Hi Roger
Sorry for this long post, it was not so easily avoidable. I typed as I went along, excuse the typos.
Note: Removal of components and value measurements was isolated to the Amplifier PCB. Solder reflow was done to the AMP PCB and TUNER PCB.
- Dwell time was between 1-2 seconds at ~300-350°C. Maybe hot but it helps me get in and out quick.
- I will focus answers to just the AMP PCB going forward.
- IC = yes, one 8 pin IC at location IC310
- Polystyrene Caps = yes, none of these have been removed
- Opamps = Maybe this is IC310 (described as a UPC 4570C), the package has 8850R appended to it.
- There are also many small transistors and diodes scattered about.
Yes, I am aware that some components are more sensitive to heat than others, albeit they all prefer little exposure to elevated it.
I do try to keep my dwell times under 3 seconds, just enough time to melt the solder. These solder joints are so small that the dwell time is typically a second or two at most. I also try to remember to keep the iron tip on the pads themselves, not component leads. I am not perfect but I try to follow this principal.
5) Thank you on the resistance readings!
Indeed, one was discolored and another cracked.
Others reveal incorrect values when measured
in circuit (as referenced by the service manual). In these cases I desoldered one lead and read the value anew. If the value didn't align with what was documented in the service manual I removed the resistor completely out of circuit and measured the value once more. If the value continued misaligned to that in the service manual I would refer to a resistor color chart to calculate the value as I wanted to rule out a potential documentation defect in the service manual. If the value arrived at from the color chart reflect the measured value I would consider the resistor as good and return it to the PCB.
After describing my process, I see that I should refer to a resistor color chart before removal just to rule out potential documentation defects in the service manual. I have already run into two scenarios; I remove out of circuit and the arrived at value of the resistor matches the service manual and other times it does not.
Sorry to be so lengthy, I wanted an opportunity to layout the steps I take so that you have some bearing on how I am approaching this. If not ideal, please let me know.
6) I have to buy a reliable ESR meter to say with confidence if the caps are healthy enough to return to the board.
Can you recommend one? Also, do you know if there are any meters that can reliably read capacitors while in circuit? Is there any truth to this or is it just a myth?
In terms of soldering skills:
- I made mention of two approaches I take that I think are important, dwell time and making contact with the pad and not the lead whenever possible. I feel pretty confident about my soldering skills though.
- The pads and traces do look very good indeed. Even where the PCB has seen elevated heat exposure.
- That said, I did unfortunately manage to loosen a pad and a portion of its trace. It turns out that this portion of pad/trace leads to R456. R456 was completely discolored, the color bands actually unreadable and the through holes visibly blackened by elevated heat. However, I do blame myself here because the evidence should have made it clear to me that at this point I was dealing with a very fragile part of the PCB. Having not clued in as I should have I left the amp on its side and began desoldering the fired resistor out. Firstly the amp should have been lying flat with the trace side facing me. At some point I realized I had to use solder braid to get some of that remaining solder off. The problem here was the solder wick got soldered on to the pad and I lost grip of it, it hung there for a moment and peeled the trace off the PCB, luckily it didn't tear off. Had I been more considerate I would have had the amp lying down not standing on its side where gravity contributed to 'pull down' on the pad/trace.
- To remedy this, I will need to decide if I want to cut off this peeled portion of pad/trace and jumper a wire from the resistor lead to the a portion of trace that remains fixed to the PCB or if the replacement resistor has long enough leads that I can solder on to a good bit of trace still adhered to the PCB substrate. Either way I may need to finish it with some solder mask ink and cure it with UV light. I will research how to glue it back on, which would be ideal, but it might be overly involved in terms of materials and time. Lesson learned here, be extra careful when dealing with burned out parts, the incident leading to failure may have impacted much more than just the component itself.
Soldering back in the original, but still good, caps now is the easiest way to get you to where your can get into troubleshooting. But if you do get the amp working, most likely you at some point want to install all new caps, which means a second round of heat exposure. So based on what you know at this point, do you feel like the pads and traces in yoru NAD would hold up to a second remove/replace heat exposure cycle in the future?
I agree on putting them back, it will speed up troubleshooting and eventually (after resolving the amp issue) I should probably replace the electrolytics with a fresh batch. I am of the opinion that the PCB should be able to cope with a few more rounds of solder/desolder/solder. I can't evaluate this presumption completely but from appearances and the reflow paces I put the PCB through don't appear to suggest difficulty in locations where caps would be soldered into. I think my solder skills are pretty good. I am not a pro but am confident and am very considerate of dwell time and where to begin administering the heat, not on the part lead but on the pad.
7 - My multimeter does have a diode mode. I have preliminarily checked the diodes but I want to learn more about measuring PNPs and NPNs. I will work on measuring them.
Post the model name/numbers of the original transistors and the one replacement. It is possible that the replacement is a good sub. Or maybe it is not.
I am not exactly sure what you mean here. I think you are referring to transistor that I suspect was replaced by someone before I got this NAD in my hands.
If so … Q430 should be the same as Q429. Q430 is not the original transistor It's compliment Q430 is.
Q429 as per nteinc.com should be
NTE37 (PNP)
https://www.nteinc.com/specs/10to99/pdf/nte36.pdf
With DMM in diode mode
Q430 - A1492 has no reading at all - I get OL every which way I read
Q429 - B817E with the DMM's black probe on Base, I get .583 from Base to Emitter and Base to Collector, all other combinations (forward and reverse bias) I get OL.
Thanks for tips on the thermal paste, yes not CPU GPU thermal paste but something proper like white silicone paste. I will inspect those micas to prevent any short. I heard that there are some non-conductive thermal pads that can replace micas and thermal compound. Have to read up on that I guess. And yes I cleaned all the old paste using isopropyl alcohol.
Living in Poland - I can order from Mouser.com even though they are in Texas. Free delivery and duty paid for if order is more than 50 or 100 dollars but there are so many places that I can order from. That's all good.
8) Agreed!
9) Hopefully I will need a bulb dim tester, that would mean some progress has been made.
post #33? - Are you sure it was post #33? I can't find anything in the text that suggest you need pics.
I am guessing it was post #31?
R456 - discolored - actual value = 2.879K, expected value = 1K

R454 - cracked - actual value = OL

C422 - blown

R307 - Another sign that someone has been inside this amp, see the ink on the PCB under the resistor? Actual and Expected Value = 56K
Thank you so much for your guidance and patience. I can't thank you enough.
I hope this reply does give you some confidence in that I have not just dived in from zero into electronics, granted I have a wealth of knowledge to absorb so in truth I have only hatched out of the egg. I would like to buy an ESR meter to really have confidence in those caps that otherwise measure fine. I would also like to check those few remaining resistors that in circuit, probably for good reason, gave me unexpected values. I am also a bit concerned that some of the smaller transistors and diodes should have their values measured. I can't do much beyond checking the state of components one by one by looking at their actual and expected values. I know that it is incredibly time consuming and doesn't really reflect any troubleshooting methodology but going this caveman approach might help me at least understand that the components in place are fit for purpose. I know that desoldering/solder on an aged PCB does have its risks. I am no where being able to perform some of the magic I see out there, for example xraytonyb
. I just have a lot to learn. I don't want to prematurely turn on this amp until I have looked at everything I can with what I know and as I learn. I appreciate that you've been there.
Cheers!
