Looking for advice on mechanical hum

[mansr]

28 minutes ago, rayl1234 said:

Yes, but that is not the definition, which is stated further down:

 

"The average value of a whole sinusoidal waveform over one complete cycle is zero as the two halves cancel each other out, so the average value is taken over half a cycle"

The average over half a period is uninteresting in this context (DC offset), so that particular definition is irrelevant. Either way, it too doesn't refer to an external earth reference.

 

Earlier, I said this:

7 hours ago, mansr said:

An AC voltage between two terminals has a DC offset if the arithmetic mean over one period is non-zero.

 

[mansr]

44 minutes ago, rayl1234 said:

PS: To be clear, I did not want to derail the thread or challenge whether DC offset is or is not the issue.  Only to say that a DMM's reading of average voltage, if it offers such, is not the right number. Need a meter that reads out DC offset, which Fluke true RMS DMMs can do.....

Most multimeters will display the DC offset if simply set to measure DC voltage. I tested it myself before posting here using both a Fluke 289 and a cheap, nasty $10 no-name meter. Neither measures the half-cycle average you're going on about (because it's hardly ever interesting).

[Don Hills]

12 hours ago, extracampine said:

Don - thanks for the specific suggestion  Yes I will look into that. Won't such a transformer dampen dynamics and limit peak draw? ...

 

Only if it's too small. Isolation  transformers are/were commonly used on building sites with power tools. The most common sizes are in the 1 to 3 kilowatt range. This is overkill for most home hifi systems. Anyway, for the test you want to do, you only need a few dozen watts to run your CD player.  

 

I think I'm showing my age... it appears that RCDs have largely replaced isolation transformers on building sites. As such, builder's merchants don't appear to stock them any more. You should be able to find one quite cheaply on the used market. If you know any builders, ask them. Otherwise, carry on trying the CD player at other locations. You could also call your local electricity supplier and say you suspect DC on the supply. They should be interested in checking, as it will be reducing the efficiency of their own transformers.

[rayl1234]

I am sorry if offense is taken from my remarks.  I only mean to correct inaccuracies.

 

1. It is absolutely correct that DC offset is not measured relative to earth. That was someone else's statement that I was trying to clarify what I think he meant. Maybe that just added more confusion.

 

2. DC offset is the LPF of the AC signal.  (Though for higher power loads, Hall effect is used instead of trying to build an LPF.) Thus any reference to average voltage, or zero, etc., is not really the best way of interpreting it.  If my being a stickler for precise definitions for engineering terms is offensive, then let me just retract the comment.

 

3. Not all DMMs will correctly measure DC offset if you just set it to DC mode.  You see crazy stuff like the OP in this: https://arstechnica.com/civis/viewtopic.php?f=11&t=100759  No, he did not really have a 20V DC offset.... Thus, I would only trust a DMM that is documented to measure it.  That's perhaps the non-academic comment I wish to make.

 

 

[mansr]

18 minutes ago, rayl1234 said:

I am sorry if offense is taken from my remarks.  I only mean to correct inaccuracies.

 

1. It is absolutely correct that DC offset is not measured relative to earth. That was someone else's statement that I was trying to clarify what I think he meant. Maybe that just added more confusion.

 

2. DC offset is the LPF of the AC signal.  (Though for higher power loads, Hall effect is used instead of trying to build an LPF.)

DC offset is the, uhm, DC component, also known as the average. Yes, a low-pass filter with a cutoff of a few Hz is one way of finding it.

 

18 minutes ago, rayl1234 said:

Thus any reference to average voltage, or zero, etc., is not really the best way of interpreting it.  If my being a stickler for precise definitions for engineering terms is offensive, then let me just retract the comment.

The standard definition of average is the integral divided by the range. Are you using some other meaning of the word?

 

18 minutes ago, rayl1234 said:

3. Not all DMMs will correctly measure DC offset if you just set it to DC mode.  You see crazy stuff like the OP in this: https://arstechnica.com/civis/viewtopic.php?f=11&t=100759  No, he did not really have a 20V DC offset....

Auto-ranging (or picking a bad range manually) can mess things up if the DC component is small compared to the AC voltage. That might be what happened there.

 

18 minutes ago, rayl1234 said:

Thus, I would only trust a DMM that is documented to measure it.  That's perhaps the non-academic comment I wish to make.

Using documented features is always best. The Fluke 87 manual suggests the method I mentioned earlier. There are of course other decent makes of multimeters, but I'm most familiar with Fluke.

[rayl1234]

4 hours ago, mansr said:

... portions deleted...

 

Using documented features is always best. The Fluke 87 manual suggests the method I mentioned earlier. There are of course other decent makes of multimeters, but I'm most familiar with Fluke.

 

I will agree that your use of the term "average" is of the typical mathematical text and I overreacted bec I read it in the AC electrical context, where it is 0.637 x peak voltage (hence will never be 0 for mains voltage).

 

However, I still urge caution on assuming the DC measurement from a DMM correctly measures the DC offset w/o your own LPF or a DMM explicitly engineered with an LPF.  Here is an article from Fluke urging caution on measuring DC when large AC component is present, which is the relevant use case for this thread.

 

http://download.flukecal.com/pub/literature/9080007_a_w.pdf

 

The "spurious offsets" referred to are what (in my opinion) causes people to erroneously read large DC values in some cases where there really isn't much DC offset. 

[mansr]

8 hours ago, rayl1234 said:

I will agree that your use of the term "average" is of the typical mathematical text and I overreacted bec I read it in the AC electrical context, where it is 0.637 x peak voltage (hence will never be 0 for mains voltage).

That would be the average absolute voltage, which is but one of many averages one might compute.

 

8 hours ago, rayl1234 said:

However, I still urge caution on assuming the DC measurement from a DMM correctly measures the DC offset w/o your own LPF or a DMM explicitly engineered with an LPF.  Here is an article from Fluke urging caution on measuring DC when large AC component is present, which is the relevant use case for this thread.

 

http://download.flukecal.com/pub/literature/9080007_a_w.pdf

Interesting, thanks. Do note that this paper is mainly about precision measurements in the nanovolt range. Some of the effects discussed are not applicable here.

 

[extracampine]

Thanks for the ongoing insights here - though some of the technical discussion is a little above my head

 

I have 4 further small findings that might help in the quest to remove the hum;

 

1. I took the case off (again) to see if the rod that holds the transformer needs tightening or loosening. What I found before getting to that however, was that the casing seems to be contributing to the noise. With the case removed, the noise is definitely less. It is still audible, though perhaps heading towards the level that would not be noticeable from the listening spot. I had tried removing the case before and don't remember a particular reduction in noise, but there you go. So I got some thin rubber pads and tried to insert them in-between the front panel and side casing. Once the screws were again tightened, this did not seem to improve the noise. Flexing or moving the top casing also seems to affect the noise somewhat, though inserting anything in-between this and the base is quite tricky. Either way I did not have the impression that inserting further bits of rubber would likely affect the noise, given my findings with the rubber at the front.

 

2. Trying again to isolate the area that the hum is coming from, I think it is coming from a different component to what I had expected. I had thought that the noise was coming from the main transformer (a big round thing which I have labelled 'A'), though it appears to be coming from somewhere nearby. The exact culprit is difficult to determine, though I'm wondering if it is coming from the component highlighted in my photo below (which I have labelled 'B'). I'm not sure what this component is but am wondering if it is also a kind of transformer?

 

 

3. I noticed that one of the bolts/pillars holding the PCB to the casing appears to have current running through it (if that is the right term) - when I touched it, I felt a small shock. I can't remember if my other hand was touching the casing or not. I also don't know if this is normal/expected or not. The unit was powered on at the time. Maybe I was stupid to be prodding around inside with the power on. I don't know if the other 4 bolts are the same (I didn't fancy testing)! I have highlighted this bolt in the picture below also.

 

 

4. I took the DAC to my workplace and plugged it in there. Same hum. It is however only about 3 miles away as the crow flies - don't know if this is relevant or not.

 

Thanks again for the help here.

 

[bobfa]

OK I am just sort of observing here.

 

When you are touching the outside of the case of the DAC can you feel the vibration?  I would assume that you could if you can hear it from a distance.  If the vibration is coming from multiple devices I would be worried about an over voltage  condition of some sort.  I am a bit "concerned" that multiple devices are exhibiting this vibration.

 

Your testing at your office should be enough to eliminate the possibility that the problem is the line transformer feeding your home and or local over voltage issues inside the home.

 

From looking at the pictures above:  

 

1.  I assume from the pictures  that this is a device with Tubes in it that means that there is high voltage stuff floating around in there.  Not a place to play around if you are not trained!

2.  If you got a "shock" from touching that standoff that appears to be grounded to the case that is a bit weird. (could it have been static electricity)

3.  Your item "B" is a transformer or a choke.

4. I cannot tell if the "B" transformer is actually bolted to the case or not, but there is one mounting leg showing that does not appear to be screwed down.

 

If the vibration is from the "B" transformer it could be vibrating against the case.  You could use some sort of hard plastic rod or the like and apply pressure to the transformer to see if the sound changes.  I would discuss this with the manufacturer.

 

Note that when I have found AC vibration in older tube type two-way radio gear it was bad transformer laminations like you can see in your item "B" where the lacquer had failed. But that stuff was decades OLD!  

 

I know this may not be of much help, but please be careful in there you could get hurt or worse.  Try feeling for physical vibrations on the equipment.  I just cannot imagine three devices with the same problem .

[extracampine]

Thanks for this reply Bob. 

 

I don't think I can feel a vibration when touching the case. Although I say that I can hear the hum/buzz from a distance, it is still relatively quiet. Regarding the over-voltage, I am going to use a multimeter to check the mains voltage in the house in the next day or two. 

 

You are correct re the tubes - 1 rectifier tube and 4 DHP tubes. I will make sure it is off in future when poking around inside  I doubt that it could have been static - the shock was a little too pronounced, and it happened twice. 

 

Item B (the transformer/choke) is bolted to the base of the case via 2 of the 4 available bolt points. I will look further into this with a plastic rod as you suggest.

 

Yes, the fact that my Classe amp hums (though to a lesser extent) and my linear PSU (now to a lesser extent also - I think it used to be worse) makes me wonder about my mains. I still plan to measure this as mentioned above, and try the DAC a bit further afield.

[extracampine]

OK, further to the above - I tested the mains voltage at an outlet in my house with a multimeter and it was reading both 245 and 246 volts. 

[One and a half]

And the rest of the measurements between earth neutral and active are.....

[mansr]

1 hour ago, extracampine said:

OK, further to the above - I tested the mains voltage at an outlet in my house with a multimeter and it was reading both 245 and 246 volts. 

That's normal for Britain.

[extracampine]

Yes, I gathered that this was normal and within the acceptable range as far as the transformers are concerned. One and a half - what other measurements would be helpful here?

[One and a half]

16 hours ago, extracampine said:

Yes, I gathered that this was normal and within the acceptable range as far as the transformers are concerned. One and a half - what other measurements would be helpful here?

Please refer to this post on the first page.

[extracampine]

OK One and a half - I looked back at the post you mentioned. Here are my findings:

 

Black probe to earth, red probe to active: 246V

Black probe to earth, red probe to neutral: nothing registered on my multimeter - 0.00V (however the lowest "setting" for AC V is 200V)

Black probe to neutral, red probe to active: 246-247V

 

You initially said that it should read 230V. Does this therefore seem quite high?

 

[mansr]

20 minutes ago, extracampine said:

OK One and a half - I looked back at the post you mentioned. Here are my findings:

 

Black probe to earth, red probe to active: 246V

Black probe to earth, red probe to neutral: nothing registered on my multimeter - 0.00V (however the lowest "setting" for AC V is 200V)

Black probe to neutral, red probe to active: 246-247V

 

You initially said that it should read 230V. Does this therefore seem quite high?

230 V is normal for continental Europe. Nominal voltage in Britain is 240 V, and readings of 250 V are not uncommon.

 

If your multimeter has a DC range of 400 V or higher, you can try checking for DC offset using this. Some meter designs can apparently give false readings in this situation, but it could still provide some information.

 

As always, be extremely careful when handling mains power.

[extracampine]

I have the Sinometer 830B - this one:

 

http://jbryant.eu/pages/DMM.htm

 

I think it has a DC range of 600V, to the left of "OFF" if I am not mistaken?

[mansr]

9 minutes ago, extracampine said:

I have the Sinometer 830B - this one:

 

http://jbryant.eu/pages/DMM.htm

 

I think it has a DC range of 600V, to the left of "OFF" if I am not mistaken?

That is correct.

 

A lot of cheap multimeters have exactly those settings. I suspect they all use the same circuit. I tested one, and it measures DC offset correctly, though that's of course no guarantee of anything.

[One and a half]

1 hour ago, extracampine said:

OK One and a half - I looked back at the post you mentioned. Here are my findings:

 

Black probe to earth, red probe to active: 246V

Black probe to earth, red probe to neutral: nothing registered on my multimeter - 0.00V (however the lowest "setting" for AC V is 200V)

Black probe to neutral, red probe to active: 246-247V

 

You initially said that it should read 230V. Does this therefore seem quite high?

 

The measurements tell me that the plug is oriented correctly. I don't believe the 0.00V reading, the meter is not capable of auto ranging and I'm not going to suggest lowering the meter scale in case you place the leads on the wrong terminals and carbonise the meter and your fingers.

 

I didn't know where you were located, not a mind reader ya know, UK is nominally 240V, 247V is only 2%, 10% on top of 240V is the limit, but transformers have a limit that's not that high. They tolerate it, but complain.  

 

This DC offset conversation is pure speculation. Unless you can organise someone who has a Powerscope and actually takes a measurement of the waveform and an FFT, there will be guesswork to what the remedy is . Any sharp rises in voltage, or aberrations on the waveform can be readily determined. If you have persistent headaches, the doctor is consulted and by process of elimination and their experience, often the remedy works, same situation here.  

 

Your local power company can be contacted, the ones that provide the stuff in the first place, they have engineers that specifically determine root causes of power problems, mention to them that the equipment doesn't hum in other places, but only at home. Perhaps the problem is on the lines in the street, then they would be interested. No need for a long term logger to be installed, a spot measurement is more than enough. 

[extracampine]

I tried measuring the mains DC with the meter set to a range of 600V DC, but it didn't measure anything. 

[One and a half]

On 1/21/2018 at 10:34 PM, extracampine said:

I tried measuring the mains DC with the meter set to a range of 600V DC, but it didn't measure anything. 

I'm sorry I can't give you any more advice without compromising your safety, cheers,

[chetthejet]

On ‎1‎/‎5‎/‎2018 at 5:35 AM, extracampine said:

Thanks for these replies.

 

 

I agree that it is not likely a signal problem. I do not have a power scope, and as sandyk mentions, I fear that doing this would be beyond my level of expertise. I also do not have a UPS.

 

 

It has been suggested to me also that it may well be DC on the mains, or DC offset. I have however tried at separate times 2 DC blockers, both feeding the power block and direct to the DAC - though neither made any difference in either configuration. Have you tried a DC blocker yourself?

 

 

The same noise persists with the lid completely removed.

 

 

Well, my understanding is that the sockets that my audio equipment is connected to is part of a ring circuit that does not include the large appliances - though I could be wrong. We do not have any dimmer switches or fluorescent bulbs. The noise was present in the middle of the night also. 

 

I will take the DAC to my work (which is around 4 miles away) and try there to see if a different power supply affects anything, and will report back.

 

I had this same problem with very audible hum emanating from my amplifier transformer.  Completely vexed me until I started unplugging every other thing that was on the same circuit.  The last thing I unplugged was a small table lamp (incandescent)....and, voila, that turned out to be the culprit.  Didn't matter whether the lamp was on or off, the hum was there as soon as the amp powered up.....but once that lamp was unplugged from the wall socket (and chucked in the dumpster!)  the hum disappeared permanently.  

[extracampine]

9 hours ago, chetthejet said:

Didn't matter whether the lamp was on or off, the hum was there as soon as the amp powered up.....but once that lamp was unplugged from the wall socket (and chucked in the dumpster!)  the hum disappeared permanently.  

 

Interesting! Did you try plugging the amp in somewhere else, other than at your house? I tried it at my work too and the hum remained the same - thus reducing the possibility of something like what you had experienced, I think. 

[chetthejet]

8 hours ago, extracampine said:

 

Interesting! Did you try plugging the amp in somewhere else, other than at your house? I tried it at my work too and the hum remained the same - thus reducing the possibility of something like what you had experienced, I think. 

Never did get that far, though it would've been an excellent experiment.  Once I found the culprit, that little search and destroy mission came to an abrupt, but satisfying, end.