NAD 3020i Hum

[otherside]

Hey guys,

Besides trying to build my own chipamp i have a NAD 3020i, i'm quite sure many of you know that amp it's a very common model. It's quite an old one around 15 years of age. It was on storage for a while and about 3 years ago i started using it again. It's no fantastic sound but also not bad.

Since the "revival" of it it has a constant hum on the speakers, this is quite static, it doesn't change with volume change or input, it's always there, not extremely loud but enough to annoy.

Today i got an oscilloscope from work to check out the frequency and shape of this noise. Frequency is around 100-110Hz and shape of a sine (well not exactly but close) amplitude is around 60mV p-p with another 60mV DC offset. As i suspected it's possibly from the psu since fully rectified 50Hz beomes double if not smoothed correctly (very very ballpark). I tried changing the large caps which are 4700u/50V with some brand new ones 10000u/63V and not much changed. I also tried bypassing them with smaller values (1u, 10u, 1n, 100n) but nothing changed.

Any ideas how can i reduce this huuuum?

If you need more details let me know.

(As far as i can tell the output stage is a pair of MJ2955/MJ3055)

Thank you very much

 

[Gold_xyz]

if you have already replaced all the cap on psu you can be tried
to short-circuit the inputs and to see from where noise produces him.

and then with the oscilloscope to go up again step by step to
the input that produces the noise...

also check all ground connection points.


(sorry for my terrific english... )


bye

 

[ash_dac]

Hi,

I had a nad 3130 before I gutted it. These things really don't last.

1. dry solder joints absolutely everywhere on mine

2. corroded rca jacks, and rusty transformer.

3. bulging capacitors everywhere!

Have you tried connecting source equipment to different rca sockets ?

I would look for bulging caps.

 

[AKN]

otherside,

How is things progressing for you?

You may check +43V and -28V voltages that comes from regulators built around Q507 and Q508. These regulators have feed from separate winding on transformer and powers pre amp and voltage amplifying stages in power amp.
Caps C516 and C517 are smoothing caps after rectifier and should be candidates to check.

 

[otherside]

Thanks a lot guys for the replies

I'll look into it on more detail,

does anyone know if/where i can find the schematics of this model ?

I tried to follow the noise and see where it's generated, it quite a wild guess without the schematics but from what i can see it's not comming from the pre-amp section, most likely the decoupling near to the output stage is damaged or just the components past their lifetime. Anyway a schematics or component ref would help a lot.

 

[AKN]

Mail me and I send schematic in return.

 

[Gold_xyz]

i have found a schematics here

 

[otherside]

thanks a lot guys, the schematics help a lot, i'll let you know how it goes.

 

[renevoorburg]

I recently bought a NAD 3020i that has a slight 100Hz hum, even after a full recap.

It appears on both channels, is not affected by the volume control but is by the bass tone control. I estimated its orgin might be located in the part of the regulator board that delivers the +28V and the -28V to the preamp .

So I did some poking around with an oscilloscope and found that the -28V output is unstable. A 100 Hz sine can be seen, measuring 40mV top to bottom. The other two outputs seem to be stable. Also, in stead of -28V it delivers -29.5 V.

I did replace the zener and replaced C526 with a 100uF/35V capacitor but with no impact. I expect Q508 might need to be replaced. The circuit board is quite darkened around it (but perhaps the previous owner did a replace?) and I guess it plays a central role in the circuit. Any advice, hints or tips much appreciated.


Thanks,
René - btw my first post -

 

[Mooly]

The slightly incorrect output voltage is no concern and is just down to inexact voltage references being used (tolerances) of the zener and LED.

Its very hard to make a judgement call on the ripple. Firstly you need to measuring from the correct ground point which really means measuring across C526 directly. Be sure the scope, if mains powered, does not introduce a ground loop. You should see a clean trace if you touch probe tip to probe ground whilst having probe ground connected to C526. Remember, positive end of C526 is ground.

Q508 is a series pass regulator and any common PNP device of suitable ratings should be fine, a BD140 comes to mind. Normally transistors should not be replaced without good reason, however these are a type known to fail in peculiar ways, particularly when they run hot... so yes, I would replace this and the NPN one as well (BD139 for that). Having said that I actually doubt these are causing the ripple but I would still replace them anyway.

Edit... and I assume 'full recap' means all the caps around on that diagram.

 

[renevoorburg]

Its very hard to make a judgement call on the ripple. Firstly you need to measuring from the correct ground point which really means measuring across C526 directly. Be sure the scope, if mains powered, does not introduce a ground loop. You should see a clean trace if you touch probe tip to probe ground whilst having probe ground connected to C526. Remember, positive end of C526 is ground.

I used the same ground for measuring the +28V output, the -28V and the +43V output. Only the -28V showed this ripple, for the others it was negligible.

Edit... and I assume 'full recap' means all the caps around on that diagram.

Those indeed, and all others .

 

[renevoorburg]

I have some SB 631 PNP transitors available here. I assume they will work in stead of a SB 649A. The latter has somewhat higher voltage and current ratings (-180V C-B, -1.5 A vs -100V, -1 A) though.
If Q508 has gone open Emitter to Base that would explain the -29.5 V I am measuring. To be continued tomorrow...

 

[Mooly]

Those transistors sound as though they will be fine. I'd be surprised at this point if they fix the ripple but it has to be tried.

 

[renevoorburg]

Those transistors sound as though they will be fine. I'd be surprised at this point if they fix the ripple but it has to be tried.

I changed Q508 but no impact indeed. Well, at least it provided a good opportunity to attach it to a bigger heat sink (see image).

Further, I experimented a little by doubling the capacity of both C526 and C523 but with no significant impact. Perhaps I should try to use bigger capacitors for both C517 and C516. These are now 330 uF but in older versions of the 3020 they used to be 470 uF.

Thanks! René

 

[Mooly]

I kind of suspected as much. What is the voltage on the collector of that transistor? Is there enough headroom for it to operate correctly.

 

[renevoorburg]

What is the voltage on the collector of that transistor? Is there enough headroom for it to operate correctly.

Will check that next time I open it up (for now it doing service again, swapped it for the Rotel RA-840BX4). In the mean time I found this posting: https://www.diyaudio.com/forums/solid-state/277555-nad-3020-recap.html#post5398713 and to quote from that

Just finished replacing the caps in my 3020B.
...
I stuck to the original value, except for the caps around the preamp regulator - here I went 1000µF 50V for C905 & C906 (originally 330µF 50V), 220µF 50V for C907 & C908 (originally 47µF 50V), 22µF 50V for C911 & C912 (originally 4.7µF 50V), and 220µF 50V for C913 & C914 (originally 47µF 50V).
...
The result is pretty good. I'd assumed the slight hum this amp always had was normal - turns out it isn't - it's dead quiet.

which at least suggests that bigger sized caps C516 and C517 (the model i equivalent of the C905 and C906 in the B, it seems) might help to remove to slight hum it has.

 

[Mooly]

Interesting. Its a quick and easy thing to try. Good luck and lets hope its a quick fix.

 

[Ian Finch]

As the -0.6V reference for the regulator is derived from its output, I'd take a look at what is happening around the zener and LED. I have no idea what their lifetime is but I know of another case where replacing them in a similar NAD 3020 model was the most likely reason that sound improved, according to the owner.

 

[renevoorburg]

As the -0.6V reference for the regulator is derived from its output, I'd take a look at what is happening around the zener and LED. I have no idea what their lifetime is but I know of another case where replacing them in a similar NAD 3020 model was the most likely reason that sound improved, according to the owner.

To be sure, I replaced that zener, but without impact. About to order bigger caps now.

 

[ubergeeknz]

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.