• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Passing light through AMP PCB reveals exposure to elevated heat

Should I continue working on this?


  • Total voters
    5
  • Poll closed .

Alex Santos

New Member
Hi everyone

For the last 4-5 days I have been buried in measuring every resistor on a NAD 7240PE's AMP PCB. The PCB suffered a failure years ago and was shelved. I thought it was time to learn electronics — wow what a trip.

Anyway, after having measured every resistor value and comparing my actual results to expected results as per the service manual I have a long list of 37 resistors to remove, 1 of which surely saw so much voltage go through it that I can't make out the colored bands. A portion of these 37 need to measure out of circuit to verify.

As I prepared for desoldering, I shined a light through one side of the AMP PCB to help me sort out where the through holes are and it's clear as day that the substrate has gone through some heavy thermals while it was last switched on. The solder side (trace side) appears to be physically intact so despite the concern the traces might be in good enough shape to have new components solder in place. I don't notice any warping to the PCB.

What are your opinions/recommendations/advice. I know that the photos are helpful but sometimes not enough. With what you can see, is it worth continuing on this component replacement path?

Photos for your evaluation
Light_pass_through_take2 - 1.jpeg Light_pass_through_take2 - 2.jpeg Light_pass_through_take2 - 4.jpeg Light_pass_through_take2 - 7.jpeg Light_pass_through_take2 - 8.jpeg Light_pass_through_take2 - 9.jpeg Light_pass_through_take2 - 10.jpeg Light_pass_through_take2 - 11.jpeg Light_pass_through_take2 - 14.jpeg
 
Last edited:
Register to hide this ad
Anyway, after having measured every resistor value and comparing my actual results to expected results as per the service manual I have a long list of 37 resistors to remove

Did you measure them out of circuit? i.e. did you lift one leg from the PCB before measuring? In circuit values will be modified by surrounding circuitry.

It looks like that entire area of circuitry has been running hot for some time; the discolouration is over a fairly wide area of the board, rather than one specific component location. This may be 'normal'... If it isn't normal, it is likely that some active component is faulty, causing abnormal dissipation in a whole bunch of components.
 
have worked on lots of stuff with similar heat stress. Usually its a sign of less than ideal cooling from new, but it doesn't neccesarily mean its not fixable. If it runs warm as part of normal operation I'd probably figure out which part(s) are toasty and elevate them off the board for cooling or add a heat sink or something to keep this from getting worse.
 
Usually its a sign of less than ideal cooling from new, but it doesn't necessarily mean its not fixable.

I only considered this to be a one time event but you're right, it' very reasonable to think that this was something that took time to develop. As you suggest, and I can see it, that over time components that functionally operate at a high temp gradually 'baked' the substrate as it were. I am glad I posted the question because I had perceived the change in the substrate of the PCB as a one time event. Good thinking!

That said, there are a few resistors that, even though elevated off the PCB by the manufacturer show the through holes as thermally stressed — they are in fact nearly black but only around the through hole. The resistors themselves however, measure accurately. On the other hand, one of the photos I posted, the one with the blue out of focus cable in the foreground, shows two empty resistors locations, R414 and R424. Those resistors were likely flush with the PCB and running very hot because the substrate was found to be solid black.

I will pay close attention to these darker areas and evaluate if I can solder replacements with more clearance while being mindful of any cables that might run near them; I wouldn't want cables to start melting. :)

Thanks a million for the reply! You certainly made some very interesting and valid observations that I never considered.
 
Last edited:
Did you measure them out of circuit?

It looks like that entire area of circuitry has been running hot for some time …discoloration may be 'normal'... If it isn't normal, it is likely that some active component is faulty, causing abnormal dissipation in a whole bunch of components.

No I've not measured out of circuit. Most of the resistor values fall very precisely in their expected spot and/or within their tolerance. Those that didn't, I double checked by remeasuring their values in case there was human error on my part.

All resistors that have values that differ from the service manual will be pulled with one leg out of circuit as you suggest and measured anew. If I still get unexpected values I will refer to a resistor chart to confirm incase there is a documentation issue with the service manual. In terms of documentation defects in the service manual, I can tell you that the PCB has locations marked for a resistor but instead I find a jumper, there are 3 such findings off the top of my head (notes are not in front of me right now). In some cases a single resistor type measures with the same value but different from that documented in the service manual. So there are some suspicions I have re the service manual on this front. Thankfully, the PCB's traces themselves look to be in visibly good condition and no warping or cracking in those locations has been detected through visual inspection. I am going in with some magnification wrapped around my head to confirm if any cracking may have occurred.

Thanks for the tip @cpt_paranoia
 
Last edited:
Alex. It is good that you are asking questions. Continue to do that and resist the newbie urge to simply replace lots of components. I have seen your posts in a couple other threads so I know that you are trying to fix a broken amp, and you have also referred to replacing lots of (or all?) capacitors.

Doing shot-gun (en masse) component replacement can easily create new problems, especially if the person replacing those parts is not experienced in this type of work. Troubleshooting can be difficult enough when there is one underlying problem, especially when troubleshooting via internet forum posts. But when more than one problem exists, troubleshooting becomes exponentially more difficult for those trying to help you. MANY repair attempts fail. So the best advice I can offer is "first, do no harm".

With help from AK posters, try to evaluate the reason for the amp failure by focusing only on components that are directly related. Don't consider replacing all capacitors or every resistor that shows some signs of prior heating.

As for the empty resistor locations: one important thing for you to attempt to determine is whether someone else has ever done any work in this amp. If so, the picture becomes more complicated. But also know that empty component locations, and locations on a PCB that are jumpered even when the PCB has a marking indicating some type of component, are common. It was common practice for manufacturers to use the same PCBs in multiple amp models. In those cases, amp models below top-of-the-line typically would have some (or many) either empty or jumpered component locations. A closer pic of the empty resistor locations that you are concerned about would be helpful.

As for your resistors, I did see in another thread a pic you posted shows one obviously cooked resistor. But generally, most types of resistors can handle heat and repeated cycles of heating/cooling. There are exceptions, notably fusible resistors and also, perhaps, carbon composition. Those two types are known to drift in value over time. Not all of them do, but they are worthy of investigation, especially the fusible resistors as they are a well known trouble maker. But not all amps have fusible resistors. You need to determine whether your amp has them.

As part of your learning electronics, it would benefit you on this project to learn to visually ID the 4 common types of resistors: carbon composition, carbon film, metal oxide, and metal film. Fusible resistors will most likely look like metal film. The three types I put into italics are generally robust and stable over time and heat cycles.

As for measuring resistance values, as has been mentioned, surrounding circuitry can cause readings that do not match the nominal resistor value. One thing to be aware of regarding this, is that in cases where surrounding circuitry does influence resistance readings across a resistor, the resulting value should read lower, not higher, than the resistor's nominal value. This is because that additional connected circuitry creates parallel circuit paths. If a resistor reads too high, in circuit or not, then that resistor is out of spec.


As for the heat discolored areas on your PCB there are several things to consider:

---Discoloration is an indicator of heat yes, but discoloration in and of itself is not a problem, and is not at all uncommon. It is not possible to determine anything with certainty from your pics. If a PCB gets too overheated, yes it can cause damage to pads and traces. But I believe that more commonly, excessive heat discoloration on a PCB is more important as a clue as to which components to check, and which solder joints to check. Too much heat can damage solder joints, and faulty solder joints are sometimes difficult to identify visually. You should google "cold solder joint" and "halo solder joint", learn to visually identify these, and inspect your amp carefully, especially in over heated areas, for bad solder joints. It only takes one bad solder joint in just the right (or wrong) place to cause serious issues.

---Too much heat for prolonged periods can also hasten the demise of electrolytic capacitors. Heat from a normally hot running component (certain resistors, diodes, transistors) can radiate to other components through the air. And heat will also travel via conduction through components' leads and PCB tracks. So any electrolytic in close physical proximity to a normally hot running component is a candidate for replacement in amps that are older. Typically, this situation would be more likely to occur in the power supply section of an amp. Also note that some components do indeed run "hot" normally, hotter than is comfortable for us to touch, and that is okay. So don't let that throw off your assessments. For resistors you can always calculate (using Ohms law) the amount of power (as heat) being dissipated and determine whether that resistor is working within its rated power capacity. As part of your learning curve...another thing that you should be working towards is the ability to identify major sections of your amp, both on the schematic and on the actual PCBs. Components are typically (but not 100% as a rule) grouped together by section: amplifier, power supply, tone, phono, protection, etc. The terminology for these sections will vary somewhat by manufacturer, another thing to be aware of.
 
Last edited:
have to agree with Roger2 on this. My usual approach is to make the thing work with as few replacement parts as possible. Basically just stuff that is measurably bad. Once it is essentially functional, I'll go further but I don't like to do mass component changes unless I have a known starting point.
 
…you have also referred to replacing lots of (or all?) capacitors.

Firstly, thank you for such a thorough and well advised reply. I sincerely appreciate it.

On the electrolytic capacitors. As I understand it, replacing old electrolytic is good practice only because of their age. I do understand that recapping is not repairing. Every electrolytic capacitor that has been desoldered was laid down on a sheet of paper, the location where came from on the board is written next to it a photograph is taken. I also took care to measure their values out of circuit, nearly all of them are in spec. What I don't have is an ESR meter but I plan to purchase Bob Parker's B2ESR from https://evbesrmeter.pt. Only two caps were no good, one had the 'can' blown straight out and lying around on the PCB, the leads still soldered in and lots of little bits of silky like paper about. The other was bulging. Other than those two, my multimeter reported back good values for those that had been removed and there was no sign of bulging or leaking. I was actually amazed because on vintage computers caps this old will have already leaked. If ESR and Farad values are good should I solder them caps back in?

Doing shot-gun (en masse) component replacement can easily create new problems.

I do agree. I hope I justified my approach when it came to electrolytic capacitors above but please advise me.

On resistors, I had to measure them all, one by one. I am documenting every action, it has taken me days just to do meter the values for resistors. I found some to be questionable in terms of actual values versus documented values. Now, I do understand that documentation (service manual) may have defects so I am being cautious about it. Here is how I am going about this, one by one, I am carefully desoldering one of the resistor leads out of circuit and measuring anew. I have already found that this method revealed what I had written down as a potentially bad resistor was actually good, which leads me to believe that another component (probably in series I don't know) was skewing results. In summary, only resistors whose values didn't match the documentation are being remeasured with one lead out of circuit. If it is good, I will update my notes and it gets soldered back in.

Don't consider replacing all capacitors or every resistor that shows some signs of prior heating.

Someone in this thread mentioned that the discoloration (PCB substrate having isolated areas that are darker - a sign of overheating) is a sign of overheating but that it likely didn't happen at once, that it happened over time. I tend to agree. As a result, I will follow your advice, it sounds reasonable. (I guess I should solder back those caps that measure nicely - will order the ESR meter soon)

As for the empty resistor locations: one important thing for you to attempt to determine is whether someone else has ever done any work in this amp. If so, the picture becomes more complicated.

To be clear, I did buy this amp used several years ago and it did work perfectly. I used it two or three times after purchase then I flew back to Ireland where I was working at the time. Half a year later, I return from Ireland, power up the amp and poof, a little bit of yellowish smoke came out the back of the unit. I don't recall hearing any noise coming from inside the amp. I turned it off as quickly as possible, unplugged it and thought to rock the amp left and right in my arms, noticing that something was rattling around inside. At that point I simply stored it on a shelf where it has been for nearly 10 years. Here I am now.

So far, all my attention and work has been isolated to the amp PCB, I have reflowed all solder joints, removed the electrolytic capacitors, removed the large PNP/NPN transistors flush with the heatsink. Ahh, with these transistors I found one that is not original to this amp so indeed, there was a repair done at one time - I made mention of this in one of the threads I posted. I checked https://www.nteinc.com/search.php to learn if any of the transistors are referenced here and some were. All the removed transistors, their mikas and screws are zip bagged individually to make it easy to put back with some fresh white silicon paste. Off memory, most measure correctly but the 'imposter' transistor measures out of spec, is of a different brand and its package is slightly different. Who knows, maybe this transistor was the troublemaker. Maybe while I was away someone cranked up the volume too high and this transistor blew cascading failure to other components? Again, everything is documented so that nothing goes out of place.

As for the empty resistor locations: one important thing for you to attempt to determine is whether someone else has ever done any work in this amp.

Sorry, maybe I wasn't clear. To be clear, the service manual has a few pages, a section of the service manual, titled "Electrical Parts List" where components are described. On another page titled, "Amplifier PCB Parts Location" it shows R461, R463, R464, and R462 as resistor locations. In these locations, the PCB had no resistors installed, instead they had what are factory installed jumpers.
Jumpers - 1.jpeg Jumpers - 2.jpeg
Jumpers_on_PCB - 1.jpeg Jumpers_on_PCB - 2.jpeg Jumpers_on_PCB - 3.jpeg

I did see in another thread a pic you posted shows one obviously cooked resistor. … it would benefit you on this project to learn to visually ID the 4 common types of resistors: carbon composition, carbon film, metal oxide, and metal film.

In this receiver, nearly all the resistors are carbon composition (as per the service manual). Those cooked ones I will remove, I think it is worth replacing those because something physical has gone on to merit the same. As noted earlier, resistor values when good were spot on but others are way off — I will check them out of circuit to be sure.

On visually identifying 4 common types of resistors … thank you! I will research this for sure.

Oh and yes, I already checked on resistor so far, out of circuit and its value was spot on. I didn't know this until I saw it. I know that capacitors should be checked out of circuit but I didn't know this applies to resistors. This is why so many had incorrect actual values … I was measuring in circuit. Learning as I go along! :)

On using the discolored areas as a reference to double check solder points, noted, I shall do this. Granted I have reflowed and when doing so up close, nothing really caught my attention. Still I will check again and research both "cold solder joint" and "halo solder joint".

The last paragraph you shared, although split up into a heat related section and a part on acquianting myself with sections of the PCB and schematics. The latter I sense is still some ways away. I spent a lot of time looking at this PCB and it does indeed seem somewhat divided equally, it's not an exact mirror but there does seem to be two equal parts as it were, suggesting that one side is for one L and the other for R speaker.

Roger, I can't thank you enough for devoting so much time and attention to what I am doing. I actually resist posting too regularly on what I am working on. I tend to usually address immediate questions by doing some research first. When I am stuck or what I read is not clear I may eventually show up here for help. I just don't want to bother people too much even though you encourage me to ask. I try to keep things off the spam scale as it were. That said, it's a blessing to have this community forum. Every interaction I have had here has been truthfully, a wonderful experience.

Thank you!
 
Alex, what seems to be wrong with the amp?
Simply? I bought this amp used years ago. It worked 2 or 3 times. I then went back to Ireland for about 6 months, came back, powered it up and it smoked. Since that time is was shelved (probably going 8 years now). I saw an opportunity to dive into the abyss of electronics. So I am trying my best to see what I can do. It may be a journey of many mistakes but I hope it leads to a repair. At this rate, it could be months. I am patient though. :)
 
Agree with everything above. See this thread- https://audiokarma.org/forums/index.php?threads/opinions-needed-nad-7240pe-worth-it.826144/

I have one of the "pe" receivers and think it's great, but it's a difficult repair even for someone with experience and lots of test equipment. The "pe" part adds a lot of complication to the circuit. I voted for you to stop working on it. IMO, you should work on a good few other things and then tackle this after you have more experience, lest you turn it into a doorstop. IMO, it's worth fixing and there are far better things to wreck first! It's likely to become a rabbit hole and sour you on the whole DIY repair idea.

I'd note that my "pe" amp also had dirty switches (and maybe pots) that didn't respond to Deoxit and required cleaning with pure isopropyl alcohol before they'd work at all. Not sure how they'll hold up over time, but you'll never fix the circuitry if you don't get the controls working well first.
 
Only two caps were no good, one had the 'can' blown straight out and lying around on the PCB,

That is likely to be the cause of your bang, smoke & rattle... Certainly replace that one, and the other faulty cap. What part of the circuit are they in?

Measuring component values isn't really the way we diagnose faults; mostly it is done functionally, knowing how the circuit should work, what voltages should exist at various points in the circuit, and tracing a signal through the amp.

We measure components where we find something wrong with the circuit function.
 
If ESR and Farad values are good should I solder them caps back in?


I rarely feel motivated to re-install old parts if I've gone to the effort of pulling them, and already have a new equivalent replacement part to go in. Your call on that, but if this were mine, I'd stab the new parts in. They're already out, and I wouldn't want to stress the board again by installing them, then removing them in the near future to install yet another cap. If you've already put the new caps in, then I definitely wouldn't un-do the work.

with these transistors I found one that is not original to this amp

not too uncommon, I've been in several amps with one single transistor replaced, sometimes with a not-exact part number. I don't like to do this personally but there are opinions about it. It obviously worked for a while.

If I have to use a non-exact replacement I will do both the bad one and it's compliment in the channel or all in the channel if its more than just 2 outputs. Once thats working I usually do the same on the other channel so everything matches but thats more optional. If this was just some piece of junk I was smashing back together to make noise in the garage I might skip it.

nearly all the resistors are carbon composition (as per the service manual).
reasonably sure those are carbon film.
 
Alex. I will address your specific questions later when I have more time. At the moment I will throw out a few quick thoughts

1) I need to correct myself. Yesterday I saw a pic someone posted of a small, clearly over-cooked small resistor and this poster was receiving troubleshooting help. I tried to find that today by searching your previous posts and I was not able to find the thread or the pic. When I made that big post above I was thinking that was your thread.

2) If you already have caps out, there is no sense in putting old caps back in. That is my opinion.

3) stop ...if you can stop making changes to the amp for a day. At this point, if you want to continue, I think you need a dedicated repair thread. I have some ideas on how to best proceed with that if you can wait 14 hours or so I will post some suggestions here.

4) unfortunately, using the amp 2 or 3 times after it came into your possession is not enough time to determine whether you received it in good health. I mention this because it can be helpful to know when and how things went from "known good" to not working.

5) the three posts after mine all give good advice. From my limited understanding, the NAD "PE" amps are significantly more complex than typical amps. The number of AK people with direct experience and/or knowledge of all of the special circuitry involved is likely to be limited. I do not have knowledge required to assist in troubleshooting.

And this is very good: "Measuring component values isn't really the way we diagnose faults; mostly it is done functionally, knowing how the circuit should work, what voltages should exist at various points in the circuit, and tracing a signal through the amp."

The resistors I see in your pics are not carbon composition. Resistors referred to in your parts list as "Carb." are most likely carbon film (unless there are a few carbon comps installed that are not showing in the pics). This is most likely not an issue anyway, and I probably shouldn't have even mentioned it as it just gives you one more thing to worry about. If there are fusible resistors however, those certainly could be an issue. I have no idea whether NAD used them and you need to find out whether there are any fusible resistors (or fuse resistors) in your amp. They could possibly also go by another name, I don't know.
 
Last edited:
That is likely to be the cause of your bang, smoke & rattle... Certainly replace that one, and the other faulty cap. What part of the circuit are they in?

Measuring component values isn't really the way we diagnose faults; mostly it is done functionally, knowing how the circuit should work, what voltages should exist at various points in the circuit, and tracing a signal through the amp.

We measure components where we find something wrong with the circuit function.

That's great if the thing works well enough to do it and doesn't smoke. I don't typically recommend it. I do most of my initial troubleshooting sans power, making measurements. When I do power up, I have fairly high confidence that things are going to work, or at least not go far south. Troubleshooting is a mental exercise, not a contact sport.
 
You should also know that charred phenolic boards can be conductive due to the carbon black. Usually fairly high resistance, but even a high resistance short between traces can be a huge problem. Ditto for electrolyte from burst capacitors. Can complicate troubleshooting and operation until fixed, I found that out the hard way.
 
That's great if the thing works well enough to do it and doesn't smoke.

That's fair. My comment about not measuring every component was a thought that arose earlier, prior to the revelation of smoke and popped capacitors... And intended to address the approach of measuring every component in a design, which, as I pointed out, is difficult to do with components in-circuit. Unsoldering every component just to measure that they are fine has potential to cause further issues, not to mention being an awful lot of work.
 
That's fair. My comment about not measuring every component was a thought that arose earlier, not to mention being an awful lot of work.

It is a lot of work! I assure you. I had to take a day off this project just because it has been so many hours already. I won't give up, many of the resistors values measured almost dead on relative to expected values. Tomorrow I will spend some time disengaging one of the leads off of resistors that are suspect. One that was suspect turned out to be spot on out of circuit.

Another thing to read on; "charred phenolic boards". In my case, the traces side of the PCB looks fine, probably because of the solder mask. The component side is where things look bad. There is like a "ground zero" as it were. Spots where the PCB is charred black but rather isolated. Around these very dark locations it vignettes out to normal very quickly.

We'll see. I know that this is a complex job and very large repair considering my skills. the charring on the phenolic substrate doesn't help.

More to do tomorrow.

Thank you all for the amazing sprinkles of information and all your efforts.
 
I've never seen darkened PCBs to be a problem except maybe some high impedance FET circuit. They have to be a bit more damaged to be significantly conductive. Hint on resistors, at least for carbon film and metal film. They rarely go low, only high. There's nothing you can put in parallel with a resistor that will raise the value, so if you see high, the part is likely defective. Pull a lead on those to be sure. If you see low, there's probably something in parallel with it. The caveats are that it's not a carbon composition and that the circuit has zero voltage left on charged capacitors. Any voltage at all will fool your meter. I can spot check a moderately complex board for resistance, diodes and small signal devices in maybe 5-15 minutes, so IMO it's worth doing. I usually put a shorting jumper on the caps feeding the board and I always remember to remove it before power-up...at least after that one time when I didn't.
 
Back
Top Bottom