[f] AMB on Facebook

No attenuation after major mess

delta1

Moderator: linux-works

No attenuation after major mess

Postby OliverD » June 29th, 2018, 2:50 pm

So, I have been building a gamma 3 (which surprisingly went without any problems, currently waiting for some freetime to order panels) and upgrading my b22 to balanced when I made a chain of stupid mistakes which ended up killing the left side of my HD 650 and one of the b22 boards.
From what I gathered measuring everything I somehow managed to have a short from v+ (or v- don't remember at this point) to g when pressure was applied at a certain spot (using offboard heatsinks in a peasante dissipante) or if everything was neatly assembled into the case. Additionally I must have mixed up connections to the XLR output and to top it all off, since this was only a minor upgrade I obviously didn't need to measure the DC offset at output.

Long story short, ~30V DC offset took care of the HD 650 and one of the b22 boards went up in smoke. All of that is fixed now (repaired the board and bought a new HD 650) but collateral damage seems to be one of the delta 1 boards:
Whatever I do only one of the boards really affects the input signal. The affected board has relais that click and I couldn't find any apparent shorts or parts that were obviously broken. And after a few minutes the ground loop breaker is too hot to touch.

tldr: I probably managed to kill a delta 1 board with a series of stupid stuff. This board won't attenuate at all and the ground loop breaker can melt ice caps.

Can you guys help me to determine if this board is indeed faulty/lost or which parts could be exchanged to save the board? Any hints what could be done or what I should check would be greatly appreciated!
OliverD
 
Posts: 104
Likes: 1 post
Liked in: 3 posts
Joined: March 26th, 2011, 4:56 am
Country: Germany (de)

Re: No attenuation after major mess

Postby amb » June 30th, 2018, 1:01 am

The ground loop breaker simply bridges signal ground and AC ground via a resistor and a capacitor in parallel. There shouldn’t be any significant voltage on either side, so it shouldn’t get hot. Maybe you have a wiring error, which connected one of these sides to a voltage source?
User avatar
amb
Site Admin
 
Posts: 10355
Likes: 175 posts
Liked in: 478 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: No attenuation after major mess

Postby OliverD » June 30th, 2018, 8:33 am

I actually forgot to mention that I am using a two box solution so the b22 case with the LCDuino is DC-only. I know I wouldn't need to populate the ground loop breaker in such a setup, but I did.
If I only connect the board with the hot side nothing happens and the ground loop breaker on that one stays cold. However once I connect the board that I think I killed with the messup described above, I have no attenuation on that second board (while the first one attenuates just fine) and as a side effect, both ground loop breakers heat up.

Is it a problem that I installed both ground loop breakers in a DC-only case?
OliverD
 
Posts: 104
Likes: 1 post
Liked in: 3 posts
Joined: March 26th, 2011, 4:56 am
Country: Germany (de)

No attenuation after major mess

Postby amb » June 30th, 2018, 1:17 pm

The ground loop breaker is not needed in a two-chassis configuration. But the resistor shouldn't get hot. You should go though your build carefully and measure for any possible sort circuits.
User avatar
amb
Site Admin
 
Posts: 10355
Likes: 175 posts
Liked in: 478 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: No attenuation after major mess

Postby OliverD » January 29th, 2019, 5:03 am

Never really took the time to do this until now since the y3 provided attenuation and I preferred listening to music over listening to my multimeter beep for short circuits.

I finally had enough of it though and actually found the culprit:
The cold (minus) side of the delta1s has a constant 16-17Ohm between output G and output L/R. Seems like this board didn't survive whatever I did to it initially. I currently only have a spare delta1 from a previous build but that's calculated for a different Zatt (50k instead of 25k iirc) so I guess I can't swap those out without also swapping out the resistors.

Could anyone provide any ideas for salvaging the delta1? I am at a total loss here as usual and really depend on someone more knowledgeable than me. Any help is greatly appreciated!
OliverD
 
Posts: 104
Likes: 1 post
Liked in: 3 posts
Joined: March 26th, 2011, 4:56 am
Country: Germany (de)

Re: No attenuation after major mess

Postby amb » January 29th, 2019, 11:50 pm

Did you examine the board carefully to make sure there are no solder bridges? Also, power up, turn the volume to maximum, then retest to see if you still have the short. If you do, then maybe the last relay is stuck.
User avatar
amb
Site Admin
 
Posts: 10355
Likes: 175 posts
Liked in: 478 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: No attenuation after major mess

Postby OliverD » January 30th, 2019, 5:17 am

Thank you for your feedback.
I carefully examined the whole board twice and found no solder bridges or other stuff that's visually off. Powering it on and either putting it at max, min volume or even some values in between doesn't change the resistance between output Signal and ouput ground although I can hear relais clicking on each step.
Shouldn't this always have infinite resistance anyway? At least that's how my other boards behave when measuring the resistance between OL and GL or OR and GR.
Edit: Obviously I am merely too clumsy to properly hold a multimeter to my other boards, these do indeed change resistance when volume is changed.

So my assumption is correct that I can't substitute a board with Zatt 25k with one that has Zatt 50k since this is a balanced build and there would be one with 25k for the hot side and 50k for the cold side?

And to top it off, if the value is due to a stuck relais and it indeed shouldn't always measure to infinite, could this explain voltage on the ground loop breaker that heats the resistor up to untouchable levels?
OliverD
 
Posts: 104
Likes: 1 post
Liked in: 3 posts
Joined: March 26th, 2011, 4:56 am
Country: Germany (de)

No attenuation after major mess

Postby amb » January 30th, 2019, 1:47 pm

If you look at the resistor value calculator schematic, you’ll see the last relay would shunt a low value resistor from signal to ground under some attenuation values. If that relay is stuck in that position then you’ll have the low resistance to ground situation at all attenuators settings.

In a balanced amp, the resistances should be the same between the hot and cold side.

Can you describe in detail what is the problem you’re encountering now?

The ground loop breaker resistor getting hot is most likely an unrelated problem. Since you have a 2-chassis build, you should just remove the ground loop breaker parts.
User avatar
amb
Site Admin
 
Posts: 10355
Likes: 175 posts
Liked in: 478 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: No attenuation after major mess

Postby OliverD » January 31st, 2019, 1:24 pm

I tried reomoving the 16Ohm resistor since I suspected more than one of the relais might have been destroyed by my previous mistake and actually now the multimeter alsways reads 12,75kOhm so I guess this board is basically lost unless I swtich out a bunch of relais without damaging the rest of the board.
At this point my best option might be to switch out the resistors from the broken board and use them on the working one I have with the wrong Zatt.

Although 12,75kOhm matches no resistor value I soldered onto the board, but it is exactly half of the RTx resistors, does that make any sense to you?
Maybe I need to remove the relais and then measure again.

My second problem (since these two seem to be unrelated) was that the transformer for the s22 powering the left channel would make a strong buzzing sound and heat up although a lot slower than the ground loop breaker parts. Since both of these problems only occur once the faulty board is involved, I guess at this point I should simply order parts for a new one since I guess I risk damage to other components when continue using this.

Maybe I will try to pull all resistors and transfer them to my spare board and see if that makes the whole thing work normally again.

Thanks for your feedback!
OliverD
 
Posts: 104
Likes: 1 post
Liked in: 3 posts
Joined: March 26th, 2011, 4:56 am
Country: Germany (de)

Re: No attenuation after major mess

Postby amb » January 31st, 2019, 2:54 pm

OliverD wrote:...
Although 12,75kOhm matches no resistor value I soldered onto the board, but it is exactly half of the RTx resistors, does that make any sense to you?

It’s possible that previous stages have resistor shunts to ground, depending on what your volume was set to before disconnecting power. These relays are latching type, meaning that they retain their positions even after power is turned off.

My second problem (since these two seem to be unrelated) was that the transformer for the s22 powering the left channel would make a strong buzzing sound and heat up although a lot slower than the ground loop breaker parts. Since both of these problems only occur once the faulty board is involved, I guess at this point I should simply order parts for a new one since I guess I risk damage to other components when continue using this.

Is there excessive current draw on that channel? And are you saying that the transform heats up only when the faulty δ1 board is connected? That’s strange because the δ1 is powered by 5V from the LCDuino-1, not the σ22s.

Maybe I will try to pull all resistors and transfer them to my spare board and see if that makes the whole thing work normally again.

Removing resistors is certainly easier than removing relays.
User avatar
amb
Site Admin
 
Posts: 10355
Likes: 175 posts
Liked in: 478 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Next

Return to δ1 relay-based R-2R stereo attenuator

Who is online

Users browsing this forum: No registered users and 3 guests

Quick Links

AMB audio DIY resources
AMB audio shop