• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Transistor Substitutions

Complement. Right

Because if you do a substitution, you want the positive and negative waveforms to match. Even if you are doing an exact replacement, You should replace it's complement. It could have been stressed during the failure of it's complement and fail in the future. Another reason... Would you feel comfortable buying a amp or other electronics item that has mismatched parts? It would seem like someone did a half-assed repair to me. I would not give that tech a good compliment!
 
The 2SB889 looks to be on the opposite stereo channel. Should I only replace complementary transistors when they are in same channel then?
 
redk9258 said:
Complement. Right

Because if you do a substitution, you want the positive and negative waveforms to match. Even if you are doing an exact replacement, You should replace it's complement. It could have been stressed during the failure of it's complement and fail in the future. Another reason... Would you feel comfortable buying a amp or other electronics item that has mismatched parts? It would seem like someone did a half-assed repair to me. I would not give that tech a good compliment!
Click to expand...
So, why would the positive and negative waveforms (which waveforms?) not match- what is it that matters as far as complementarity is concerned and why would a device that was perfectly functional prior now no longer be acceptable? The other reasons are just opinion, and I won't comment on them one way or another.
 
Audio waveform passing through the transistors in question. Maybe it would match because of the amp circuit. I don't know. The perfectly functional device MAY have received damage when the other transistor shorted. Why would you trust it?

I have a question for you. If you were doing a tune up on a car, and 3 out of 8 spark plugs looked perfectly functional, would you replace them?

Sometimes people like to replace things in sets- faucet handles, tires, light bulbs (even if one still works), etc.
 
redk9258 said:
Audio waveform passing through the transistors in question. Maybe it would match because of the amp circuit. I don't know. The perfectly functional device MAY have received damage when the other transistor shorted. Why would you trust it?

I have a question for you. If you were doing a tune up on a car, and 3 out of 8 spark plugs looked perfectly functional, would you replace them?

Sometimes people like to replace things in sets- faucet handles, tires, light bulbs (even if one still works), etc.
Click to expand...
What audio waveform? Transistors are not spark plugs.
Are you referring to a class AB output stage- if so, what do you think is the matter at issue here.
 
I would seriously consider replacing a transistor's complement in the case where the other transistor of the pair had failed. But it would depend on the failure mode of the faulty transistor - if it shorted C-E for example and especially if I didn't have an exact replacement, I would replace them both. But if the transistor in question failed for a less serious reason, and if I DID have an exact replacement - then maybe not.
 
Hyperion said:
I would seriously consider replacing a transistor's complement in the case where the other transistor of the pair had failed. But it would depend on the failure mode of the faulty transistor - if it shorted C-E for example and especially if I didn't have an exact replacement, I would replace them both. But if the transistor in question failed for a less serious reason, and if I DID have an exact replacement - then maybe not.
Click to expand...
So, lets assume that you did replace just the one. If it is a class AB stage are you assuming that the output is self biasing at a given standing current or are you going to be adjusting a pot or something similar? Or, are you just accepting whatever current you get as being acceptable?
 
wyn palmer said:
So, lets assume that you did replace just the one. If it is a class AB stage are you assuming that the output is self biasing at a given standing current or are you going to be adjusting a pot or something similar? Or, are you just accepting whatever current you get as being acceptable?
Click to expand...
It isn't about that, you are missing the point - it's about whether or not half of a failed transistor complementary pair has been damaged when its complement has failed catastrophically - why can't you understand this? - you have almost been bullying redk9258 with an interrogation about this very point which you have singularly failed to grasp!
 
Hyperion said:
It isn't about that, you are missing the point - it's about whether or not half of a failed transistor complementary pair has been damaged when its complement has failed catastrophically - why can't you understand this? - you have almost been bullying redk9298 with an interrogation about this very point which you have singularly failed to grasp!
Click to expand...
Bullying? Hardly. Yes, I get the point. If it's simply a matter of perhaps the device has been damaged and therefore it should be replaced then fine- but in my experience it is quite likely that a single device can fail and not cause measurable damage to the other one, and when it does the result is difficulty in setting the bias current. Rather than just making unqualified statements about things why don't we add a bit of nuance and explanation...
 
I am trying to figure out if Mr. Palmer is asking this sincerely or out of a state of corporate arrogance.

If sincerely it can be answered in part by COA......letting out magic smoke from a owners just repaired pretty vintage unit does cause strained relationships and direct cost to ones pocketbook.

One might study the history of PNP transistor development back 40-50 years ago to get a further perspective.

If arrogance, start repairing units for soon to be ex friends......you'll learn soon enough.
 
c_dk said:
I am trying to figure out if Mr. Palmer is asking this sincerely or out of a state of corporate arrogance.

If sincerely it can be answered in part by COA......letting out magic smoke from a owners just repaired pretty vintage unit does cause strained relationships and direct cost to ones pocketbook.

One might study the history of PNP transistor development back 40-50 years ago to get a further perspective.

If arrogance, start repairing units for soon to be ex friends......you'll learn soon enough.
Click to expand...
Possibly arrogance- who knows?
However, I am profoundly aware of the historical underpinnings of transistor development- and not just PNPs.
Perhaps the answer is as I said- some further explanation would be nice for those who are less informed and are here to learn. Have you ever heard of the Socratic Method?
 
Hyperion said:
You may need more experience of repairing Quality DC Coupled HiFi Audio Amplifiers.
Click to expand...
I've had limited recent experience in such matters, but a substantial level of experience in design of amplifiers and their ilk. Perhaps there's more to it than a simple "swap the devices out until it works" approach, and a bit of explanation might go a long way...
 
dlucy said:
Im a hobbyist, not an E.E. or pro tech, so take anything I suggest and carefully double-check with people who actually know.

That said, yes, you should look for a device that has a simialr or higher hfe gain. The various hfs gain codes (like MJE15030 "E" vs "G") is what you should look at. The gain codes will tell you "when I get these new parts, how much gain will they have".

The ft is a bit different. If I understand ft correctly, it is the frequency at which this device stops working (i.e. Amplifying). Because this is an audio frequency application, you require a ft of at least 20k Hz. Your original part was 60 MHz and the new part is way higher. That means both will amplify your audio frequencies just fine. The _trouble_ would be if your circuit is delivering higher frequencies, i.e. RF, and these newer, higher ft devices would happily amplify them. That's not good as you could get oscillation (motor boating) and the amp will work harder with no audio benefit.

Circuit designs, especially newer ones, have tiny capacitors in an amongst the driver or adjacent stages to act as a low pass filter and cut out any frequencies above 20k Hz (or so). The trouble if you use a newer device with higher ft would be if your circuit IS picking up RF and DOES NOT low-pass filter it.

Now someone with much more experience and knowledge can point out where I'm wrong and give you a better answer. I think your above choices are good ones.
Click to expand...
FT has a precise definition for transistors. It's the frequency at which the current gain becomes unity. It has other implications, but that's how it's defined.
 
wyn palmer said:
Why will replacing a device with an FT of 10MHz with one with an FT of 4MHz be better than replacing it with one of 20MHz.
Click to expand...
Because of availability of proper or ideal replacements. But I managed to fix the problem with same original devices from a local source.
 
dlucy said:
Im a hobbyist, not an E.E. or pro tech, so take anything I suggest and carefully double-check with people who actually know.

That said, yes, you should look for a device that has a simialr or higher hfe gain. The various hfs gain codes (like MJE15030 "E" vs "G") is what you should look at. The gain codes will tell you "when I get these new parts, how much gain will they have".

The ft is a bit different. If I understand ft correctly, it is the frequency at which this device stops working (i.e. Amplifying). Because this is an audio frequency application, you require a ft of at least 20k Hz. Your original part was 60 MHz and the new part is way higher. That means both will amplify your audio frequencies just fine. The _trouble_ would be if your circuit is delivering higher frequencies, i.e. RF, and these newer, higher ft devices would happily amplify them. That's not good as you could get oscillation (motor boating) and the amp will work harder with no audio benefit.

Circuit designs, especially newer ones, have tiny capacitors in an amongst the driver or adjacent stages to act as a low pass filter and cut out any frequencies above 20k Hz (or so). The trouble if you use a newer device with higher ft would be if your circuit IS picking up RF and DOES NOT low-pass filter it.

Now someone with much more experience and knowledge can point out where I'm wrong and give you a better answer. I think your above choices are good ones.
Click to expand...
Ic opamps can use devices that have FTs that are between a few hundred MHz and a few GHz. The devices are rarely used as low pass filter elements for audio, rather a specific pole or poles are introduced to control the bandwidth. This is sometimes done to restrict the RFI impact but more generally for stability issues and to maintain immunity to effects like Transient Intermodulation Distortion...
 
Sometimes lower FT devices can compromise design stability as the effective Cbc is higher and the output impedance looks more inductive and this can both reduce the phase margin of a design and result in localized oscillations. Generally higher Ft devices are a better choice- the safest choice being the original type of course.
 
Hello Again,

Continuing the question of FT, do you think it could hurt to replace Rotel RA-01 15Mhz output transistors (2SB817/2SD1047) with 30MHz toshiba or Fairchild 2SA1943/2SC5200 ?
 
You must log in or register to reply here.
Back
Top Bottom