Hi zaib4tsu,
overlooking the brief tension, I want to offer just my 2 cent, as a question to you without offending you.
Did you happen to measure the voltage drop over the resistors in the area that had blown and replaced transistors ?
Perhaps a resistor burned out shortly after a transistor died, yet never got replaced.
Also, as suggested prior, did you inspect the soldering and traces visually ?
It may not hurt, if in doubt, to ring through some traces....
For starters all the basic parameter, such as, Ic, VCEo, Pc or Ptot, fT, Hfe should be close. Don't expect to find a perfect one on one match for a replacement part. Eventually experience leads a professional tech to choose a certain replacement part, knowing what parameter can be traded in on a replacement part and still make it work ok. Science and art become somewhat blurry...
As already stated prior, well matched paired transistors may be imperative for this particular amp to function stable. As said, true and well matching is generally not as easy as to take a single reading of some DMM. A curve tracer is needed for that. Quite a specialized and expensive measurement tool that most amateurs never heard of. However, more important than having the tools, is actually knowing how to use them effectively.
You were stating you have a basic understanding on BJTs. Have you further investigated why on some of your measurements B and E voltages are the same ?
Are you familiar with the different failure modes of BJTs ?
Again, no offense, just trying to lead you to look into things for yourself so you get the AHA effect and learn something a teacher could never tell you.
overlooking the brief tension, I want to offer just my 2 cent, as a question to you without offending you.
Did you happen to measure the voltage drop over the resistors in the area that had blown and replaced transistors ?
Perhaps a resistor burned out shortly after a transistor died, yet never got replaced.
Also, as suggested prior, did you inspect the soldering and traces visually ?
It may not hurt, if in doubt, to ring through some traces....
For starters all the basic parameter, such as, Ic, VCEo, Pc or Ptot, fT, Hfe should be close. Don't expect to find a perfect one on one match for a replacement part. Eventually experience leads a professional tech to choose a certain replacement part, knowing what parameter can be traded in on a replacement part and still make it work ok. Science and art become somewhat blurry...
As already stated prior, well matched paired transistors may be imperative for this particular amp to function stable. As said, true and well matching is generally not as easy as to take a single reading of some DMM. A curve tracer is needed for that. Quite a specialized and expensive measurement tool that most amateurs never heard of. However, more important than having the tools, is actually knowing how to use them effectively.
You were stating you have a basic understanding on BJTs. Have you further investigated why on some of your measurements B and E voltages are the same ?
Are you familiar with the different failure modes of BJTs ?
Again, no offense, just trying to lead you to look into things for yourself so you get the AHA effect and learn something a teacher could never tell you.
Hi,
You need to have equal or better ratings for VCE, Ic, and power dissipation. Gain should be somewhat close and is normally given as a range of values. fT should be in the same ballpark and will sometimes work even though far off. The biggest issue here might be possible oscillation (output transistors and drivers).
There is obviously a certain amount of "give" when matching parameters that will depend on the exact circuit. You can sometimes use input pairs with lower Ic ratings if the circuit will never get close to the limits. The same holds true for VCE breakdown ratings. If your amp uses +/- 60 volt rails, you should use parts rated for 140 VCE or better. Using parts rated at 120 VDC breakdown leaves no safety buffer and high AC line voltage could put the transistors in a circuit with +/- 130 VDC applied. There is an unknown safety buffer in the design of the transistors, but you never, ever want to go there. It's clear you could use 250 V devices without a problem in a circuit that would call for 140 VDC devices, but 100 VDC devices would probably (as in should) blow. I've seen units working for a couple years with outputs rated for under the total supply voltage. Why they didn't fail soon after installation was just luck.
Anyway, make your list. It's great learning material and gets your head thinking along the right lines for fixing this amplifier.
Matching. Differential input pairs must always be matched for beta. The tighter, the better. I would also match any other pairs of transistors that appear to be differential. Matching compliments isn't required, but matching can improve performance. Output transistors should also be matched in each bank. So NPN parts matched together, the same for PNP transistors. Motorola (On-Semi now) once found that by matching the output transistors, you could reduce distortion by a factor of 10X before feedback is added. I would say that matching is worthwhile given how much attention we give to reducing distortion everywhere else in amplifier circuits.
When you test parts, things like transistors are checked out of circuit. Every other component is the same, you have to lift at least one leg to get an accurate reading. There are situations where you can measure a part in circuit and get accurate readings, but until you know more, assume they all have to come out.
-Chris
You need to have equal or better ratings for VCE, Ic, and power dissipation. Gain should be somewhat close and is normally given as a range of values. fT should be in the same ballpark and will sometimes work even though far off. The biggest issue here might be possible oscillation (output transistors and drivers).
There is obviously a certain amount of "give" when matching parameters that will depend on the exact circuit. You can sometimes use input pairs with lower Ic ratings if the circuit will never get close to the limits. The same holds true for VCE breakdown ratings. If your amp uses +/- 60 volt rails, you should use parts rated for 140 VCE or better. Using parts rated at 120 VDC breakdown leaves no safety buffer and high AC line voltage could put the transistors in a circuit with +/- 130 VDC applied. There is an unknown safety buffer in the design of the transistors, but you never, ever want to go there. It's clear you could use 250 V devices without a problem in a circuit that would call for 140 VDC devices, but 100 VDC devices would probably (as in should) blow. I've seen units working for a couple years with outputs rated for under the total supply voltage. Why they didn't fail soon after installation was just luck.
Anyway, make your list. It's great learning material and gets your head thinking along the right lines for fixing this amplifier.
Matching. Differential input pairs must always be matched for beta. The tighter, the better. I would also match any other pairs of transistors that appear to be differential. Matching compliments isn't required, but matching can improve performance. Output transistors should also be matched in each bank. So NPN parts matched together, the same for PNP transistors. Motorola (On-Semi now) once found that by matching the output transistors, you could reduce distortion by a factor of 10X before feedback is added. I would say that matching is worthwhile given how much attention we give to reducing distortion everywhere else in amplifier circuits.
When you test parts, things like transistors are checked out of circuit. Every other component is the same, you have to lift at least one leg to get an accurate reading. There are situations where you can measure a part in circuit and get accurate readings, but until you know more, assume they all have to come out.
-Chris
Sometimes you get a problem that is very hard to find.
I had one amp where I took out every semi and checked them as OK.
In the end it turned out to a bad transistor anyway which had HFE of 1 !
It should have been at least 100. I had to test with a HFE checker.
I bought a Maplin 50 watt amp which was faulty.
All components checked ok.
Turned out someone had put in a pnp transistor instead of an npn. The transistor checked fine on a semi analyser.
The transistors number had rubbed off so I couldn't see.
I had one amp where I took out every semi and checked them as OK.
In the end it turned out to a bad transistor anyway which had HFE of 1 !
It should have been at least 100. I had to test with a HFE checker.
I bought a Maplin 50 watt amp which was faulty.
All components checked ok.
Turned out someone had put in a pnp transistor instead of an npn. The transistor checked fine on a semi analyser.
The transistors number had rubbed off so I couldn't see.
thing like keeping the beta's within a narrow range so as to not upset the operating point of the next stage (it is direct coupled) is important. and Gbp i once selected a sub that had way higher frequency of operation and caused ultra sonic oscillation that destroyed outputs.
it been a long time since and i've probably forgotten more than i currently remember about the A700 and as someone who has himself been school'ed by Anatech i'll leave you in his capable hands...
it been a long time since and i've probably forgotten more than i currently remember about the A700 and as someone who has himself been school'ed by Anatech i'll leave you in his capable hands...
i have repaired a lot of japanese amps in the 80's and early 90's
and the first thing i would do is to hunt down the schematics or
service manuals.....
i then take tine to study and compare with the actual amp.
amps seldom burn two channels at a time so most likely
there is still one good channel to compare....
about replacing parts, make sure you got them all...
otherwise the amp still smokes...
i have found that a series lamp tester about 100watt
incandescent lamp will be invaluable in your repairs...
and the first thing i would do is to hunt down the schematics or
service manuals.....
i then take tine to study and compare with the actual amp.
amps seldom burn two channels at a time so most likely
there is still one good channel to compare....
about replacing parts, make sure you got them all...
otherwise the amp still smokes...
i have found that a series lamp tester about 100watt
incandescent lamp will be invaluable in your repairs...
links to documentation....https://www.hifiengine.com/manual_library/yamaha/a-700.shtml
the series lamp tester will turn bright then very dim when your repair was successful.
otherwise if still have problems, the lamp will be bright and stay bright....
a good amplifier will have its output offset voltage at no more than +-100 mV,
of course zero is better if you can set it, or if that amp has an output offset adjustment...
additionally, a good amp will have its output bias adjustments working...
so these are two things to look out for....the rest is up to you....
good luck....
the series lamp tester will turn bright then very dim when your repair was successful.
otherwise if still have problems, the lamp will be bright and stay bright....
a good amplifier will have its output offset voltage at no more than +-100 mV,
of course zero is better if you can set it, or if that amp has an output offset adjustment...
additionally, a good amp will have its output bias adjustments working...
so these are two things to look out for....the rest is up to you....
good luck....
Hi Fellas,
That lamp trick can do some mighty funny things with some amplifiers that are normal. I really, really strongly suggest that you buy a little variac and use that to power these things up. A little 2 ampere model will power most things up, you don't need to use it to power test an amplifier. You only need to use it to power an amplifier up so you can check bias currents, offsets, things like that. Or to partially power things up so you can troubleshoot something with a repeatable environment.
Having said this, I have just used my own variac to find a really weird problem in a Yamaha P-2200. Never went above 1 ampere and in most cases after the surge, less than 1/4 ampere. A variac really is the way to go. Keep an eye on Ebay for one. I think a small brand new one runs about $120 Canadian. Just guessing as I haven't bought one myself for at least 25 years, except for my Ebay find that was about $60 for a B&K 1655 AC power source. Meters, switches, leakage tests. That was a really good deal had it not had problems when I got it.
-Chris
That lamp trick can do some mighty funny things with some amplifiers that are normal. I really, really strongly suggest that you buy a little variac and use that to power these things up. A little 2 ampere model will power most things up, you don't need to use it to power test an amplifier. You only need to use it to power an amplifier up so you can check bias currents, offsets, things like that. Or to partially power things up so you can troubleshoot something with a repeatable environment.
Having said this, I have just used my own variac to find a really weird problem in a Yamaha P-2200. Never went above 1 ampere and in most cases after the surge, less than 1/4 ampere. A variac really is the way to go. Keep an eye on Ebay for one. I think a small brand new one runs about $120 Canadian. Just guessing as I haven't bought one myself for at least 25 years, except for my Ebay find that was about $60 for a B&K 1655 AC power source. Meters, switches, leakage tests. That was a really good deal had it not had problems when I got it.
-Chris
as with any direct coupled amplifier you have to check all parts and replace anything
connected with the failed parts......soon as you power up a direct coupled amplifier with even one small bad resistor or leaky transistor the problem only gets worse...
My first Marantz repair was a 4300.. and I had missed one transistor under the heat sink..
it was leaky from base the emitter... I had rebuilt the whole channel and checked every part or thought I had...... POWER ON ,, RELAY CLICKED... BOOM !! SPEAKER BOTTOMED OUT ,,SMOKED... THEN FIRECRACKER NOISE... ITS CALLED THE SCHOOL OF HARD KNOCKS... don't let anyone tell you they never screwed up....mine cost me 110.00 back in 1979...........my only advice... check every part for correct value before applying power........yeah its a pain to pull some of the parts.. even the best of us sometimes have problems with a complicated amplifier like that......
connected with the failed parts......soon as you power up a direct coupled amplifier with even one small bad resistor or leaky transistor the problem only gets worse...
My first Marantz repair was a 4300.. and I had missed one transistor under the heat sink..
it was leaky from base the emitter... I had rebuilt the whole channel and checked every part or thought I had...... POWER ON ,, RELAY CLICKED... BOOM !! SPEAKER BOTTOMED OUT ,,SMOKED... THEN FIRECRACKER NOISE... ITS CALLED THE SCHOOL OF HARD KNOCKS... don't let anyone tell you they never screwed up....mine cost me 110.00 back in 1979...........my only advice... check every part for correct value before applying power........yeah its a pain to pull some of the parts.. even the best of us sometimes have problems with a complicated amplifier like that......
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.