Ground Loop(?) leads to blow speaker


Aloha 'goners!
          I'm going to do my best to get the details down here as I am a bit stumped on this one.

I have a Wadia 121 DAC that was recently repaired.  The issue was the RCA/XLR outputs produced mostly noise and little music.  The headphone side was fine.

As the unit wasn't functioning properly I hadn't had the chance to hook it up and use it.  When I got it back and hooked it up there was a nasty ground loop(?) that led to a blown speaker (Thiel 1.6).  The buzz produced was the angry bee buzz followed by a loud electrical sounding crack and the woofer was gone.  The amp (Creek Classic A53) went into protect mode and I shut it off.  I then disconnected the Wadia and hooked up a Mytek Brooklyn+ and there was no buzz.  Same cables same outlets same everything except the DAC.  I tested the amp and it seems to be ok through another pair of speakers.

At this point I wasn't sure what was going on.  I tested 3 different amps with the Wadia and 2 of the 3 buzzed.  With the Mytek none of them buzzed.  Of the 2 amps that buzzed one of them had a 2 pin connection and the other a 3 pin.  The only amp that didn't buzz was a 2 pin Adcom 535 MkII.  The other amps were a 3 pin Creek and a 2 pin Carver.

Here is the full chain:

Primare CD31 --> Madrigal AES Cable --> Wadia/Mytek DAC/preamp --> Chord Cobra Vee RCA --> Creek/Adcom/Carver amps --> Tara Labs spkr cable --> Thiel 1.6/B&W CDM1SE speakers

I used the same wall outlets for the DACs and amps.  The same interconnects were used between components as well as from the amps to speakers.

I tested the wall outlets with a cabling tester and it said they were wired correctly.

Could there be something else that I'm missing that would be causing this?  The only amp I am comfortable with using to test is the Carver as it has a variable level on the front panel.

Any help or thoughts on this would be greatly appreciated.

solobone22
I'm not an expert of any kind, but what leaps out at me is the Wadia has a bad problem that is passing a very destructive signal down the chain. I think the volume must have been turned up a great deal as it was loud enough to blow a speaker (I'm surprised it was a woofer and not a tweeter).

I'm relatively sure what you're hearing is not a ground loop, but a sound I've heard unhappy digital equipment make when something is very wrong in their circuitry. I would throw away the Wadia, or at the very least, have it repaired properly. Sometimes, though, that can happen when something is not connected properly. So, check everything you can. Good luck,
Dan
Post removed 
It’s not clear to me what you mean when you refer to a "2 pin connection" and a "3 pin connection." Initially I thought you might have meant RCA vs. XLR, but several rear panel photos I found as well as the manual for the Creek amp (which you described as "3 pin") show it to only have RCA inputs. And the cable you indicated was being used between the DAC and the amps is an RCA cable. Can you clarify that?

Also, by any chance were you using XLR-to-RCA adapters between the DAC and any of the amps?

In any event, I agree with Dan that it does not sound like the problem was caused by a ground loop.

Good luck. Regards,

-- Al



P.S: Now that I think of it I presume that "2-pin" and "3-pin" were references to the number of prongs on the power plugs of the amps. And since the problem appeared when you used the Carver which has a two-prong plug that would seem to further confirm that the problem was unrelated to a ground loop.

Regards,

-- Al
The Adcom 535 Mk II is capacitor coupled at the input and the Creek and Carver are not. What you have is DC at the output which the Creek passed through to the speaker before going into protection mode. The Adcom's input cap blocked the DC and didn't pass the buzz-causing DC to the speaker. DC is enough to blow out a woofer.

The problem in the DAC can be anywhere in the active circuit. I would present a repair bill for the Thiel to whoever worked on the Wadia.
+1 to everything gs5556 said. A ground loop would be characterized as a hum. What you say you heard was a buzz, that blew out a woofer, almost definitely DC 
Thanks Folks!

With regards to the 2 pin 3 pin I should have posted that I was referring to the power cable.

Working on measuring the output on the unit now. The place that did the repair wants the unit back however I’d like to measure it in its current state and then again after they send it back.
I have a Klein Tools MM300.  I set the meter to 200m and tested the RCA output touching the positive line inside the connector and the negative line to the outer barrel.

The left channel read 1.0.
The right channel read -0.1.

This looks to be the culprit.
Ran the same test on all the DAC preamps I have:

Cambridge DACMagic+

L: 0
R: 0

Mytek Brooklyn+

L: 0.6
R: -0.1

MSB Analog:

L: 0.3
R: -0.3

None of the other preamps have the buzzing problem with the Creek amp.  Compared to the measurements from the Wadia this doesn't look so bad.
gregdude 

51 posts   
04-23-2019 10:52pm


A ground loop would be characterized as a hum. What you say you heard was a buzz, that blew out a woofer, almost definitely DC


Not always. Not if there are higher frequency harmonics riding on the equipment grounding conductor.
https://www.youtube.com/watch?time_continue=5&v=FBRPHojSGAs

https://www.youtube.com/watch?v=Nakpj_Mee0Q
.
solobone22 OP

7 posts
04-24-2019 11:04am

I have a Klein Tools MM300. I set the meter to 200m and tested the RCA output touching the positive line inside the connector and the negative line to the outer barrel.

The left channel read 1.0.
The right channel read -0.1.

This looks to be the culprit.

@ solobone22 ,

How about setting the meter to DC volts and preform the same tests.

Then just for the heck of it set the meter to AC volts and preform the same test.
.

Hi Jim (Jea48),

Looking at a photo of the particular meter shown here I see that there are two settings of the mode switch marked "200m." One is for DC current, which it seems safe to assume is not the one the OP used, and the other is the most sensitive of the DC volts settings. So it sounds like the worst case of any of the readings the OP reported was 1 mv (millivolt) DC, which won’t hurt anything.

I second your suggestion of trying an AC volts measurement, although the most sensitive AC volts setting provided by this meter has a full scale of 200 volts.

Best regards,

-- Al
Al,

Thanks for the better picture of the meter. I wondered why the OP was measuring DC current.

https://data.kleintools.com/sites/all/product_assets/hires/klein/mm300_photo.jpg

gs5556 theory makes sense to me. I just thought the DC voltage at the outputs of the DAC would have been higher.

The left channel read 1.0.
The right channel read -0.1.


Jim
I tested using meter set to the V with the 3 dashes below the solid line.  That is the setting for DC.  Anyone have further thoughts on this?

@gs5556 Is this something that would present itself all of the time or only intermittently? I'm referring to the DC issue.
I tested using meter set to the V with the 3 dashes below the solid line.  

That's correct, and that is what I assumed you were doing.

I suppose it is possible for a large DC output to occur intermittently, but it seems unlikely.  Especially if the problem has occurred when no signal was being provided to the DAC (i.e., when a CD was not being played).

Regards,

-- Al
 

Hi Al!

I'm looking at testing the power supply now.  Per the Wadia documentation here:

http://www.wadia.com/ContentsFiles/Wadia%20121%20Brochure%20web.pdf

• Multiple stages of regulation were designed into the main circuit board to ensure quiet, stable DC power is available for all sensitive circuits.
• A dedicated voltage regulator for the master oscillators is used to provide clean pure DC power for maximum clock stability.

I wouldn't know where to start testing those pieces and do not want to dig around inside of the unit.

On the repair bill they stated that the internal power supply was replaced as well as an IC.

My aim again with all of this is to get some measurement of the issue prior to sending it back to them.
Not sure what to suggest at this point, other than measuring the DAC’s output on the 200 volt AC scale as was suggested (while no music is playing).

It would be nice if it were possible to also measure the output voltage(s) of the DAC’s external switching power supply, but that would seem to be impracticable due to the nature of the connector and the lack of definition as to how it is wired.  I suppose it's possible that it has been intermittently misbehaving, and caused the failure of the internal supply.

Good luck. Regards,

-- Al


@almarg your suggestion is most appreciated - will do that this evening
The only way to get a handle on what's going on is to play a 50 hz 0 dBFS test signal through the Wadia and measure the output voltage. It should be 2.0 volts RMS (the 50 hz tone will be more accurately measured since most DMM's are optimized for 50/60 hz). The DC offset can vary as a signal is passed through so the 1.0 VDC reading with no signal won't give you the extent of the problem.

So the answer is the no signal condition will be there most of the time but the actual offset may not.
The Adcom 535 Mk II is capacitor coupled at the input and the Creek and Carver are not. What you have is DC at the output which the Creek passed through to the speaker before going into protection mode. The Adcom's input cap blocked the DC and didn't pass the buzz-causing DC to the speaker. DC is enough to blow out a woofer.
Can you put a capacitor between  Wadia 121 output and input of the Carver to confirm that?
Update: I went to test the Creek and it wouldn't power on. Opening it up I found the 2 fuses on the board were blown. Unfortunately they didn't blow before the speaker was damaged. The unit even worked a short period after the initial incident. Checked the fuses and they were the correct rating.

Testing with 50/60 hz test tone later today.
I'm trying to think of a way that a simple ground loop, DC or ultrasonic noise would destroy a woofer and leave the amp intact.
None of those seem to be the issue; a ground loop can't do that, DC is blocked in any sane design by coupling capacitors and ultrasonics take out tweeters, not woofers.
So what could cause a really loud angry buzz before outright failure of the woofer? I'm guessing the 'snap' was the woofer voice coil bottoming out. AC line voltage is one possibility- but this happened without the Creek being damaged.
I tested the amp and it seems to be ok through another pair of speakers.
I'm wondering if the DAC is grounded correctly. If the green wire and white wire were exchanged on the IEC it might be able to do something like this. A DVM should tell the story- measure the resistance from the center pin of the IEC connector to the chassis and see what you get. It should read 0.something ohms- shorted. What is the resistance to ground of the other two pins? If either is shorted to ground its a problem!

@atmasphere

You must have missed the OP’s last post.
solobone22 OP

11 posts   
04-28-2019 2:47pm Update:

I went to test the Creek and it wouldn’t power on. Opening it up I found the 2 fuses on the board were blown. Unfortunately they didn’t blow before the speaker was damaged. The unit even worked a short period after the initial incident. Checked the fuses and they were the correct rating.


You must have missed the OP’s last post.
I did see that. But what we know is that the amp didn't fail- those fuses provide power to the amp, but the amp worked because filter caps in the power supply still held some voltage. That's my assumption that checks all the boxes anyway. We'll know more after the fuses are replaced and the amp tested.
Post removed