reddish75 said:
The steps I would take are:
1. Ensure the bias voltages at various are correct, most transistors(except protection transistors) should be forward bias, ie vbe = about 0.65v. The bias voltages at different point must be within design parameters. no load connected(*)
2. once ok the output voltge if it is dc ampl. should be less than 50mv. any larger should really be unacceptable, amp, maybe ok but o/p is too high(i/p transistors or jfets not matched). This is measured when the input is close to gnd, not left floating. no load connected(*)
3.apply i/p sinusoidal voltage to full power when o/p is loaded with 8ohm load.this ensures the o/p transistors can handle full power.optional test with 4ohm load if your speaker is 4 ohm.
hope this helps. cheers.
* amp must be unconditionally stable with no load connected.
just a thought. If you have two identical circuits you can connect red lead of volt meter to circuit 1 and black lead to circuit 2, and ideally it should be zero volts if its working. If it is not working right you will know. Of course you have to probe both sides in the exact same place or part. You do not reference to ground when doing this. Its just a general quick check. I am talking about an amp with left and right channel. This info is just an idea to help in troubleshooting.
when it starts crackling, begin spraying semiconductors (diodes and transistors, and op amps) with freeze mist one at a time (except for differential transistors, spray both of them evenly at the same time). move from input stage to output stage, and wait 30sec between each one. usually such popping and hissing is the input stage, but not always. the reason it's difficult to troubleshoot (not impossible, just difficult) a typical DC coupled amp with a scope is that from the colloctors of the diff amp to the base of the voltage amplifier stage, the signal is in the form of current, not voltage, and the viltages are usually about 0.7 volts less than one of the power supply rails.
I work on PA amps all day long and here is what i do with an amp of unknown working condition.
1- Remove cover(s) and inspect. Check for loose wires, burned parts etc.
2- Bring amp up on a metered Variac (no load) slowly watching current meter. I have a 10 amp variac i use to bring up most amps with. If the amp is drawing an amp or more by 20-30 volts. there is a problem. Some amps will draw current then fade as the caps charge up and some amps with soft start circuits and toroids you need to be careful with as the soft start resistor will smoke. I usually bypass soft start circuits when powering up by variac.
3- If the amp comes up ok. then I check for DC at the output. If less then .5 volts, then i inject a 100hz signal into one channel at a time and use a scope to observe the output.
4- For noise issues like you describe. I run the amp without a load. I may check the amp with a load briefly but while poking around inside i leave the load disconnected. this can help save an amp. depending on the noises nature. with a load connected. poking around and wiggling things, the amp could suddenly draw large currents through the load and smoke parts. without a load, the amp is more likely to survive and not cause further damage to the amp, you or the load! Once you have been able to observe the issue with a scope and resolve it. then i proceed to test with a load.
5- Once both channels have checked out ok, I then connect an 8 ohm load and check each channel with a load to clip. Looking for oscillation, clip summitry, etc.
6- 1/3 power test, I run the amp at a calculated 1/3 power for 15-20 minutes and check for overheating.
7- full power test. run the amp just under clip for 15 minutes. check for overheating/shut down etc.
Of course this is a large overview. along the way i check Bias, offset , DC rail ripple etc. But i pretty much follow these basic steps every time. I have never had a problem running a solid state amp without a load. with the exception of some old old 1970's Sherwood receivers with died if no load is connected.
1- Remove cover(s) and inspect. Check for loose wires, burned parts etc.
2- Bring amp up on a metered Variac (no load) slowly watching current meter. I have a 10 amp variac i use to bring up most amps with. If the amp is drawing an amp or more by 20-30 volts. there is a problem. Some amps will draw current then fade as the caps charge up and some amps with soft start circuits and toroids you need to be careful with as the soft start resistor will smoke. I usually bypass soft start circuits when powering up by variac.
3- If the amp comes up ok. then I check for DC at the output. If less then .5 volts, then i inject a 100hz signal into one channel at a time and use a scope to observe the output.
4- For noise issues like you describe. I run the amp without a load. I may check the amp with a load briefly but while poking around inside i leave the load disconnected. this can help save an amp. depending on the noises nature. with a load connected. poking around and wiggling things, the amp could suddenly draw large currents through the load and smoke parts. without a load, the amp is more likely to survive and not cause further damage to the amp, you or the load! Once you have been able to observe the issue with a scope and resolve it. then i proceed to test with a load.
5- Once both channels have checked out ok, I then connect an 8 ohm load and check each channel with a load to clip. Looking for oscillation, clip summitry, etc.
6- 1/3 power test, I run the amp at a calculated 1/3 power for 15-20 minutes and check for overheating.
7- full power test. run the amp just under clip for 15 minutes. check for overheating/shut down etc.
Of course this is a large overview. along the way i check Bias, offset , DC rail ripple etc. But i pretty much follow these basic steps every time. I have never had a problem running a solid state amp without a load. with the exception of some old old 1970's Sherwood receivers with died if no load is connected.
and with any intermittent noise issue. I use a calibrated plastic toothbrush handle to poke the circuit board. tap devices. wiggle things. 9 out of 10 times its a bad solder joint near a heat producing component, or plug socket etc. But as these PSA-2 are now getting to that age. bad caps is a real possibility. I have a LARGE stack of these amps waiting for me to fix.
Re: it looks
hey thanks for posting. very helpful! how to interpret the current between the stages if it's not .7V from the supply rails to the colloctors of the diff amp and the base of the voltage amplifier stage, as you say?
i will start to freeze spray the semis as you suggested. many thanks.
i wrote the rest of this before you posted..
i've been getting my head around this beast for about a year now.. she certainly is a complicated device and i treat her with great respect and 'caution'. i've already done a lot of work on her, but to no avail RE: 'the crackle'. the sound is excellent, but we continue to get this hiss/pop in ch1 after she's warmed up.
any advice for the not so gifted on how to trace the signal to find the source of this disturbance?
when it starts crackling after an hour of warming up, i'm getting softness/hairiness and spikes on the negative half of the scope when probed at the speaker output.
to be more specific.. the crackle is independent of the signal, not affected by the volume level pots, nor does the behavior respond to light tapping or nudging of components and boards.
all 4 output modules have been rebuilt with new components.. all drivers and outputs are now MJ1402xG and MJ2119xG. all volume level pots and all electrolytic caps replaced, except for the main filters (of which i've swapped channels, with the problem remaining on ch 1). i've swapped the output modules between the channels and the problem is still on channel 1. the problem appears to be isolated to the main module so i'm looking at tracing the signal there.
How do i look for oscillation? and what's a calibrated toothbrush?
cheers!
hey thanks for posting. very helpful! how to interpret the current between the stages if it's not .7V from the supply rails to the colloctors of the diff amp and the base of the voltage amplifier stage, as you say?
i will start to freeze spray the semis as you suggested. many thanks.
i wrote the rest of this before you posted..
i've been getting my head around this beast for about a year now.. she certainly is a complicated device and i treat her with great respect and 'caution'. i've already done a lot of work on her, but to no avail RE: 'the crackle'. the sound is excellent, but we continue to get this hiss/pop in ch1 after she's warmed up.
any advice for the not so gifted on how to trace the signal to find the source of this disturbance?
when it starts crackling after an hour of warming up, i'm getting softness/hairiness and spikes on the negative half of the scope when probed at the speaker output.
to be more specific.. the crackle is independent of the signal, not affected by the volume level pots, nor does the behavior respond to light tapping or nudging of components and boards.
all 4 output modules have been rebuilt with new components.. all drivers and outputs are now MJ1402xG and MJ2119xG. all volume level pots and all electrolytic caps replaced, except for the main filters (of which i've swapped channels, with the problem remaining on ch 1). i've swapped the output modules between the channels and the problem is still on channel 1. the problem appears to be isolated to the main module so i'm looking at tracing the signal there.
How do i look for oscillation? and what's a calibrated toothbrush?
cheers!
"2- Bring amp up on a metered Variac (no load) slowly watching current meter. I have a 10 amp variac i use to bring up most amps with. If the amp is drawing an amp or more by 20-30 volts. there is a problem. Some amps will draw current then fade as the caps charge up and some amps with soft start circuits and toroids you need to be careful with as the soft start resistor will smoke. I usually bypass soft start circuits when powering up by variac. "
actually, i prefer the light bulb method. if you've ever worked on an Acoustic 360, you'll know why. if you haven't, i'll explain:
Acoustic 360 amps work great when they're ok. when the outputs go, everything goes. i once rebuilt a 360 three times (it was for a friend), replacing all of the silicon in the power amp. when bringing the amp up S L O W L Y on a metered variac, everything seemed normal until around 60V, then it was TINK! the ammeter would suddenly slam, the variac fuse would blow, and everything on the amp board was shorted. i thought "maybe there's something i forgot to check" went through, replaced all of the silicon, checked all of the passives, checked the power supply components, etc.... i gave up after the third time (besides, i ran out of output device stock). later, talking to other techs, i find i'm not alone. once a 360 goes up in smoke, leave it alone is the general opinion. later i discover the light bulb method, and it is effective at limiting damage to the first component to short, or to protect everything if, for instance, a bias transistor is open. if the equipment is drawing excessive current, it self-limits the applied voltage, and the amp can be troubleshot "semi-live" without worrying that something else will fry in the process.
actually, i prefer the light bulb method. if you've ever worked on an Acoustic 360, you'll know why. if you haven't, i'll explain:
Acoustic 360 amps work great when they're ok. when the outputs go, everything goes. i once rebuilt a 360 three times (it was for a friend), replacing all of the silicon in the power amp. when bringing the amp up S L O W L Y on a metered variac, everything seemed normal until around 60V, then it was TINK! the ammeter would suddenly slam, the variac fuse would blow, and everything on the amp board was shorted. i thought "maybe there's something i forgot to check" went through, replaced all of the silicon, checked all of the passives, checked the power supply components, etc.... i gave up after the third time (besides, i ran out of output device stock). later, talking to other techs, i find i'm not alone. once a 360 goes up in smoke, leave it alone is the general opinion. later i discover the light bulb method, and it is effective at limiting damage to the first component to short, or to protect everything if, for instance, a bias transistor is open. if the equipment is drawing excessive current, it self-limits the applied voltage, and the amp can be troubleshot "semi-live" without worrying that something else will fry in the process.
another idea for you. If you have the diagram to the beast, signal trace it using another audio amp. say even a small LM386 headphone amp etc. Probe each stage until you hear the crackle. Work from preamp towards the output or the other way. From the preamp toward the output might be the better choice for me. At least you can eliminate the possibilities. I haven't taken a look at the diagram, but the like the other forum member said, the parts that are in any area that gets hot, or draws a substantial amount of current should be checked first, visually and with power off and unplugged check the component values against the other known good side. Yes the solder connections can become fractured and you can usually see them with a good eyepiece. The center pin is not making contact or the base of the solder is not making contact, or the solder will look porous if it is a bad solder job, which will act like a diode in some cases because of the bad solder job.
So I hear anyway. Check the voltage reference points in the diagram and see if any look weird. Start with the signal tracing method, then when you find the bad stage, go for the gusto, and find the problem, using methods already mentioned by other forum members, poking the board near the parts, excessive heat compared to other side etc.
good luck. Do you have the diagram to your unit? If not maybe we can help. It would be much easier to diagnose. I like the ideas given by other members. If one could learn and remember all that is said, it would be a wonderful thing. I would love to know all the secrets of the trade, but found out one could never know it all, only a section of the vastness there is. But even so It is good for ones mind to keep it busy learning all you can. I'm looking fore the schematic so far no luck, but i will keep looking. Found, a service manual pdf download from Crown.
So I hear anyway. Check the voltage reference points in the diagram and see if any look weird. Start with the signal tracing method, then when you find the bad stage, go for the gusto, and find the problem, using methods already mentioned by other forum members, poking the board near the parts, excessive heat compared to other side etc.
good luck. Do you have the diagram to your unit? If not maybe we can help. It would be much easier to diagnose. I like the ideas given by other members. If one could learn and remember all that is said, it would be a wonderful thing. I would love to know all the secrets of the trade, but found out one could never know it all, only a section of the vastness there is. But even so It is good for ones mind to keep it busy learning all you can. I'm looking fore the schematic so far no luck, but i will keep looking. Found, a service manual pdf download from Crown.
yes, i agree
LM386 HeadPhone low watt low power amplifier Chip Amp
LM386 is VERY easy to use in many applications like for amplifier testing
It is not ultimate amplifier quality, but good enough for most any audio.
And not too much distortion in one LM386.
Datasheet:
http://www.national.com/mpf/LM/LM386.html
http://www.national.com/ds.cgi/LM/LM386.pdf
LM386 HeadPhone low watt low power amplifier Chip Amp
LM386 is VERY easy to use in many applications like for amplifier testing
It is not ultimate amplifier quality, but good enough for most any audio.
And not too much distortion in one LM386.
Datasheet:
http://www.national.com/mpf/LM/LM386.html
http://www.national.com/ds.cgi/LM/LM386.pdf
Did you say you swapped power supplies as well? I was not certain you said that for sure. my apologies for not reading you replies in depth. One thing you said you did not do was swap the output speaker network. The inductor toroid could be shorting. or trying to.
In the meantime I will look over the schematic for ideas. I am trying to upload it for all to see as well.
In the meantime I will look over the schematic for ideas. I am trying to upload it for all to see as well.
I put together a download file for all. You could go to crown and download each file individually as I have done, but I saved the time and put them all together as one download for ease.here is the link to download the psa2 pdf's ,all 10 are packed using winrar to cut down on space. just use winrar to unpack. user manual,repair manual and schematics. Read number 6 of 7 page 14 where it tells about how each stage works in detail. How to test and adjust. You will just have to take your time and do it. Its not going to be a quickie, but when youre done, your done.now we will be able to help you better as we can follow along. oh the link sorry.
http://www.4shared.com/file/62820220/a72bfd3a/crown-ps2.html
http://www.4shared.com/file/62820220/a72bfd3a/crown-ps2.html
Hey all.. thanks for all the advice. Gives me a great starting point for moving forward with this and I feel that my beloved PSA-2 and I are in very good hands.
I must apologize for not mentioning that I do have the Service Manual and schematics printed out, and I've read the theory of operation section and 6 of 7 page 14, which describes the signal path and feedback network. Many thanks to yourownfree for posting the goodies for anyone following along.
To clarify: I've swapped the main filter caps between the channels. And I've replaced the caps and resistor on the Output Terminal Module (A-1). I haven't swapped the inductor toroids as yet (would that be L100/L200?). I'm looking at the schematics pdf page 2 which has the Main Module for my particular vintage along with the Output Modules and Output Terminal Module. With my limited understanding of signal paths, etc, would I be able to eliminate the toroid if I end up finding the crackle on the main module? As you can probably tell, I really need my hand held through this because I'm learning as we go..
I can't get my hands on freeze spray just yet. I think I'll start checking voltage reference points now, but I love the suggestion of using another amp to trace the signal and listen for the illusive crackle. If I don't have or can't get a LM386, could I use another amp, eg my trusty D-150A Series II or a Yamaha receiver? If I'm checking points on the main module before the output stage, the signal will be OK to input into one of those amps?
And what about the suggestion further up: "If you have two identical circuits you can connect red lead of volt meter to circuit 1 and black lead to circuit 2, and ideally it should be zero volts if its working"
Does that apply for the problem I'm troubleshooting? Would this mean probing identical points between ch1 and ch2 on the Main Module?
Thanks again to everyone for the terrific responses and suggestions. I'm very determined here and I really enjoy learning about this stuff. Will report back soon.
jake.
EDIT: just paraphrased a bit of my Australian English
I must apologize for not mentioning that I do have the Service Manual and schematics printed out, and I've read the theory of operation section and 6 of 7 page 14, which describes the signal path and feedback network. Many thanks to yourownfree for posting the goodies for anyone following along.
To clarify: I've swapped the main filter caps between the channels. And I've replaced the caps and resistor on the Output Terminal Module (A-1). I haven't swapped the inductor toroids as yet (would that be L100/L200?). I'm looking at the schematics pdf page 2 which has the Main Module for my particular vintage along with the Output Modules and Output Terminal Module. With my limited understanding of signal paths, etc, would I be able to eliminate the toroid if I end up finding the crackle on the main module? As you can probably tell, I really need my hand held through this because I'm learning as we go..
I can't get my hands on freeze spray just yet. I think I'll start checking voltage reference points now, but I love the suggestion of using another amp to trace the signal and listen for the illusive crackle. If I don't have or can't get a LM386, could I use another amp, eg my trusty D-150A Series II or a Yamaha receiver? If I'm checking points on the main module before the output stage, the signal will be OK to input into one of those amps?
And what about the suggestion further up: "If you have two identical circuits you can connect red lead of volt meter to circuit 1 and black lead to circuit 2, and ideally it should be zero volts if its working"
Does that apply for the problem I'm troubleshooting? Would this mean probing identical points between ch1 and ch2 on the Main Module?
Thanks again to everyone for the terrific responses and suggestions. I'm very determined here and I really enjoy learning about this stuff. Will report back soon.
jake.
EDIT: just paraphrased a bit of my Australian English
I think the most important thing you have said is that the noise appears on the scope on the negative half of the waveform. this gives you a head start on which side of the amp to start looking.
The old crowns u have worked on in the past have had some large high watt resistors that either fail, or get so hot that the solder joint is compromised. I would remove, clean, test and replace any large resistors on the boards.
The old crowns u have worked on in the past have had some large high watt resistors that either fail, or get so hot that the solder joint is compromised. I would remove, clean, test and replace any large resistors on the boards.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.