Hey, I also have questions about replacement transistors...
I know that Q707 is a 2N5232 and Q705 is a 2N5322, but I'm not sure about the others...
I found somewhere that Q709 is a 2N5320, but don't know if that's correct. Anyone know for sure?
Same thing for Q703, which I read somewhere that a 2N3440 was a good replacement for it, but again, I'm not sure about that. Any help is greatly appreciated.
I am hoping to have all of these parts ordered by tomorrow night. Thanks!
I know that Q707 is a 2N5232 and Q705 is a 2N5322, but I'm not sure about the others...
I found somewhere that Q709 is a 2N5320, but don't know if that's correct. Anyone know for sure?
Same thing for Q703, which I read somewhere that a 2N3440 was a good replacement for it, but again, I'm not sure about that. Any help is greatly appreciated.
I am hoping to have all of these parts ordered by tomorrow night. Thanks!
Let me make this clear, you need to test every transistor, and you need a way to current limit the amp when you replace them and power it up for the first time. I start with a simple diode check of both junctions, and checking for C-E shorts both ways, with an ohm meter.
These are what I suggest for drivers unless you want to use older vintage parts. These may have a very high voltage rating but it is much better as far as SOA goes:
If you want to go MJE, I'd suggest the MJE15034 and complement - it has better hfe at lower currents where it is important for this design. Again add a small heat sink.
Pete B.
These are what I suggest for drivers unless you want to use older vintage parts. These may have a very high voltage rating but it is much better as far as SOA goes:
If you want to go MJE, I'd suggest the MJE15034 and complement - it has better hfe at lower currents where it is important for this design. Again add a small heat sink.
Pete B.
Q703 a 40408 is the VAS. It is a 90V, .7A, 1w part, hfe 40 to 200, ft = 100 Mhz.
The 2N3440 will probably due, it is slower and a bit more powerful, however in this application the speed is swamped by the large Miller cap. Best thing about the 2N3440 is that the case is the same. I would probably choose a better part.
The 2N3440 will probably due, it is slower and a bit more powerful, however in this application the speed is swamped by the large Miller cap. Best thing about the 2N3440 is that the case is the same. I would probably choose a better part.
paco161 said:
I'd say that's right as I think you got it from the service manual.
I don't know that part but from what I saw online it is a 50V 100mA part with hfe min of 200. 2N5210 should do if that is correct, but the 100 mA max is a bit low in my opinion.
Better is a PN4250a, I prefer from Fairchild for all these small signal types.
Careful of pinout, this case type comes with different pinouts.
You'll probably want to do both channels if you like the results.
I like the 2N3773 for output replacements if you ever need those.
Ok, so I'll still go with the 2N5232 for Q707, 2N5322 for Q705, and now a KSC2690A for Q703 as well as a PN4250a for Q709. I believe that should be everything.
However, I must go back to what I said earlier. If using these parts is going to make significant changes to the sound of the right channel, than I'd rather not do that, but if not, than I'll go for it.
And while we're here, I must dig up an old question from this thread that I'm fairly certain went unanswered and that is; what made the parts fry to begin with? At the very least, I still don't know myself. If it was because of a bad resistor on the PCB, that is no longer an issue since they've all been checked/fixed. If it was caused by something else that I don't know about, isn't there a highly likely chance that as soon as I get new parts in and try them out with full power, they'll fry again? Or was it possibly just a fluke that caused me this enormous headache??
However, I must go back to what I said earlier. If using these parts is going to make significant changes to the sound of the right channel, than I'd rather not do that, but if not, than I'll go for it.
And while we're here, I must dig up an old question from this thread that I'm fairly certain went unanswered and that is; what made the parts fry to begin with? At the very least, I still don't know myself. If it was because of a bad resistor on the PCB, that is no longer an issue since they've all been checked/fixed. If it was caused by something else that I don't know about, isn't there a highly likely chance that as soon as I get new parts in and try them out with full power, they'll fry again? Or was it possibly just a fluke that caused me this enormous headache??
Can you provide the exact list of parts that failed and how. I think I asked this already.
When did it fail, and how, were you probing around, did a probe slip? Were you driving the amp with a test signal, was there an output load?
Often when drivers go, so do the outputs.
It should sound the same since it is a high feedback design.
If you find all the failed parts, and get it back together correctly, and avoid fake parts, it should work.
You should set the bias, check for DC offset, and see that it passes a signal before you go full power. Even let it warm up a bit and determine that the bias is stable before you go full power.
When did it fail, and how, were you probing around, did a probe slip? Were you driving the amp with a test signal, was there an output load?
Often when drivers go, so do the outputs.
It should sound the same since it is a high feedback design.
If you find all the failed parts, and get it back together correctly, and avoid fake parts, it should work.
You should set the bias, check for DC offset, and see that it passes a signal before you go full power. Even let it warm up a bit and determine that the bias is stable before you go full power.
Just to be on the safe side, I am going to replace all the transistors, whether or not they're blown....
As for when they actually blew, it was the first time I plugged the amp in after having installed the new diff pair transistors, which is weird since I'm fairly certain those transistors are good. However, if I should find a new pair, I will do so, but for the time being, I probably won't. I know the R721 failed since it broke in half when I was unsoldering it. If my hunch is correct and R721 failed like I think it did, than that would compromise Q709, Q707, and Q703, which seems logical considering I saw smoke pour out of Q707 for what it's worth...
As for when they actually blew, it was the first time I plugged the amp in after having installed the new diff pair transistors, which is weird since I'm fairly certain those transistors are good. However, if I should find a new pair, I will do so, but for the time being, I probably won't. I know the R721 failed since it broke in half when I was unsoldering it. If my hunch is correct and R721 failed like I think it did, than that would compromise Q709, Q707, and Q703, which seems logical considering I saw smoke pour out of Q707 for what it's worth...
paco161 said:
What do you think is the most logical conclusion?
You got the diff pair transistors installed incorrectly.
I didn't think those came in different pin outs but check.
It would have not been an issue if you had current limiting.
Could also be that you created a solder bridge while working on it.
Or perhaps bad connections at the connector. Hope you gave these a shot of De-oxit.
I use the higher concentration brush on formula.
The failure is odd because it seems that the VAS (pre-driver) was turned full
on - probably due to an error in the diff pair. This normally should not really cause a problem
just drive the output to the negative rail. The upper driver should have turned off, which means
that the resistor should not have failed. Still, it seems that something caused high current in the
VAS, leading to the bias transistor and drivers failing. Was the output shorted?
Are you aware that the can of the metal transistors is the collector? Don't let anything brush
up against them when it is on. Scope probe ground for example.
You need to build some confidence in testing transistors.
I suggested a good meter, that is for measuring bias etc.
Do you have a simple analog ohm meter? This is what I prefer for a simple go no-go test
of the semis. Use the 1K ohm scale, and check junctions forward and reverse biased, and for C-E leakage both ways.
Pete B.
Web page on Harman Kardon Citation 12
I've recently started putting together a web-page on the Citation Twelve Power amp...You might find some interesting things there:
Citation 12 Power Amp
I've recently started putting together a web-page on the Citation Twelve Power amp...You might find some interesting things there:
Citation 12 Power Amp
Seasoned Tech scratching his head here. DC Offset issue
I Have a Citation 12 on my bench, with approx 100mv DC offset in both channels.
Here is what I have done/checked:
All on-board electrolytic caps changed
Tried new matched hfe input diff pairs
Supply rails are good
Retensioned the headers
Verified bias (shouldn't affect anyway)
ALL semis are original
It's not an amount of DC offset that will cause problems, I would like to see it lower if possible.
Thanks
Tweakjunkie Winnipeg
I Have a Citation 12 on my bench, with approx 100mv DC offset in both channels.
Here is what I have done/checked:
All on-board electrolytic caps changed
Tried new matched hfe input diff pairs
Supply rails are good
Retensioned the headers
Verified bias (shouldn't affect anyway)
ALL semis are original
It's not an amount of DC offset that will cause problems, I would like to see it lower if possible.
Thanks
Tweakjunkie Winnipeg
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.