Amplifier hum

[timelessbeing]

Hey guys.

I've got a little 30W stereo for my office desk. It's not exactly hii-fi, but I like it because the main unit and speakers have a very slim profile, which allow me to put them right against the wall and use very little precious desktop real estate. Anyway, I need help because it produces a prominent mains hum. It's loud enough to be annoying when there's no music playing. I know it's not being coupled in through the aux input, from my laptop for example, because it's still there when I unplug it. It is present in all modes (tuner, tape, aux) however in CD mode it is very faint. The volume setting has no effect on it.

Are there any hacks for making the 60Hz hum go away?

[timelessbeing]

Here are some photos of the guts. The shell says it was manufactured Feb 2005, but from the inside it looks like it could have come from the 80's.

[timelessbeing]

The speaker terminals are mounted to the board with the big heat sink, so that must be the amplifier module. Incidentally the mains transformer outputs go directly into that board too, so perhaps it's doing double duty as the power supply as well (poor design decision?).

Maybe if I measured the main rail voltage, I could rip out the transformer (there is only one tap on the secondary) and use an external switcher power brick to feed it the required DC?

[cvanc]

Wow, that thing claims a 2005 build date?  You're right - it looks thirty years older inside.

What is the model number?  Low chance of finding schematics online for Jensen stuff but worth a try.  Almost certainly one (or more) (or all!!) electrolytic caps on the power supply/amp board are failed...

What test equipment do you have?  How far down the rabbit hole do you feel like going here?

[BMack]

Change the filter caps on the output portion of the board. You won't find a schematic on that. The majority of the caps will be fine.

Easy way to do it is to trace your speaker out back to the output transistor(s) then change the caps in that portion of the circuit. I'm surprised you aren't having tact switch issues.

[dan3460]

+1 on the caps.
Most of this type of this cheap radios, also apply to tvs, were made using dirt cheap Chinese components. In the close up of the power supply looks like the top right hand side cap is melted (maybe). You may spend more time and money repairing this thing that what the radio is worth.
 

[dcac]

Are there any hacks for making the 60Hz hum go away?

Are you sure it’s 60Hz hum and not rectified 120Hz ripple, 60Hz would not be very audible with small speakers.

But either way you could always check if any cables run too close to the transformer. Did the hum i.e. change when you run the amplifier with the backpanel open.

[cat87]

Try bodging a mains filter on the thing. Something like a common mode choke and some X-rated caps across the lines.

If that doesn't do it, the culprit can be either the hum is being coupled somehow to the  rails of the amplifier (the suggested re-caping might do it) OR bad design in the stage before the main amplifier (if it has something like that) With that monolithic am there's not much you can do...

[CJay]

Could also be mechanical hum, I've a DAB radio which has an audible hum in standby until you pick it up or press on the 'right' bit of the cabinet.


But, I'd replace the caps anyway, they look cheap and nasty.

[mzacharias]

Just about has to be a bad filter cap associated with the bridge rectifier or perhaps with a regulator transistor. Recommend replacing all such caps seen along the top right in pic 2, and especially the larger one seen partially enclosed by that heat sink.

[timelessbeing]

I got it out onto the bench today. No caps are bulged or leaking. I probed the rectifier output with my scope and it's pretty clean. I hooked up a speaker to examine the output on the scope, and sure enough the hum was gone. I leaned the stereo in closer to the transformer (which was loose on the bench at this point) and the hum came back.

I think I'll go ahead with my idea of hooking up an external power supply.

[timelessbeing]

  No good.

I finally got around to modding the stereo. I hooked up a DC laptop supply, and now the hum is REALLY LOUD.

I don't get it though. I tested the supply and there was very little ripple. Maybe a few hundred mV under load. Certainly not significantly more than what the old transformer had.

I'll try adding some capacitance and see if that helps.

[FlyingHacker]

Is the hum dependent upon the volume control setting? Or is it a constant level regardless of volume?

[timelessbeing]

The hum is not affected by the volume dial.

[DaJMasta]

How is the pitch of the hum tone (usually right at 120Hz in my experience for mains hum with 60Hz cycles)?  Higher frequency noise would probably point to something else.  If you can pitch match (singing or whatnot), just put on a tone generator and give it a listen, you should at least to get an idea of what frequency it's at.  120Hz is low, think lower end of an average male singing voice range.  60Hz can only really be reached by very low voices (somewhere around a B natural an octave under the bass clef staff, if you're familiar with music), so it's often easy to rule it out unless it sounds really low.


Can you see the hum at the speaker and can you trace it back through the signal path?

It could be that a dried up cap is making something oscillate which is coupling into a signal line... the chassis looks metal so it's unlikely that it's just outside interferance... but since you said moving the speaker affected it, it could be something radiating inside the case.  This again could be a dried out cap or something, but maybe test a layer of shielding between the speaker and the rest of the unit and get it close like before.  If that bit of shielding reduces the hum, it may be a viable solution, or at least may point towards the component that is introducing the noise.

[timelessbeing]

It's still 60Hz.

I never mentioned speakers. I said moving the transformer got rid of the hum, which is why I got rid of the transformer.

[FlyingHacker]

No speakers? Are we talking about physical vibration hum? That sounds like a transformer design issue, and possibly a physical mounting issue leading to some type of resonant enclosure.

[timelessbeing]

  You guys aren't listening.

Of course the sound is hum through the speakers. I was responding to DaJMasta, who said "since you said moving the speaker affected it ...  shielding between the speaker and the rest of the unit". The speaker is far away and has nothing to do with it.

As I've said a few times now, the transformer is gone.

I'll try powering it up from my bench supply. That should tell me if the source of the buzz is the power source, or the stereo circuitry. It sounds bang on 60Hz so my money is on the supply.

[Lucian84]

I guess the transformer is poorly designed, cheap, and it vibrates on the chassis. Maybe replacing the transformer with a more powerful one and putting some silicon glue will fix the hum. It's strange that you still got the hum with the external PSU though.

[timelessbeing]

WTF.  The transformer doesn't vibrate!

[mzacharias]

Just replace the larger capacitors and you should be fine. If you have a scope and are familiar with using it to troubleshoot, observe the DC voltage at each leg of each larger cap with respect to ground (mostly the positive leads) and one of them will show a large, unfiltered ripple component. Replace that one. It's pretty obvious from looking at the pictures that the caps are "el cheapo" so it would be a good idea just to replace them all on general principles.

[dan3460]

I think we all are tuning to different satiations. I'm a little confused. Can we reset?
- Tell us what is the setup that you have right now. Seems to me that you have bypassed the power supply and powering directly from a bench power supply.
- What kind of instrumentation do you have, can you determine the frequency of the hum? If you have a smart phone there are some apps that can pick up a sound and tell you the frequency. This is crucial to determine if the hum is power related of may be some kind of feedback.

[timelessbeing]

Sure. You can see a photo of the stereo in my first post. That's the main unit. The speakers are separate and connect via long leads to terminals on the side.

As I said earlier, I opened up the main unit to investigate. I found a mains transformer mounted to the inside of the case, and fairly close to the amplifier board. I found that the hum would disappear when the mains transformer sat on my bench, away from everything else. Thus, I concluded that the audio signal was picking up interference from the transformer, so I removed it completely.

The setup now:
I am powering it with a cheap chinese laptop power supply (20V, 90W), which I have connected where the transformer used to be, ie rectifier input. (I put it here to utilise the diode drop to bring the voltage down to what I measured on the rectifier output in the factory setup, around 19V).

[timelessbeing]

Here is some info on the buzz right now.

I am attaching an MP3 file of the sound itself, and photo of the signal on my CRO.

Vertical scale is 10mV/div. Timescale is 1ms/div making the period roughly 8ms, which equates to a frequency of 125Hz (probably actually 120). Though, If I listen to a generated 120Hz tone it sounds higher to me, but that could be the distortion in the waveform throwing it off.

[timelessbeing]

I'm a lot closer to the source of the problem now. I was able to get rid of the buzz completely. Here's what further investigation revealed... It gets much weirder!

I think it's some kind of ground loop effect. It only happens a bunch of conditions are met. Sorry, it gets a bit complicated, but here goes...

Condition #1:
Stereo is plugged into a certain outlet. If I move it to a far away outlet, buzz disappears.

Condition #2:
I have a laptop, which I connect via DisplayPort to a monitor. Monitor takes digital audio signal from DP and converts to analog audio for output on a 3.5mm jack. When I connect this analog audio to the AUX input on the stereo (+Condition#1 is met) then I get buzz. Even if I connect the laptop headphone audio to the AUX input there is buzz, but only while the external monitor is connected. Once you take the monitor out of the equation, laptop headphone audio is fine. For any other isolated audio device (ie smartphone), the audio is fine.

Since this is a cheap chinese power supply, it may have some poor grounding design. In the stereo, it's a cheap 1-sided phenolic PCB, but the traces did look like they used star grounding. I might try probing around with a grounded lead to see if that makes any difference.