The output devices in my Musical Fidelity A300 integrated amp blew out on the left channel and I need to replace them. There are 4 Sanken SAP15 NY 9819 and 4 Sanken SAP15 PY 9013 devices in the amp.
I can get replacement ones from Profusion PLC for $5 each. They're from the same gain group (Y). Or I can send it to the importer for replacement at $25 each + work + 2-way shipping + at least 2-week wait.
So the question is, how tight are the tolerances in a gain group like this? Would it be ok to match them with random ones from the same gain group?
Please, I really need this info.
I can get replacement ones from Profusion PLC for $5 each. They're from the same gain group (Y). Or I can send it to the importer for replacement at $25 each + work + 2-way shipping + at least 2-week wait.
So the question is, how tight are the tolerances in a gain group like this? Would it be ok to match them with random ones from the same gain group?
Please, I really need this info.
Hi,
are there individual source resistors for each FET in the power outputs?
If yes then group them into Hi Vgs & Low Vgs, using the Hi groups in Ch1 & Low groups in Ch2.
If No then closer matching will be my recommendation but await other opinions.
BTW I think the letter code grading will be for admittance(yfs) which you may find difficult to measure consistently. Varies with temp during each reading.
You can easily measure Vgs at a prescribed current. Try to match this to Iq.
are there individual source resistors for each FET in the power outputs?
If yes then group them into Hi Vgs & Low Vgs, using the Hi groups in Ch1 & Low groups in Ch2.
If No then closer matching will be my recommendation but await other opinions.
BTW I think the letter code grading will be for admittance(yfs) which you may find difficult to measure consistently. Varies with temp during each reading.
You can easily measure Vgs at a prescribed current. Try to match this to Iq.
Thanks. The Sanken SAP15N/P are monolithic bipolar Darlingtons with internal temperature compensation diodes. The Y-code is the gain group, but it's really wide, from 8000 to 20000 hfe.
Hi,
mistakenly thought SAP were FETs.
Just substitute Vbe for Vgs & follow previous.
The crossover distortion is likely to be the most serious concern when replacing outputs and that's why I think matching Vbe @ Iq will be more effective. Current gain disrepancies further up the voltage swing will mean unmatched currents but the emitter resistors (if fitted) will help balance them out.
mistakenly thought SAP were FETs.
Just substitute Vbe for Vgs & follow previous.
The crossover distortion is likely to be the most serious concern when replacing outputs and that's why I think matching Vbe @ Iq will be more effective. Current gain disrepancies further up the voltage swing will mean unmatched currents but the emitter resistors (if fitted) will help balance them out.
Ok. So it seems that it's probably more cost effective to send the unit to the importer than buy a bunch of the devices and match them by hand.
No , not a bad amp.. in fact ,rave reviews...
http://www.soundstage.com/revequip/musicalfidelity_a300.htm
Can't find a schematic but I don't think matching is a problem
with these sankens (High end- made just for audio- OP devices)
Since these are darlingtons with diode comp.
it's a good bet they are IC driven.
Amazing for the high end (now I really want to build my sanken/
LM4702 amp)
Good luck, OS.
🙂
BTW .. something to note ..internal emitter resistor can be bypassed by connecting an external one.
you might have just burned internal resistor-data says
it is designed to fail before transistor junction does(s). strange.
http://www.soundstage.com/revequip/musicalfidelity_a300.htm
Can't find a schematic but I don't think matching is a problem
with these sankens (High end- made just for audio- OP devices)
Since these are darlingtons with diode comp.
An externally hosted image should be here but it was not working when we last tested it.
it's a good bet they are IC driven.
Amazing for the high end (now I really want to build my sanken/
LM4702 amp)
Good luck, OS.
🙂
BTW .. something to note ..internal emitter resistor can be bypassed by connecting an external one.
you might have just burned internal resistor-data says
it is designed to fail before transistor junction does(s). strange.
- Status
- Not open for further replies.