

Signal tracer looks OK.


I've looked again and the resistances seem OK actually.

Between Pins 1 and 2 there is a path to ground from pin 1 through Q101 and to ground also from pin 2 through R107 (100 Ohm). Add the resistance of Q101 and R107, you get the reading you measured.

From Pin 2-3 is R107, so you get around 100 Ohms there.

From Pin 3-6 you are measuring the resistance of Q101 which is basically in parallel with pins 3-6 through the ground.



It is weird that the voltage on Pin 5 stays at -15 volts. So it does not change when the DDT LED is on? Or does that not happen any more?

Can you check and post voltages at R117, R115, R114, CR102, both with the DDT LED on and off?
Agent24 is offline   Reply With Quote
Old 09-30-2017, 06:24 AM   #99
5inc
Senior Member
 
Join Date: Dec 2016
City & State: Barcelona
My Country: Spain
I'm a: Knowledge Seeker
Posts: 120
Default Re: Crest CPX1500

what a wonderfully detailed explanation, thanks! Now I do understand resistance of pins 1-2.

Now, I just checked voltage on R114/115/117 CR102 in defeat (DDT led off), all carry around -15v, then I switched to enabled whitch should turn on the led but it didn't. While probing the different components the led would light up and turn off when the probe was lifted. Voltage was also -15v on all four when the led lit.

I made the same tests on channel B and voltages were similar, but with CGL enabled DDT led wouldn't light when probing, so in this case I couldn't check with DDT led on.

So to answer your questions:
-The voltage doesn't change when the DDT led is on, it is allways around -15v. Or that or my multimeter trips only channel A's led to light up and the measurements I made should all count as with led off.
5inc is offline   Reply With Quote
Old 10-03-2017, 03:01 PM   #100
Agent24
I see dead caps
 
Agent24's Avatar
 
Join Date: Oct 2007
City & State: Hiding inside a plated-through hole
My Country: New Zealand
Line Voltage: 230VAC 50Hz
I'm a: Hobbyist Tech
Posts: 4,359
Default Re: Crest CPX1500

Thanks, I just wish I could actually explain what is wrong with it! But I really am not sure. I don't think I understand the operation of this well enough.

I'm really hoping someone else will be able to give some more advice at this point.

All I can think is that maybe one of the IC was damaged again with that -25v.
Agent24 is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump



Badcaps.net Technical Forums 2003 - 2017
Powered by vBulletin ®
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
All times are GMT -6. The time now is 09:35 PM.

Did you find this forum helpful?