Hi again
It's good to have checked and identified what you have. Those may be original emitter resistors and the outer casing seems similar to carbon composition resistors which have a phenolic moulded shell. If they read 1 ohm and the wires are solidly fixed, they will be fine - there is no point buying parts if you have sufficient resistors. As they are wire wound, replace with same but use low inductance type.
Earlier, I described the circuit, pointing out how many output transistors there were and how many drivers. They may look the same but may be numbered differently. Look carefully at the schematic of the output stage which clearly marks the OP transistors as A8 on the +rail and A6 on the -rail. The drivers associated with the output transistors will be wired as the schematic, i,e. different to the outputs, so don't confuse the roles and note the quantities. Top row, driver plus 5 OP. Lower row, driver plus 6 OP (all A8 NPN type)
The negative rail is a mirrored arrangement using A6 type (PNP). The schematic is clear enough to show all this. Note the 1R resistors with a parallel diode connecting the common emitter rails to the output. That accounts for the last 4 out of 48 resistors, The rest are associated with the 44 OP transistors.
OK, hopefully you are looking at the schematic, counting the parts referred to and realising that the resistors don't all do the same job.
You could expect differences accordingly. Verify that each channel measures the same differences and, assuming the resistors and transistors are intact, you should be OK to refit heavier wiring where the traces have lifted. Whilst you are looking , check to see if transistors have been replaced with different types or the damage goes further than already thought, before testing with a load.
You may be confident but I don't know what parts are there, since you haven't confirmed it.
It's good to have checked and identified what you have. Those may be original emitter resistors and the outer casing seems similar to carbon composition resistors which have a phenolic moulded shell. If they read 1 ohm and the wires are solidly fixed, they will be fine - there is no point buying parts if you have sufficient resistors. As they are wire wound, replace with same but use low inductance type.
Earlier, I described the circuit, pointing out how many output transistors there were and how many drivers. They may look the same but may be numbered differently. Look carefully at the schematic of the output stage which clearly marks the OP transistors as A8 on the +rail and A6 on the -rail. The drivers associated with the output transistors will be wired as the schematic, i,e. different to the outputs, so don't confuse the roles and note the quantities. Top row, driver plus 5 OP. Lower row, driver plus 6 OP (all A8 NPN type)
The negative rail is a mirrored arrangement using A6 type (PNP). The schematic is clear enough to show all this. Note the 1R resistors with a parallel diode connecting the common emitter rails to the output. That accounts for the last 4 out of 48 resistors, The rest are associated with the 44 OP transistors.
OK, hopefully you are looking at the schematic, counting the parts referred to and realising that the resistors don't all do the same job.
You could expect differences accordingly. Verify that each channel measures the same differences and, assuming the resistors and transistors are intact, you should be OK to refit heavier wiring where the traces have lifted. Whilst you are looking , check to see if transistors have been replaced with different types or the damage goes further than already thought, before testing with a load.You may be confident but I don't know what parts are there, since you haven't confirmed it.
Hi again
It's good to have checked and identified what you have. Those may be original emitter resistors and the outer casing seems similar to carbon composition resistors which have a phenolic moulded shell. If they read 1 ohm and the wires are solidly fixed, they will be fine - there is no point buying parts if you have sufficient resistors. As they are wire wound, replace with same but use low inductance type.
Earlier, I described the circuit, pointing out how many output transistors there were and how many drivers. They may look the same but may be numbered differently. Look carefully at the schematic of the output stage which clearly marks the OP transistors as A8 on the +rail and A6 on the -rail. The drivers associated with the output transistors will be wired as the schematic, i,e. different to the outputs, so don't confuse the roles and note the quantities. Top row, driver plus 5 OP. Lower row, driver plus 6 OP (all A8 NPN type)
The negative rail is a mirrored arrangement using A6 type (PNP). The schematic is clear enough to show all this. Note the 1R resistors with a parallel diode connecting the common emitter rails to the output. That accounts for the last 4 out of 48 resistors, The rest are associated with the 44 OP transistors.
OK, hopefully you are looking at the schematic, counting the parts referred to and realising that the resistors don't all do the same job.
You may be confident but I don't know what parts are there, since you haven't confirmed it.
Great. Forgot to mention, I have an Oscilloscope (Owon PDS-5022s), which I had bought sometime ago but never used. Will this be of any help in diagnosing the same?? Since I had only one bad resistor per channel, is it ok to leave the others alone or change the entire set?
Use the new resistors on their own, parallel with the diodes so that the others share normally, as described #21.
Certainly, if you can load-test the amplifier, you could find a fair bit of info on the 'scope.
It sounds easy, but requires very large non-inductive resistor loads. These are BIG amplifiers and need really big loads to test up to 100W at least. There are many RF dummy loads but not many 8R or 4R dummy loads. You can innovate with water cooled resistance wire but it's tricky and risky. Next, you need to feed a very clean, variable sinewave into the amp of around 0.5 V at 100, 1k, 10kHz in order to verify that the LF bandwidth, which is somewhat dependent on cap quality, is OK and HF testing tells you if the front end is fully functional, among other things.
Before setting all this up, you need to look at posts here where someone who knows what they are doing is posting 'scope pics. Look at what they are posting. Some is square wave and clipping tests which are tough tests and old designs like this won't necessarily look so good. Then, I have never tested or worked on one so I can't assume too much.
Please look up scope testing and read any manual thoroughly. I believe there is a good Babani book on analog scopes, at least. There is the complexity of the controls as one difficulty but the real problem is trying to understand what the time domain reveals of the behaviour of DC circuits and AC signals (SS amplifiers). It is anything but simple to know whether a trace is telling you good or bad news, even though the instrument looks wonderful and traces look meaningful. Take it very slowly and be careful as the ground of scope probes is usually mains grounds too. That causes problems for some amplifiers with balanced lines or floating signal earth. There, both inputs must be used in differential mode.
Certainly, if you can load-test the amplifier, you could find a fair bit of info on the 'scope.
It sounds easy, but requires very large non-inductive resistor loads. These are BIG amplifiers and need really big loads to test up to 100W at least. There are many RF dummy loads but not many 8R or 4R dummy loads. You can innovate with water cooled resistance wire but it's tricky and risky. Next, you need to feed a very clean, variable sinewave into the amp of around 0.5 V at 100, 1k, 10kHz in order to verify that the LF bandwidth, which is somewhat dependent on cap quality, is OK and HF testing tells you if the front end is fully functional, among other things.
Before setting all this up, you need to look at posts here where someone who knows what they are doing is posting 'scope pics. Look at what they are posting. Some is square wave and clipping tests which are tough tests and old designs like this won't necessarily look so good. Then, I have never tested or worked on one so I can't assume too much.
Please look up scope testing and read any manual thoroughly. I believe there is a good Babani book on analog scopes, at least. There is the complexity of the controls as one difficulty but the real problem is trying to understand what the time domain reveals of the behaviour of DC circuits and AC signals (SS amplifiers). It is anything but simple to know whether a trace is telling you good or bad news, even though the instrument looks wonderful and traces look meaningful. Take it very slowly and be careful as the ground of scope probes is usually mains grounds too. That causes problems for some amplifiers with balanced lines or floating signal earth. There, both inputs must be used in differential mode.
Thanks for the info. Why can't I use my speakers as the Load instead of a load resistor? Your explanation was informative, though I need to clarify two things...
I feed the power amp with 100Hz sine wave & play it through the speaker as the load, & start tracing the signal from the RCA inputs of the Amp following the path of the signal. This has to be on AC weighting or DC?
Secondly, as I had mentioned earlier, I have Grounded the chassis of my amp to the Earth (pin of the IEC inlet). So while probing the Amp, I can safely connect the Black lead of the probe to the Chassis itself?? My entire Audio system is connected via an Isolation transformer, so connecting the Oscilloscope to the wall outlet will be safe in this case I guess..?
The speaker is a dynamic load. Any measurement has to then factor in the load variation in the speaker as it responds to the signal. It is much simpler to use a fixed, accurately known resistive load without reactance. i.e, pure resistance without capacitive or inductive components to confuse the measurement.
I think you would probably see how pointless it would be to measure with any varying load connected. You would mainly be measuring speaker response rather than amplifier linearity.
I just reread your comments on the mains power, Jolida. You say you have an isolating transformer but the chassis and signal earth are still connected to mains protective earth. Is that correct? If the signal earth is also connected to chassis earth (which it typically is not with DIY or many commercial amplifiers), then 'scope connections won't be a problem.
I realise you wan't answers to everything here, like how to use the scope too but I ask that you read proper written instructions first, such as any that came with the scope and a good text which you will find on the internet even but it needs to be from a reputable source. Yours are basic questions and they will be answered in any explanation of basic oscilloscopes. We don't have the time or space to write a book for you and you should be capable of study, if you got this far.
This is not going to get your amplifier working properly unless you can interpret the traces and the degree of any errors seen. For the moment, it will just help you to see and understand how amplifier function is represented by a scope trace and what they are capable of, before it becomes a useful diagnostic tool for you. Unfortunately, that is going to take some study and further experience so get started as soon as possible if you want to see the results as you proceed.
Note re: the sinewave level. Audio signal generators always have a variable output to avoid overdriving and set a precise level for an amplifier being tested. Don't simply blast 0.5 VAC signal into the input as there is likely no means to attenuate it. Bring it up from zero with a preamp if necessary, measuring the amplifier output voltage as average RMS with your DMM if it is only 100Hz or 1 kHz AC and noting that the scope is read in peak to peak volts. RMS voltage is like it says, at RMS of 1/2 Peak-peak voltage.
Monitor the amplifier output voltage and keep it less than 50% peak to peak voltage of the total DC voltage across the rails - around 70V peak-peak max. - Use only 25% with the load connected, unless you want to burn it. Remember, 100W is much less than this amplifier can supply, given a chance.
I realise you wan't answers to everything here, like how to use the scope too but I ask that you read proper written instructions first, such as any that came with the scope and a good text which you will find on the internet even but it needs to be from a reputable source. Yours are basic questions and they will be answered in any explanation of basic oscilloscopes. We don't have the time or space to write a book for you and you should be capable of study, if you got this far.
This is not going to get your amplifier working properly unless you can interpret the traces and the degree of any errors seen. For the moment, it will just help you to see and understand how amplifier function is represented by a scope trace and what they are capable of, before it becomes a useful diagnostic tool for you. Unfortunately, that is going to take some study and further experience so get started as soon as possible if you want to see the results as you proceed.
Note re: the sinewave level. Audio signal generators always have a variable output to avoid overdriving and set a precise level for an amplifier being tested. Don't simply blast 0.5 VAC signal into the input as there is likely no means to attenuate it. Bring it up from zero with a preamp if necessary, measuring the amplifier output voltage as average RMS with your DMM if it is only 100Hz or 1 kHz AC and noting that the scope is read in peak to peak volts. RMS voltage is like it says, at RMS of 1/2 Peak-peak voltage.
Monitor the amplifier output voltage and keep it less than 50% peak to peak voltage of the total DC voltage across the rails - around 70V peak-peak max. - Use only 25% with the load connected, unless you want to burn it. Remember, 100W is much less than this amplifier can supply, given a chance.
Last edited:
5676 or 5876??
The schematic states that the transistors used are the 5878 & 5676. But Mr. Nelson pass says that it was the 5878 & 5876...
http://www.diyaudio.com/forums/pass-labs/38659-threshold.html
Great that you have improvement! Yes, it will likely work fine at low power, even with some of the paralleled output transistors toasted. Set your DMM to lowest resistance range or even to continuity buzzer (if it has one) and read the value in ohms across the resistors (power off) to verify how many need replacement of 5 + 6 in series in the +ve rail (2N5878) and 5+6 in series in the -ve rail (2N5676)that that I count from the schematic.
The big composition emitter resistors still look small for the job but that may be a safety feature. Nelson Pass would likely know whether you should substitute another type as these carbon composition ones are now rare and way too expensive, if that's what they actually are or even need to be.
Looking closer, the type specified on the schematic is 1 ohm, not 0.3 or 3 ohm. If the solder work there looks new, I'd say you have more a**hole work to deal with. You never know, the transistors associated with the damaged resistors may still be OK. Replacing the transistors, which may require matching for a proper restoration, would not be cheap for these obsolete parts and modern, drop-in equivalents may just not be available.
Check out those resistors and carefully measure the base-emitter voltage drop across each power transistor. These are the 2 pins and the collector connection is the case, as you can see by the connections on the PCB to the mounting bolts. NPN will read reverse polarity to PNP types but both should measure around 0.6Volts DC (power on and take care not to slip or short anything as it's curtains if you do.
Note that the transistors and resistors sets are in parallel, so don't be confused by the fact that you could be reading 5 or 6 at once - the resistors need to be the broken ones or unsoldered at one end to get individual transistor measurements. After you check B-E voltages, remove the resistors in one bank at a time of paralleled ones and replace all with equal value ones, rated about 2 Watt as I estimate, depending on what you see as the correct type and correct resistance.
Sorry if it begins to seem vague now, but hacked amps are not easy to sort out without good info. as well as the schematic. Check the schematic posted for your own assurance.
The schematic states that the transistors used are the 5878 & 5676. But Mr. Nelson pass says that it was the 5878 & 5876...
http://www.diyaudio.com/forums/pass-labs/38659-threshold.html
Yes, that seems reasonable but why not download the data sheets as I advised and check the specifications like similar Vceo, maximum power dissipation, PNP v NPN, TO3 case etc. It will help you to investigate for yourself. Go to Datasheetarchive site or simply Google the part number and select a reference like that one. Hint: 5676 is a TO66 size can and 5876 is a TO3 can. Your output transistors are TO3, right?
I guess you are looking for part numbers because there may not be any on the transistors or are they re-coded A8 and A6 as I said earlier? That is probably a spares code, for matched sets if they are necessary. It won't take a genius to distinguish which type they refer to and it is clear on the schematic anyway, assuming you can read it. If you have difficulty reading, just remember the 8 and 6 are the last digits for the the NPN and PNP types respectively.
As you posted in that 7 year old thread, it will be amusing if anyone, including NP replies.
I guess you are looking for part numbers because there may not be any on the transistors or are they re-coded A8 and A6 as I said earlier? That is probably a spares code, for matched sets if they are necessary. It won't take a genius to distinguish which type they refer to and it is clear on the schematic anyway, assuming you can read it. If you have difficulty reading, just remember the 8 and 6 are the last digits for the the NPN and PNP types respectively.
As you posted in that 7 year old thread, it will be amusing if anyone, including NP replies.
Yes, that seems reasonable but why not download the data sheets as I advised and check the specifications like similar Vceo, maximum power dissipation, PNP v NPN, TO3 case etc. It will help you to investigate for yourself. Go to Datasheetarchive site or simply Google the part number and select a reference like that one. Hint: 5676 is a TO66 size can ano 5876 is a TO3 can. Your output transistors are TO3, right?
I guess you are looking for part numbers because there may not be any on the transistors or are they re-coded A8 and A6 as I said earlier? That is probably a spares code, for matched sets if they are necessary. It won't take a genius to distinguish which type they refer to and it is clear on the schematic anyway, assuming you can read it. If you have difficulty reading, just remember the 8 and 6 are the last digits for the the NPN and PNP types respectively.
As you posted in that 7 year old thread, it will be amusing if anyone, including NP replies.
The hum goes away if there is a power fluctuation in the mains but comes back after restarting the amp. Also im beginning to get some screechy sounds from my left channel at certain frequencies now. Its easy to inspect the top section of the amp by just opening the lid, but I have no idea how to access the lower set of resistors, as the transformer is fixed to the lower part of the chassis
Hi, nearly 2 years later.
Having just reread the thread, I notice you never discussed replacing the electrolytic capacitors. I assume you still haven't and so I suggest you do, to ensure that you have a clean power supply. Usually, there is a constant, loud buzz when the electrolytics fail. I am not familiar with this in all amplifiers so perhaps it is just on the edge of failure or only a certain number of caps have completely failed. Who knows, without tests.
It should be fairly obvious that if the hum comes and goes after restarting, that the amplifier must be in a different electrical condition when you do so. Given that this is just a power amplifier, the only way this can occur is by the main power supply capacitors being faulty (dead probably). It's time to do something about them - all of them.
Having just reread the thread, I notice you never discussed replacing the electrolytic capacitors. I assume you still haven't and so I suggest you do, to ensure that you have a clean power supply. Usually, there is a constant, loud buzz when the electrolytics fail. I am not familiar with this in all amplifiers so perhaps it is just on the edge of failure or only a certain number of caps have completely failed. Who knows, without tests.
It should be fairly obvious that if the hum comes and goes after restarting, that the amplifier must be in a different electrical condition when you do so. Given that this is just a power amplifier, the only way this can occur is by the main power supply capacitors being faulty (dead probably). It's time to do something about them - all of them.
Hello, I've got a NAD C272 power amp driving a pair of sub-woofers, giving an audible hum with no music playing. Here's what I've found . . .
Speakers only connected - silent
Input leads connected at amplifier end only - loud hum
Input leads connected normally to upstream components - quiet hum, audible from listening position (6 ft from each speaker) with no music playing
No output or input connections
DC across output terminals L 125mV. R 100mV slowly rising with time from power-up
AC across output terminals. L. 22mV. R. 22mV stable with time
What can be deduced from these symptoms?
Martin
Speakers only and silent is good. Hum when connecting "open ended" input leads is normal. If you apply a short to the end of each input lead it should still be silent.
Assuming it is, then you have to find the cause. Is it a ground loop ? Do you get the hum if you connect (touch) the still sorted input leads to the audio ground of the component that would be feeding the NAD ? It should really be silent.
If the hum is pure and deep then it sounds like a mains ground issue. If there is harshness and a rasping quality to it then it could be some poorly implemented grounding somewhere internal to the source components.
DC offset is moderate but won't cause hum. The AC readings are probably not to be relied on if done with a DVM in the presence of the DC offset.
Assuming it is, then you have to find the cause. Is it a ground loop ? Do you get the hum if you connect (touch) the still sorted input leads to the audio ground of the component that would be feeding the NAD ? It should really be silent.
If the hum is pure and deep then it sounds like a mains ground issue. If there is harshness and a rasping quality to it then it could be some poorly implemented grounding somewhere internal to the source components.
DC offset is moderate but won't cause hum. The AC readings are probably not to be relied on if done with a DVM in the presence of the DC offset.
Deep and pure does sound like a true ground loop which could be tricky to find a simple solution to.
Does the hum still appear if the upstream components are all connected up but only the NAD power amp is on ?
You need to build the system up one piece at a time and see where the issue starts.
1/ Power amp and speakers are OK
2/ As above but with input leads attached and shorted are still OK
3/ Now connect those leads to the appropriate source component and disconnect all other input leads to that component. So you have NAD + leads + whatever it connects too and nothing else apart from the mains leads.
Does it hum ?
Does it hum if the source component is powered off ? (and if it does you have a mains ground loop)
The outcome of that determines the next step.
Also.. does the NAD have a rear volume control ? It looks like it has two sets of inputs from pictures. If so, does the hum vary if you use the adjustable input and alter the control ?
Does the hum still appear if the upstream components are all connected up but only the NAD power amp is on ?
You need to build the system up one piece at a time and see where the issue starts.
1/ Power amp and speakers are OK
2/ As above but with input leads attached and shorted are still OK
3/ Now connect those leads to the appropriate source component and disconnect all other input leads to that component. So you have NAD + leads + whatever it connects too and nothing else apart from the mains leads.
Does it hum ?
Does it hum if the source component is powered off ? (and if it does you have a mains ground loop)
The outcome of that determines the next step.
Also.. does the NAD have a rear volume control ? It looks like it has two sets of inputs from pictures. If so, does the hum vary if you use the adjustable input and alter the control ?
Whether or not powered......
Two things come to mind. Firstly that this is a true ground loop caused by the two ground connections in the two mains leads (do both NAD and source have 3 core earthed leads ?)
Does it hum if you disconnect the mains lead to the source component ? I'm guessing not.
That leaves either the above mentioned loop or it still hums when plugged in but off because the mains side of things is still active and live (a standby rather than true on/off switch).
Another possible cause. If the source component really is powered all the time (even when "off" because it has a standby function) then physical proximity of the two items could cause hum pickup. That's only likely if one is sat on top of the other. Worth mentioning though.
Two things come to mind. Firstly that this is a true ground loop caused by the two ground connections in the two mains leads (do both NAD and source have 3 core earthed leads ?)
Does it hum if you disconnect the mains lead to the source component ? I'm guessing not.
That leaves either the above mentioned loop or it still hums when plugged in but off because the mains side of things is still active and live (a standby rather than true on/off switch).
Another possible cause. If the source component really is powered all the time (even when "off" because it has a standby function) then physical proximity of the two items could cause hum pickup. That's only likely if one is sat on top of the other. Worth mentioning though.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.