NAD transformer humming-buzzing/Nichicon KA series

[Timeslice]

Hi
I took out of storage my faithful old NAD 3020i only to find a hum/buzz from the transformer. The unit works and the audio channels are not affected but the buzz is loud enough to detect over low volume playback, and well, it shouldn't be there!

I'm not well informed on what the causes might be. Is it terminal? Do I need to replace or could it be alleviated down the line (caps?) I am planning on replacing all the caps anyway (pre and power amp,) which leads to my second question; does anyone have experience with the new Nichicon KA series?

Thanks in advance.

 

[Ian Finch]

Search the NAD 3020, 3020i, 3020e topics in this forum. There are dozens of threads, most centred on age issues. A buzzing transformer is not uncommon.

 

[Timeslice]

Thanks Ian, have done but not seen my particular problem. Will try again.

Anyone know about the Nichicon KA series capacitors?

 

[djoffe]

Sometimes I've seen (or heard) this when one of the diodes in the bridge blows. You can actually lose two diodes in the bridge. If they blow open, the only issue is that the output becomes half wave, rather than full wave, but that keeps nearly full voltage (especially at light loads) and kind of rattles the transformer as the ripple frequency drops in half.

The last time I saw this, the package of one of the diodes had exploded...Probably a transient had exceeded the PIV rating of the diode.

Akitika GT-101

 

[Timeslice]

Thanks, I'll look into the bridge but on quick inspection looks OK.

I just went through the NAD service manual (courtesy NAD 3020 | Owners Manual, Service Manual, Schematics, Free Download | HiFi Engine) for my model, the 3020i and saw that amongst all the electrolytic caps there are two that are listed as Ceramics - a pair of 330uF@63v (C516 and C517) Would they need replacing, and if so, I presume with ceramics?

Cheers

 

[Ian Finch]

Here is a typical thread with all the possible answers but no cigars until you get down to the end. http://www.diyaudio.com/forums/solid-state/220449-i-destroyed-my-nad-3020i-help-guys-4.html

330 uF ceramics? I don't think so - C516 at least, is polarized. Ceramics are non-polar and at the time those models were built, almost all were small, flat discs with a wire attached to each face. They look nothing like the vertical, cylindrical shape of radial leaded (to fit PCBs) electrolytics. The good news is, you only need to replace the electrolytics, because like the name implies, they are wet chemical devices which, over years, dry out and lose their capacitance so things, like power supplies then tend to deliver pretty spiky current that affects the sound and the transformer behaviour.

A thread on the 3020i about 1 year ago spent ages eliminating the wild guesses only to finish with no more options but replace the main (yes, the big ones C508,9) power supply electrolytics and lo and behold - much less buzz. I don't say that this is the only problem, but it is quite likely the major contributor. Considering this noise only became apparent after storage, the issue is deterioration and that's what electrolytics do. I hope you find this helpful, though even if it is not completely successful, the effort of recapping won't be wasted.

Modern caps are smaller than original so it won't hurt to uprate the capacitance from 6,800uF with anything up to 10,000 uF. A higher voltage is OK if necessary, too. Panasonic THSA grade is a good, economical, widely available cap. grade, as are many established brands and you can buy more expensive ones too but IMHO, there is unlikely to be any benefit in a NAD 3020 series amplifier. Check dimensions and lead spacing before buying caps.

 

[Timeslice]

Further to my last post. I'm obviously exposing myself as completely ignorant, but in the NAD service manual, under parts, does say next in the capacitor listing that two of the caps, I presumed were electrolytic because they look like the othere, are 'Cer. Cap. 330uF 63v' Does this mean ceramic? And if so why have a ceramic cap in the main amp. The circuit schematic indicates a polarized cap with those specs.

I wondered about this and looked up ceramic caps and it doesn't seem that these specs are even manufactured!

Does one generally just replace the electrolytic or all the capacitors on an old amp? Do the other cap types deteriorate as much as the electrolytics?

Thanks

 

[Ian Finch]

Is this a mix-up of schematics and parts lists with the 3225 design? Anyway, as you realize, there are no such values of ceramics. A maximum ceramic size for NAD designs would be 0.1 uF. Typically, there won't be many >1 nF. I don't know what is meant by Cer. there, though it's likely an error. What do you see written on the components? What shape/size are they?

Rather than blast away, fitting all new caps, it might be safer for a newb, to replace the prime suspect parts first then tackle the rest, one channel at a time, so that you always have a working reference amplifier in case of a problem. It also gives a clearer indication of the source of the problem and a sense of proportion for the value in doing this kind of maintenance.

 

[Timeslice]

Thanks Ian, my last message was sent before I read your reply. You make a lot of sense. The only reason I would look at replacing all the caps, well, at least buying them, is that the postage is about the same as the cost of the components! So might as well just do it once.

But I will do as you say and replace the big ones first, test and then proceed from there.

Thank you for your kind help. (Well done in the ODI's against SA!)

Regards

 

[mabu0815]

Hi,

I had the same problem (humming transformer) with my 3225PE.
I exchanged all 8 Rectifier diodes and the humming was a lot quieter.
Its still there, but you have to move the ear close to the amp when no music is playing to hear it.

 

[AndrewT]

Do not replace any of the small capacitors, nor any of the plastic film capacitors.

Build and use a Mains Bulb Tester for your next power ON.
It will protect your transformer from damage if you make a mistake installing any replacement items.

Can your digital voltmeter (DMM) measure AC in the presence of a high DC voltage?
i.e. can you measure the AC voltage on the +ve and -ve supply rails. That would give you a big clue to the condition of the main smoothing capacitors.
Some DMM cannot do this and give a silly measurement reading. On these instruments you can add a capacitor in series with one probe and that blocks the DC voltage and lets you read the remaining AC ripple voltage.
The capacitor MUST be rated higher than the supply rail voltage

 

[jaycee]

In this case I would say "Cer" probably means Elna Cerafine.

 

[Goldie99]

.....or, I suspect more likely, is just a mis-print.

 

[fabrice63]

........
Anyone know about the Nichicon KA series capacitors ?

I have used many of them to recap the préamp section of my Microméga minium amp , they "sound" as well as the Panasonic's FC , almost

.

 

[sgrossklass]

I had the same problem (humming transformer) with my 3225PE.
I exchanged all 8 Rectifier diodes and the humming was a lot quieter.
Its still there, but you have to move the ear close to the amp when no music is playing to hear it.

But none of them behaved abnormally during a simple diode test, I guess?

That's interesting. Presumably at least one of them was leaky under high reverse voltage, something that's not easy to test for these days.

If swapping out the electrolytics does it, then at least one of these is leaky enough to draw substantial current (it would also warm up).

Transformers tend not to be too much into loads that are not symmetrical, and either kind of defect could cause this.

 

[mabu0815]

I measured all 8 Diodes with my cheap Transistor/Cap/diode tester and all showed up with nearly the same measurements. So the defect must be in a way that a simple measurement cannot detect it.

 

[sgrossklass]

Capacitor testers from the vacuum tube days were often equipped with a leakage test facility which allowed subjecting a DUT to a variable amount of DC voltage (up to hundreds of volts) while monitoring current. Something like that would have been handy here.