• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join free Log in
    Want zero display ads or expanded classifieds tools? Compare plans.

Replacement power transistors for Acoustic 220

#21
agedhorse said:
Sorting and matching are totally unnecessary for modern small signal silicon transistors in a competent circuit design.

Sorting and matching are very much a DIY hobby avocation, no different than the mythology of recapping everything they see because they “know” the caps are bad. It’s an absurd reality imo.
Click to expand...

I was talking about NOS germanium transistors needed to be sorted, not silicon transistors. Nobody claims that vintage pedal circuits are well designed. The right hfe range is needed to sound the best and maintain a low noise floor.

As for modern silicon transistors for audio circuits, stock I buy from digikey and mouser are mostly fine. I do find DOA transistors from time to time and prefer to test them.
 
#23
Back to sorting transistors. As an example, in vintage Neve circuits such as the 1073 preamp, they used (the now obsolete, options are available for modern revisions) BC184C transistor. The specs indicate a minimum hfe at a given voltage and current. Wanting to use a minimal inventory part count, BC184C’s were screened for higher hfe’s for input transistors to ensure low noise. The design could have been improved…
 
#24
agedhorse said:
DOA small signal silicon parts? I can’t recall the last one I have encountered, and that would be over quantities of hundreds of thousands.
Click to expand...


At a company that I worked at, wafers were 100% tested under static conditions and not at the maximums conditions such as junction temperature. They extrapolated to drew conclusions from those tests. The next step was packaging with machine inspection. Then further testing was performed. These were high cost items.

For low cost small signal transistors, it would be nice if 100% testing was performed under maximum operating conditions. Somehow I don’t expect that it is.
 
#25
Since I haven't experienced any failures in as long as I can remember, it's not something I worry about or even consider these days.

Screening for gain is a "way outside the box" sort of application. It's so much easier to design using the design min/max and add a little local feedback to account for any natural variation. These days, there's no problem achieving suitable gain.

This is how we get our parts (in reels of ~2000 pieces), the insertion machines also can test each part on the varifyier module at the parts pass under the test head. Since the parts we use are first quality and easily meet the data sheet specs with zero fall-outs, we typically only verify for proper function and polarity in the event of a component getting accidentally flipped on the tape. I haven't seen it happen except where the reel was installed flipped. The semiconductor manufacturers also test the parts on ATM's this way but much more comprehensively. The number of defects is minuscule under the worst of conditions.

1720132796282.png
 
  • Like
Reactions: Jeff Scott
#26
mmon77 said:
I've also never heard of output transistors "going bad". They are not like tubes. Every case I've seen, they either work or they don't. When they fail, they will likely blow a fuse, and many times take other components with them.

I would not trust that diagnosis.
Click to expand...
Sometimes you might find leaky transistors, but more often than not, it's as you describe.
 
  • Like
Reactions: JacklegBass
#27
JacklegBass said:
Although you gotta do what ya gotta do, it sure would be a shame to give up on the 220. It's probably one of the easiest amps in the world for any reasonably competent tech to repair with its simple fixed select resistors for bias and symmetry and single-ended power supply. It's not clear from your OP if the clipping is symmetrical and at what drive level. Could be bias or voltage starvation or something going on in the drive circuit, but a tech shouldn't have a problem diagnosing.
Click to expand...
Well, my amp doctor made a small drawing, something like this:
IMG_6365.jpeg

the amp doctor has a very good reputation here and I trust him for all my amps. I gave him the information about replacement transistors and he also said that these are available but that it is difficult to be sure they are no rebranded other stuff.

What do you guys think? Is this amp worth the extra investment of something like 300 euros?
I don’t understand anything about matching or biasing… :)

I really love this amp, it was all the sound i need. But with my ts-50 preamp and a power amp i can get close…

Really appreciate all the help! <3
 
#28
Audumbla De Koe said:
Well, my amp doctor made a small drawing, something like this: the amp doctor has a very good reputation here and I trust him for all my amps. I gave him the information about replacement transistors and he also said that these are available but that it is difficult to be sure they are no rebranded other stuff.

What do you guys think? Is this amp worth the extra investment of something like 300 euros?
I don’t understand anything about matching or biasing… :)

I really love this amp, it was all the sound i need. But with my ts-50 preamp and a power amp i can get close…

Really appreciate all the help! <3
Click to expand...

Well that waveform is obviously not symmetrical, crushed at the top and not at the bottom. According to the schematic in the service manual symmetry is controlled by resistor R8 within a range of 68K and 220K, typical value = 150K. Hopefully your amp doctor has already checked that.

300 euros is another matter though, that's about what the whole amp would be worth if it was working properly. Hard to justify that much repair cost. It didn't occur to me that you were overseas, that make getting parts shipped in from Electronics Surplus in the USA less attractive too.
 
#29
It’s not uncommon for there to be asymmetrical clipping on these amps due to the specific topology.

If the asymmetry is at low frequencies (say below 100 Hz), that’s an indication of one type of problem.

How close to rated power is it when the amp begins to clip?

This is almost certainly not output transistors.
 
  • Like
Reactions: mmon77 and JacklegBass
#30
agedhorse said:
It’s not uncommon for there to be asymmetrical clipping on these amps due to the specific topology.

If the asymmetry is at low frequencies (say below 100 Hz), that’s an indication of one type of problem.

How close to rated power is it when the amp begins to clip?

This is almost certainly not output transistors.
Click to expand...

Good point. Since he has the service manual hopefully his tech was using the specified 1kHz test wave, otherwise all bets are off.
 
#31
agedhorse said:
It’s not uncommon for there to be asymmetrical clipping on these amps due to the specific topology.

If the asymmetry is at low frequencies (say below 100 Hz), that’s an indication of one type of problem.

How close to rated power is it when the amp begins to clip?

This is almost certainly not output transistors.
Click to expand...
Well indeed, the clipping happens more in the highmids and not only when i am playing loud, it never was a superclean amp anyway and I am not a clean player as well…

Still crazy how loud this amp is for 120 watts at 4ohm. Last rehearsal i played with an ampeg svt350h with volume nearly maxed. The volume on the acoustic was always around 9,5 (o clock) and could do a little more…

Just tested the poweramp of the acoustic with a different preamp and the clipping is still there, so must be something in the poweramp. Actually it is not that terrible and for rocking it still sounds great, think i’ll let it go for little money. Someone will be happy!
Guys, thanks for all the tinkering!
 
Last edited:
#36
JacklegBass said:
Perhaps not, although I wouldn't rule it out. That said, the test point voltages and setup notes in the service manual, if followed, should provide an orderly roadmap for where to look.
Click to expand...
I would rule it out until the more common causes are evaluated. There's no reason to worry about the .1% likely cause while ignoring the far more likely causes. In this case, you might notice that global feedback is involved in stabilizing the operating point of the power amp at ~1/2Vcc because it's not a bipolar power supply. That's why output and driver transistors aren't going to be the problem, there's no voltage gain in these stages so they either work of they really don't work. Not much in between.
 
#37
Schematic I have identifies the outputs as 2N6254, but it is EXTREMELY unlikely the output semis are faulty. Power transistor failure is typically E-C short, in which case the amp doesn't work at all or blows the power fuse (or in the more typical case of dual+- supply rails, a speaker on fire). If a driver was failed, again no output. This is a very simplistic bare-bones circuit with a rudimentary comp/limiter tacked on, no way to adjust the bias or DC balance - no signal DC balance is always the first test. If the voltage at the hot side of the output coupling capacitor is close to 1/2 the supply rail my guess without actually having the amp on my bench would be a problem in the positive side of the overcurrent protection circuit, thus the sensitivity to load impedance.
 

Latest posts

Top Bottom
Back