All On-Semi product (New product is now marked MJ15003G) is supported by dealer networks and distributors. There will be many industrial service parts suppliers in Italian cities and throughout Europe. In particular, international ones like RS components: RS Components | componenti elettronici ed elettrici who also sell BUV20 in small quantities on-line. Arrow electronics are major distributors too. Local Sales Support in ITALY
You are correct with BDY56, I had not seen this smaller NAP110 model before but the rest is identical and it is a scaled-down size die BDY58. It will be difficult if not impossible to source old parts like these. Bear in mind that as a DIY repairer, small mistakes mean buying more parts again and the cost can be enormous if you must have originals. The cheap parts on Ebay etc. will be just that since originals were made by Semelab or possibly SGS- Thomson and RCA in the USA. They were obsolete more than 30 years ago. New parts offered on Ebay etc. will be cheap new copies or fakes with the only right part being the number I imagine, since no recognized manufacturer apart from Chinese factories offers to supply them.
You are correct with BDY56, I had not seen this smaller NAP110 model before but the rest is identical and it is a scaled-down size die BDY58. It will be difficult if not impossible to source old parts like these. Bear in mind that as a DIY repairer, small mistakes mean buying more parts again and the cost can be enormous if you must have originals. The cheap parts on Ebay etc. will be just that since originals were made by Semelab or possibly SGS- Thomson and RCA in the USA. They were obsolete more than 30 years ago. New parts offered on Ebay etc. will be cheap new copies or fakes with the only right part being the number I imagine, since no recognized manufacturer apart from Chinese factories offers to supply them.
"Little Diode" is usually the most expensive source for semiconductors - anywhere!
I can only guess that their sources are genuine and the product is close to original specification -
but with only a generic TO3 image and no reference to their source, who can say?
Donberg are offering SML product, which I have no idea of the source because that is a another
generic brand and again, the transistor could be anything that meets just the basic specs.
There is no datasheet offered by the manufacturer for proper comparisons which makes any
decision a guess. If they rely on the original part detail specifications to characterise their own
product, then they should state that and show or refer to that datasheet. With no assurances,
you really only can be certain of the external appearance and marking.
It's your choice but really, there should be no reason for almost any TO3 not to function, even if they are just rebranded 2N3055 or similar. It seems as though others have used generic parts anyway, from the pics posted earlier.
I can only guess that their sources are genuine and the product is close to original specification -
but with only a generic TO3 image and no reference to their source, who can say?
Donberg are offering SML product, which I have no idea of the source because that is a another
generic brand and again, the transistor could be anything that meets just the basic specs.
There is no datasheet offered by the manufacturer for proper comparisons which makes any
decision a guess. If they rely on the original part detail specifications to characterise their own
product, then they should state that and show or refer to that datasheet. With no assurances,
you really only can be certain of the external appearance and marking.
It's your choice but really, there should be no reason for almost any TO3 not to function, even if they are just rebranded 2N3055 or similar. It seems as though others have used generic parts anyway, from the pics posted earlier.
I since managed to read SGS on the second example and this should, if a genuine mark, be the correct part. Consider that this brand became ST micro in 1987 so the transistor will not be looking brand new after at least 27 years. (I would test the mark with solvent). If the price is acceptable and you feel better about the part number being right, this will be the only choice. Good luck and buy a spare set as well!
Yes the price is acceptable, they require a minimum 30 euros order though.
Also I've found an alternative: some aftermarket BDY56 from Comset Semiconductor... it is a belgian company that produces
"aftermarket for discontinue parts like transistors, SCR and Triac from major MFRs such as PHILIPS.
COMSET SEMICONDUCTORS was founded in 1974, Our aim is to reproduce discontinued, mature and end-of-life Transistors, Triacs, SCR’s and IC’s to extends product life cycle. Our product’ line is focused mainly on PHILIPS semiconductors that were produced in the eighties."
BDY56 in stock buy now | View parts datasheet pdf
Comset Semiconductors Home page
Also I've found an alternative: some aftermarket BDY56 from Comset Semiconductor... it is a belgian company that produces
"aftermarket for discontinue parts like transistors, SCR and Triac from major MFRs such as PHILIPS.
COMSET SEMICONDUCTORS was founded in 1974, Our aim is to reproduce discontinued, mature and end-of-life Transistors, Triacs, SCR’s and IC’s to extends product life cycle. Our product’ line is focused mainly on PHILIPS semiconductors that were produced in the eighties."
BDY56 in stock buy now | View parts datasheet pdf
Comset Semiconductors Home page
I'm not sure that they are original SGS...
Silly question probably but isn't an aftermarket BDY56 more original than a different transistor (MJ15003)???
This is where we started. It's just my opinion that MJ15003 are better transistors all round for audio and I would suggest as before, use them for both channels because they are fairly cheap, tough and replaceable - no issues with old, uncertain parts or poor quality copies if you buy from an authorized seller. They are also proven good in Avondale Audio's NCC200 Naim style amplifiers.
OK, other guys think that the good sound quality of NAP110 is in the original parts types which include the output transistors so you see pics of them with "aftermarket" BDY56 transistors fitted when replacements are needed. Some of these may be similar to the original parts but probably only the original part brands, such as SGS. There is resale value to consider too, if you wish to sell at a good price. If the parts are obviously not original brand or type, value will be less, no matter how good the sound.
The term "aftermarket" means much the same as generic. The parts meet or exceed the minimum general specifications for voltage, current, Ft and a few other important qualities but that does not make them equivalent to the original any more than a modern car that substitutes for a 40 year old one. Inside, the dies will be very different since manufacturing methods changed a lot over the decades. I hope I am making sense here.
OK, other guys think that the good sound quality of NAP110 is in the original parts types which include the output transistors so you see pics of them with "aftermarket" BDY56 transistors fitted when replacements are needed. Some of these may be similar to the original parts but probably only the original part brands, such as SGS. There is resale value to consider too, if you wish to sell at a good price. If the parts are obviously not original brand or type, value will be less, no matter how good the sound.
The term "aftermarket" means much the same as generic. The parts meet or exceed the minimum general specifications for voltage, current, Ft and a few other important qualities but that does not make them equivalent to the original any more than a modern car that substitutes for a 40 year old one. Inside, the dies will be very different since manufacturing methods changed a lot over the decades. I hope I am making sense here.
Of course you are. Thank you again.
Just a last thing: do the MJ15003's need to be matched?
No, you don't benefit much from matching because the Quasi-complementary design results in a very poor match at the crossover point anyway. This means that whatever you achieve by matching is overcome by that inherent non-linearity.
Actually, matching power transistors is not easy. You need a curve tracer which tests at currents of 1 amp or more for this, which is way too expensive and demanding for DIY or basic repairs. With modern parts, bought or taken from same lots in original boxes, matching will be close enough anyway. That's another reason why I suggest you buy from a proper On-semi agency where you can at least ask for same batch/pack parts. Again, it's not essential and some people have good results with random parts too. Note that genuine new parts are now marked MJ15003G. http://www.onsemi.com/pub_link/Collateral/MJ15003-D.PDF
Actually, matching power transistors is not easy. You need a curve tracer which tests at currents of 1 amp or more for this, which is way too expensive and demanding for DIY or basic repairs. With modern parts, bought or taken from same lots in original boxes, matching will be close enough anyway. That's another reason why I suggest you buy from a proper On-semi agency where you can at least ask for same batch/pack parts. Again, it's not essential and some people have good results with random parts too. Note that genuine new parts are now marked MJ15003G. http://www.onsemi.com/pub_link/Collateral/MJ15003-D.PDF
Last edited:
Finally I've replaced some parts: power transistors (cheap ones for the moment), driver transistors, burned resistors, some diodes (these parts have a yellow dot in the pic).
Switched on and no smoke, all seems fine.
But I still have 30v from speakers sockets.
Bias is 3,9mv across both emitter resistors.
The others transistors seems fine.
Help.
Switched on and no smoke, all seems fine.
But I still have 30v from speakers sockets.
Bias is 3,9mv across both emitter resistors.
The others transistors seems fine.
Help.
Attachments
These were the failed parts. Hope this does make sense for you.
Other than these I've also replaced TR9, TR10 and TR12.
Attachments
Hi. 'Good to know you are still working on the amplifier.
30V on the output says you may have had a further problem, as suggested by sgrossklass in #5. The 200 mV you measured originally was probably only due to leakage current through the burnt transistors. When you replaced them, the original fault reappeared, I suspect.
What polarity is the 30V offset and are the power supply rails about the same +/- 30V with respect to ground?
Are the IN4005 (BFY406) diodes correctly polarized, because if they have blown again,
someone must have done something very wrong, like putting mains power on the speaker leads to melt the terminals and blow the amplifier in the first instance - or so it seems? The diodes are not actually part of the circuit - just there to remove transient voltage spikes.
Let's also check the voltage amplifier to see if it's OK. Measure the collector voltages of TR4,6 and then the base voltages of TR9,10 in both channels for comparison. Note all measurements with no connections other than mains power.
30V on the output says you may have had a further problem, as suggested by sgrossklass in #5. The 200 mV you measured originally was probably only due to leakage current through the burnt transistors. When you replaced them, the original fault reappeared, I suspect.
What polarity is the 30V offset and are the power supply rails about the same +/- 30V with respect to ground?
Are the IN4005 (BFY406) diodes correctly polarized, because if they have blown again,
someone must have done something very wrong, like putting mains power on the speaker leads to melt the terminals and blow the amplifier in the first instance - or so it seems? The diodes are not actually part of the circuit - just there to remove transient voltage spikes.
Let's also check the voltage amplifier to see if it's OK. Measure the collector voltages of TR4,6 and then the base voltages of TR9,10 in both channels for comparison. Note all measurements with no connections other than mains power.
Hi Ian,
offset polarity is negative. PS rails are +33 and -33.
Yes they are, they haven't blown again since I replaced them.
Collector voltages of TR4-6 are -32,7 in the failed channel and +1,2/-1,2 in the working channel.
Base voltages of TR9-10 are -32,5 in the failed channel and +1,2/-1,2 in the working channell.
Last edited:
Thanks for the measurements. It seems that TR4 is not functional but TR6 may still be OK. You will not find these parts (ZTX653/753 or similar) so easy to find though they are still manufactured by Diodes Inc. I had assumed Vbe measurements had been taken and were OK at the input end of the amplifier but seeing your meter does not give reliable results you should recheck all measurements now, and whenever there is time to repeat the test. Don't forget to include TR7,8.
All the measurement confirmed.
TR7-8 have voltage collector: -6,90 and -30 in the failed channell; 1 and 0,4 in the working channell; VBE 0,45 and 0,96 in the failed channell; 0,22 and 0,82 in the working channell.
Others VBE are fine but TR4 has 0v?
ps. TR4-6 are ZTX752-652 even harder to find, may I replace them with ZTX653/753?
TR7-8 have voltage collector: -6,90 and -30 in the failed channell; 1 and 0,4 in the working channell; VBE 0,45 and 0,96 in the failed channell; 0,22 and 0,82 in the working channell.
Others VBE are fine but TR4 has 0v?
ps. TR4-6 are ZTX752-652 even harder to find, may I replace them with ZTX653/753?
Sure, no problem substituting ZTX653/753 - the difference is only max. voltage ratings.
Continue with Vbe measurements of the rest of the small transistors if not all checked yet. Check particularly TR6, the current source for TR4, for an E-C short which may also result in a large offset.
However, TR4 and maybe TR5 usually fail when the output stage fails, so this is not surprising. As transistors are DC coupled, they can fail in a complete chain back to the input transistor, falling like dominos. Consequently, it may seem easier to simply replace all semiconductors but you learn nothing and still may not solve the problem that way. Are you using a lightbulb limiter to protect against further damage?
Continue with Vbe measurements of the rest of the small transistors if not all checked yet. Check particularly TR6, the current source for TR4, for an E-C short which may also result in a large offset.
However, TR4 and maybe TR5 usually fail when the output stage fails, so this is not surprising. As transistors are DC coupled, they can fail in a complete chain back to the input transistor, falling like dominos. Consequently, it may seem easier to simply replace all semiconductors but you learn nothing and still may not solve the problem that way. Are you using a lightbulb limiter to protect against further damage?
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.