Hi is there a specific way to fit the metal audio transistors before I start, I think I may have damaged the last set I bought, thanks in advance
Yes, there is a specific way to fit TO3 transistors but it's pretty obvious from the original fitted parts that there is an insulating washer of thin mica and special, thermally conductive - not electrically conductive grease, applied to both sides. When reassembled, each transistor should be tested for any shorts between transistor case (collector) and heatsink with the ohmmeter or the continuity buzzer selected on your DMM.
Normally, new mica washers are fitted with the new parts and a layer of grease is sparingly applied to both sides of the washer using care not to trap grit, dirt etc. in there and spoil the thermal conductivity or even puncture the washer. There must be sufficient grease to form a virtually continuous film between both sink to washer and transistor to washer faces so that the full rating can be achieved. Any excess grease (there should be a tiny amount at least) should just ooze evenly all round the edges but large excess should be removed and wiped cleanly to avoid sticky surfaces and grime build-up. Note that any grease wiped accidentally on the leads will prevent tinning and properly soldering them. The grease is often silicone oil based and this can seep over everything and does not like solder or electrical circuits in any form.
There are also thermally conductive silicone rubber washers that don't require grease and these are suitable in most cases too. Don't try to use grease with silicone washers - it defeats the purpose and does not improve thermal transfer with conformal materials like rubber.
The collector connection is made via one of the mounting bolts and this must make good, tight contact whilst both bolts must be insulated from the hole they pass through in the heatsink. Small bushes (or ferrules inthe US) are often supplied in mounting kits or can be bought separately for this. The flat face of the bush is placed on the PCB side for this connecting bolt, otherwise it won't be doing much connecting 😉 Don't forget to test that this connection to the collector is good at the PCB too.
Regarding having to replace the transistors again.....
You may have assumed that all was well when there was actually more damage upstream from the output transistors - if they were actually the problem. How did you determine they had failed? Often if one of these has failed the replacement outputs will follow in short order so check the parts - all of them by at least checking the Vbe (voltage between base and emitter, irrespective of +or -) is ~0.7V on all BJT transistors before attempting to fit speakers and you did use a Bulb-tester (Light bulb in series with the mains supply) to power up with and safely check first, didn't you??
The drivers in these NAD3020 type design amplifiers are a bit fragile for current-thirsty old dogs like 2N3055/MJ2955 and would probably fail too if those output transistors failed. The original driver types, 2SD669A/2SB649A are unobtanium now but there are plenty of cheap copies about - Profusion PLC have a good copy (Unisonic) which would be preferred to the common Ebay fakes, if required.
Normally, new mica washers are fitted with the new parts and a layer of grease is sparingly applied to both sides of the washer using care not to trap grit, dirt etc. in there and spoil the thermal conductivity or even puncture the washer. There must be sufficient grease to form a virtually continuous film between both sink to washer and transistor to washer faces so that the full rating can be achieved. Any excess grease (there should be a tiny amount at least) should just ooze evenly all round the edges but large excess should be removed and wiped cleanly to avoid sticky surfaces and grime build-up. Note that any grease wiped accidentally on the leads will prevent tinning and properly soldering them. The grease is often silicone oil based and this can seep over everything and does not like solder or electrical circuits in any form.
There are also thermally conductive silicone rubber washers that don't require grease and these are suitable in most cases too. Don't try to use grease with silicone washers - it defeats the purpose and does not improve thermal transfer with conformal materials like rubber.
The collector connection is made via one of the mounting bolts and this must make good, tight contact whilst both bolts must be insulated from the hole they pass through in the heatsink. Small bushes (or ferrules inthe US) are often supplied in mounting kits or can be bought separately for this. The flat face of the bush is placed on the PCB side for this connecting bolt, otherwise it won't be doing much connecting 😉 Don't forget to test that this connection to the collector is good at the PCB too.
Regarding having to replace the transistors again.....
You may have assumed that all was well when there was actually more damage upstream from the output transistors - if they were actually the problem. How did you determine they had failed? Often if one of these has failed the replacement outputs will follow in short order so check the parts - all of them by at least checking the Vbe (voltage between base and emitter, irrespective of +or -) is ~0.7V on all BJT transistors before attempting to fit speakers and you did use a Bulb-tester (Light bulb in series with the mains supply) to power up with and safely check first, didn't you??
The drivers in these NAD3020 type design amplifiers are a bit fragile for current-thirsty old dogs like 2N3055/MJ2955 and would probably fail too if those output transistors failed. The original driver types, 2SD669A/2SB649A are unobtanium now but there are plenty of cheap copies about - Profusion PLC have a good copy (Unisonic) which would be preferred to the common Ebay fakes, if required.
Last edited:
Hi and thanks for your reply, there was a definate short between emitter and collector on one transistor so I changed all four. Also the 180 ohm resistor keeps frying. 669a and 649a was damaged so changed them but off ebay maybe they are cheap copies.
Last edited:
Sorry but can you tell me if I have to add resistance to the output stage, I read the old components had this built in and needs to be don due to thermal runaway
The 1 ohm resistors R455,456 I think, are in series with the output transistors and offer some stability benefit but these 700 series amplifiers were built in 1992, near 15 years after the early 3020 models which did use hometaxial output transistors but they would have been scarce even then, in the late 1970s.
The transistors in the 700 series models I repaired some years ago were all common epitaxial types and of course the amplifiers were all the same in that series. I don't think there is any issue here or need to fit more resistance for epitaxials in this much later series model.
I assume by 180 ohm resistors you mean R453,454 which connect the driver emitters. These will conduct a lot of current if either the drivers or output transistors are shorted and will die in the effort of trying to pass more current to shorted output transistors than the drivers can safely supply. They will likely fail too but that's a symptom, not the cause of the problem. When there is excessive current, you must use a current limiting device to safely troubleshoot the hot spot problems before they fail.
There is a manual procedure (see Main Amplifier - Alignment procedures) for removing and refitting solder bridges to the bias resistors when replacing output transistors - make sure you read it carefully and understand that it applies to an amplifier that is working first and that is just the bias current setting method: NAD 701 Manual - AM/FM Stereo Receiver - HiFi Engine .
As before, make sure you work safe and use something like this when you power-up, with a 60-100W incandescent lightbulb in series with the mains supply: http://www.diyaudio.com/forums/equipment-tools/252386-bulb-limiter-testing.html
If you don't, you can expect a lot more burnt parts before you can even get the opportunity to measure what the problem is.
Don't use Ebay parts for repairs - It just makes no sense to invest hours and money in a broken amp, to find the cheap fake and even reasonable copies too, are often only good for a fraction of their required duty. Buy from a bona fide local, authorized components seller. If not, then regular channels of international agencies should be useful too.
The transistors in the 700 series models I repaired some years ago were all common epitaxial types and of course the amplifiers were all the same in that series. I don't think there is any issue here or need to fit more resistance for epitaxials in this much later series model.
I assume by 180 ohm resistors you mean R453,454 which connect the driver emitters. These will conduct a lot of current if either the drivers or output transistors are shorted and will die in the effort of trying to pass more current to shorted output transistors than the drivers can safely supply. They will likely fail too but that's a symptom, not the cause of the problem. When there is excessive current, you must use a current limiting device to safely troubleshoot the hot spot problems before they fail.
There is a manual procedure (see Main Amplifier - Alignment procedures) for removing and refitting solder bridges to the bias resistors when replacing output transistors - make sure you read it carefully and understand that it applies to an amplifier that is working first and that is just the bias current setting method: NAD 701 Manual - AM/FM Stereo Receiver - HiFi Engine .
As before, make sure you work safe and use something like this when you power-up, with a 60-100W incandescent lightbulb in series with the mains supply: http://www.diyaudio.com/forums/equipment-tools/252386-bulb-limiter-testing.html
If you don't, you can expect a lot more burnt parts before you can even get the opportunity to measure what the problem is.
Don't use Ebay parts for repairs - It just makes no sense to invest hours and money in a broken amp, to find the cheap fake and even reasonable copies too, are often only good for a fraction of their required duty. Buy from a bona fide local, authorized components seller. If not, then regular channels of international agencies should be useful too.
Last edited:
I made a bulb tester and powered the amp up with it but again the 180 ohm resistor just burnt out, i wanted to fix it my self but might have to put in a shop get it done its a bit above my experience with electronics
That may be a wise decision - The bulb tester, if wired in series correctly, could avert that problem by limiting the transformer's primary current to about 250 mA and indicating a bad fault immediately. The resistor might then only burn if the transistors were already shorted. As suggested though, that's only a 5p symptom of a ₤5 problem.
Be prepared for stiff charges to rectify mistaken repairs. Techs can't really afford to be sympathetic to amateur repair attempts, which can make it difficult to identify all the damage and take much longer to rectify than the original, usually well-known fault.
Whilst it may be too late to suggest this, I assume one channel was still working when you replaced all output transistors. It would have been been better to leave the working channel intact as a reference for voltage comparisons when checking your work on the bad channel. Believe me, when you are inexperienced, you really need all the help that an example like a working channel provides.
Be prepared for stiff charges to rectify mistaken repairs. Techs can't really afford to be sympathetic to amateur repair attempts, which can make it difficult to identify all the damage and take much longer to rectify than the original, usually well-known fault.
Whilst it may be too late to suggest this, I assume one channel was still working when you replaced all output transistors. It would have been been better to leave the working channel intact as a reference for voltage comparisons when checking your work on the bad channel. Believe me, when you are inexperienced, you really need all the help that an example like a working channel provides.
Hi Ian, no there was nothing working just faint noise and then it went into protection mode, I will have one last go with the new parts I have just received and new insulators if after that I will maybe put in a shop when I have time or if there's anyone on here near me who could help i would apreitiate it, thanks again ian
As said, one of the bolts will connect the TO3 transistor case (which is bonded to the transistor's collector internally) to the the rest of the circuit on the PCB. The other bolt appears to be isolated (I couldn't recall TBH) since it will be at the relevant + or - supply voltage but with no need to duplicate the connection.
Unfortunately, the partial manual I linked does not illustrate the copper PCB pattern but there is a full manual here: NAD 701 SM Service Manual free download, schematics, eeprom, repair info for electronics which shows the copper pattern faintly but well enough to see that only the left side bolts are connected (see page 10). You only need look at the copper pattern to confirm.
Unfortunately, the partial manual I linked does not illustrate the copper PCB pattern but there is a full manual here: NAD 701 SM Service Manual free download, schematics, eeprom, repair info for electronics which shows the copper pattern faintly but well enough to see that only the left side bolts are connected (see page 10). You only need look at the copper pattern to confirm.
With the power off, use the ohmmeter or continuity buzzer of your DMM to measure for continuity between the case (the collector) of the transistors and their relevant circuit node shown on the schematic. As you will be working from the top side, it will probably be easier to use the layout diagrams (manual pages 10,13) to identify the more obvious corresponding points to use.
Q416 2N3055 collector should be connected to nearby test point P406. Likewise, Q418 collector should be connected to nearby P408.
The MJ2955 collectors are a little different for access from the top. The easiest place to spot these points might be at the driver transistor collectors, i.e the centre pins of Q414 and Q418 case should be connected, as should the centre pin of Q413 with Q417 case.
Electrically, these points should connect and be at the same potential as the negative power supply for the MJ2955s, at ~ -36V when powered up. Likewise the 2N3055 cases should both be at the same potential (a tad lower due to the series 1 ohm resistors) as the positive power supply at ~ +36V. I don't suggest powering up now to prove the point but If those voltages were not present on the cases, logic dictates that you simply wouldn't have a connection.
Q416 2N3055 collector should be connected to nearby test point P406. Likewise, Q418 collector should be connected to nearby P408.
The MJ2955 collectors are a little different for access from the top. The easiest place to spot these points might be at the driver transistor collectors, i.e the centre pins of Q414 and Q418 case should be connected, as should the centre pin of Q413 with Q417 case.
Electrically, these points should connect and be at the same potential as the negative power supply for the MJ2955s, at ~ -36V when powered up. Likewise the 2N3055 cases should both be at the same potential (a tad lower due to the series 1 ohm resistors) as the positive power supply at ~ +36V. I don't suggest powering up now to prove the point but If those voltages were not present on the cases, logic dictates that you simply wouldn't have a connection.
Last edited:
Thanks for that dude I will have a go sunday and let you no how it goes, I hope it works this time
Sorry 1 more thing before I begin, the heat sink compound I have is zp360 is this ok or should I get a differant one, thanks Ian
I know little about computer CPU cooler compounds like ZP360, other than they don't need to be great electrical insulators as they are applied to the rear side of the CPU chip, which is an insulator. This one is said on 'overclockers' forums to be an average performer but some of these materials can still be somewhat electrically conductive and a little too stiff when they should not be for mounting power transistors. The grey colour associated with silver powder in the formulation may just be for appearances but I would not risk the performance and safety of your amplifier.
Compounds specified for mounting power transistors, are typically plain white and all are very good electrical insulating materials, since that may be critical in their application. Here are some economical types available retail and on-line in the UK: Heat Transfer Compounds | CPC UK
Compounds specified for mounting power transistors, are typically plain white and all are very good electrical insulating materials, since that may be critical in their application. Here are some economical types available retail and on-line in the UK: Heat Transfer Compounds | CPC UK
Thanks again for your help, I don't think I'm gona get around to having another go this weekend now as things have been a bit hectic, we're at a funeral tomorrow so try and get the white thermal grease as soon as. Thanks again
Caps
Hi Ian if I replace the caps in the amp do I have to use the same ones or can they be a differant voltage/uf thanks and that bay looks nice where you are
Hi Ian if I replace the caps in the amp do I have to use the same ones or can they be a differant voltage/uf thanks and that bay looks nice where you are
Hi
A general rule for electrolytic type caps is that you can use higher voltage types without problems other than physical size and lead spacings, types etc. Never replace any with lower voltage rated caps than those originally specified.
It's similar for capacitance value but you can usually fit even -20%, +50% values with few problems, particularly where newer parts are now a lot smaller than the originals. The tolerance range of electrolytic caps can be very wide so don't put a lot of faith in the precision of "uF" ratings - they're good to only one significant figure.
It's not often you'll have problems with film or ceramic caps that are also fitted, since they don't deteriorate like electrolytics and there is no reason to routinely replace them.
As you're approaching another power up and I'm not clear on whether you have checked the condition of all the small transistors as well (remember the Vbe measurements?) I think you should build a bulb limiter to prevent further expensive parts damage. This is essentially an incandescent lightbulb, 60W or so, wired in series with the mains voltage supply (one lead only) to the power transformer. You could build an external type that mounts in an insulated box with approved mains connector sockets, if that suits you better. Note: thoroughly insulate and take other appropriate safety measures with the assembly of mains wiring.
Here's a short example: https://www.youtube.com/watch?v=BhBhWi7c8us
Solid state amps also flash brightly with inrush current at power-up but revert to dim until some serious power is being used as in a short circuit or overload condition.
Yup, the coastline is picture-perfect here - for hundreds of km north and south. As it's spring and the days are warming up, I guess it's time for a bit of basking out there 😎
A general rule for electrolytic type caps is that you can use higher voltage types without problems other than physical size and lead spacings, types etc. Never replace any with lower voltage rated caps than those originally specified.
It's similar for capacitance value but you can usually fit even -20%, +50% values with few problems, particularly where newer parts are now a lot smaller than the originals. The tolerance range of electrolytic caps can be very wide so don't put a lot of faith in the precision of "uF" ratings - they're good to only one significant figure.
It's not often you'll have problems with film or ceramic caps that are also fitted, since they don't deteriorate like electrolytics and there is no reason to routinely replace them.
As you're approaching another power up and I'm not clear on whether you have checked the condition of all the small transistors as well (remember the Vbe measurements?) I think you should build a bulb limiter to prevent further expensive parts damage. This is essentially an incandescent lightbulb, 60W or so, wired in series with the mains voltage supply (one lead only) to the power transformer. You could build an external type that mounts in an insulated box with approved mains connector sockets, if that suits you better. Note: thoroughly insulate and take other appropriate safety measures with the assembly of mains wiring.
Here's a short example: https://www.youtube.com/watch?v=BhBhWi7c8us
Solid state amps also flash brightly with inrush current at power-up but revert to dim until some serious power is being used as in a short circuit or overload condition.
Yup, the coastline is picture-perfect here - for hundreds of km north and south. As it's spring and the days are warming up, I guess it's time for a bit of basking out there 😎
Last edited:
Apologies, had forgotten you already used a bulb limiter but can I suggest using a lower wattage bulb if you can, for better protection. Around 40W should be plenty but the range of halogen bulbs probably differs from the older types often referred to.
Ok I thought the higher rating bulb would offer more protection I'll change it from 60 to 40, never been a great lover of the heat me (if it's hot there ) and I wouldn't cool down in the sea, knowing my luck I'd get eaten by a shark lol oh and I bought another nad amp this one is a t 751 so the broke one can be done at my lesure now and my boss has two cabinets in his stables with lots of little draws in them full of components he said help your self to anything I need how goods that ?
Last edited:
- Status
- Not open for further replies.