NAD 3020i Hum

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Before you start changing things, I would look to see if anything is drawing excessive current off that rail. Maybe a component downstream of it is bad causing excess draw, and therefore higher than normal ripple. Just a thought. I would start with any tantalums or electrolytics going from that rail to ground.

Thanks. A sound advice, I guess. I don't seem to be the only one who has a Q508 that runs very hot, coloring the circuit board around it, but that doesn't mean nothing is wrong there, perhaps something is straining it too much, with that ripple as a side effect.

I am not that experienced yet in fixing amps. Do you have any tips in how to proceed with that? Is the easiest way by checking the dc resistance of those caps connection the -28V rail and the ground? I don't have a meter to check caps.

Thanks, René
 
I just replaced the regulator filter caps C516 and C517. Here is what I measured at the -28V (? -29 V) output:

330uF -> 38 mV ripple (before)
680uF -> 20 mV ripple (first attempt, not yet satisfied)
1000uF -> 15 mV ripple (a bit of a push to get the wires through ...)

I did expect the bigger caps to have more effect. Listening to the result, the level of hum is more or less half of what is was (with the 330uF caps). Nearly negligible, but what is itching is that I don't get what causes it. Indeed, I will try to hunt that down a next time.

Cheers, René
 
Last edited:
Administrator
Joined 2007
Paid Member
You may be trying to improve the performance beyond what the basic design is capable of. Many commercial amps have slight hum/buzz due to constraints in the layout and design. 'Nearly negligible' is probably normal tbh.

I just wonder now whether the ripple you are seeing isn't in fact ripple due to a problem with the regulators but simply an effect of the layout and wiring and circulating currents.
 
I had all models of 3020 and several times each model.
what i can tell you is that i had several amps with buzz and others not.
I have recapped a dozen with or without effects on the buzz.
I tried to search on several devices and I ended up dropping.
I still have a 3020 "serie 20" full of origin that is silent as a grave and a 3020 A fully recapped that is noisy.
I would be happy if you found the "why of how"
 
After tripling the capacity of the filter caps of the regulator (330 uF to 1000 uF, see a previous post) I doubled the capacity of all other electrolytic capacitors in the regulator. The ripple is now reduced to just 5 mV peak to peak. I think I'll consider it fixed, more or less, since probably I just lessened the symptoms, not tackled the root cause (?).

To sum things up: the light 100 Hz hum I could hear over the speakers was traceable in the -28 V output of the regulator, visible on the scope as a sine wave.

original situation -> 38mV peak to peak ripple
filter caps to 680 uF -> 20 mV peak to peak
filter caps to 1000 uF -> 15 mV ripple
doubled other regulator caps -> 5 mV ripple.

Cheers, René
 
Administrator
Joined 2007
Paid Member
Pleased to hear you have lessened the hum.

I doubt the transformer buzz is anything new. You'll probably find it a bit variable and better at some times than others due to purity of the mains (no harmonics) and perhaps even if the mains is a little low in value (less chance of saturation).
 
My 3020 has a buzz coming from the PSU transformer, it is heard approaching the ear to it, but it is not amplified by the circuit that remains silent as the first day.
By the way, the only problem I had was a continuous and very high hum after a "plop" very pronounced when turned on, so much so that I thought it would ruin a speaker.
The change of the diode bridge and the filter capacitors solved everything.
But the transformer's physical buzz is still there.


NAD 3020 amp upgrades?
 
Thinking about the possible root cause of the hum I am beginning to suspect the mute-on-startup circuit. That is comnnected to the -29V part of the regulator, the part that has the ripple on it. Further, I've learned that adding capacity / capacitors towards the input of the regulator seems to have more impact and that indeed is where the mute circuit connects. Moreover, the circuit board and components in that area look like they have experienced some stress.

In the schematic I used there is a Q516 tranistor which I don't understand. It has it base connected over a 47K resistor to the -50V lead of the regulator:

629474d1502234164-nad-3020i-thermistor-seen-nad-3020i-muting-detail-png



However, I just found a different schematic, supposedly also of the 3020i. There Q516 is also connected to a transistor Q517 and a thermistor, which does seem to make some sense:

629447d1502220290-nad-3020i-thermistor-seen-nad-jpg


So as a next thing I'll check which schematic my 3020i follows. I will check the state of Q516 and D601 (latter schematic). Further, I will just disconnect the leads from the regulator to the mute circuit (should not be a problem I gather) and test if that effect the ripple. I think I am getting close now.
 
Last edited:
In the face of a persistent hum that is weakened but not eliminated with new filter caps and increased filter capacitance, academia50's suggestion that the diode bridge may be going bad seems as though it might have some merit.

...By the way, the only problem I had was a continuous and very high hum after a "plop" very pronounced when turned on, so much so that I thought it would ruin a speaker.
The change of the diode bridge and the filter capacitors solved everything...

NAD 3020 amp upgrades?
 
In the face of a persistent hum that is weakened but not eliminated with new filter caps and increased filter capacitance, academia50's suggestion that the diode bridge may be going bad seems as though it might have some merit.

Might be indeed. I just checked Q516, Q517 and D601 and they were all fine. I fully disconnected the mute-on-startup circuit (except for the FETS) and it had no effect whatsoever on the ripple.

To be continued...
 
In the face of a persistent hum that is weakened but not eliminated with new filter caps and increased filter capacitance, academia50's suggestion that the diode bridge may be going bad seems as though it might have some merit.

Regrettably, that didn't help.

If it would have helped, I would be wondering now how a broken rectifier bridge could result in a 100 Hz noise, when I think 50 Hz is to be expected. Or am I mistaken?

Think I'll move over to a next project. Thanks and see you later.
 
Regrettably, that didn't help.

If it would have helped, I would be wondering now how a broken rectifier bridge could result in a 100 Hz noise, when I think 50 Hz is to be expected. Or am I mistaken?

Think I'll move over to a next project. Thanks and see you later.



I remember that I measured the original diode bridge and one of the four was wrong. It drove in both directions - with the simple test of the tester - although the leak was very slight, it was not shorted or fully open. But it was a leak! I oversized the amperage and voltage of the new bridge, but not the Uf value or the voltage of the new electrolytic filter capacitors.
(They lower the ripple, until they reach pure DC and from that point they should be like a chemical battery, but it seems that it never becomes totally like that, I found out about it here (I think it was DF96) It is not necessary to oversize the capacitors of the PSU, the circuits of the PCB and some components are not prepared for that over current availability.
A teacher many years ago told us that the electrolytic capacitors should be replaced by the same maximum voltage, not over-dimensioning them, because this modifies the capacity too. And he had written several electronic books, so I took it literally since then.
The first time a colleague introduced me to the NAD 3020, after listening carefully to the low solids of the amplifier, he introduced me to the amplifier "without the clothes" and when I saw the size of the filter capacitors, I could not believe it. Obviously, engineers take many factors into account when designing a circuit.
The buzz that persists in my amplifier is originated by the PSU transformer, probably it was always, and it will continue there, since it is not amplified.
Good luck and I hope there is a happy ending.
 
You may be trying to improve the performance beyond what the basic design is capable of. Many commercial amps have slight hum/buzz due to constraints in the layout and design. 'Nearly negligible' is probably normal tbh.

I just wonder now whether the ripple you are seeing isn't in fact ripple due to a problem with the regulators but simply an effect of the layout and wiring and circulating currents.

I think it can be like this .....
Do not forget that the great performance of the first Nad was characterized by an excellent development in the circuit´s , and the accessibility of its components, being very easy to get / replaceable their transistors. (Line Phillips BC /BD) But its components were not up to par, which on the other hand would have increased its price much. Nothing of Alps potentiometers here!
There are no free lunches once again.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.