Emotiva UPA-2: Lost Left Channel

mattlach

mattlach

Junior Audioholic
Hey all,

Almost 2 years ago, I picked up an Emotiva UPA-2 amp used on craigslist.

Up until today it had been working perfectly, until without warning the left channel just stopped working. The unit has not been dropped banged or mishandled in any other way, so the sudden loss of the left channel is perplexing me.

Currently I have it set up like this:
Schiit Modi Multibit DAC -> Sys Preamp (used as source switch) -> Schiit Jotunheim Headphone Amp (used both for headphones, and as pre-amp for the UPA-2).

I also have the line outs on my UPA-2 hooked up to my Subwoofer.

I just don't have sound out of the left channel, but the little blue LED's on the front of the UPA-2 that indicate the channel is operational are both on.

So, I have started troubleshooting:

- First I checked the connections to my speakers to see if they had come out (they hadn't).

- Then I decided to check if it was my source or the amp that was acting up, by listening to my headphones. The headphones have sound on both channels, so the problem is downstream from the Jotunheim headphone amp.

- Next I noticed that the Subwoofer also does not play any sound when there is input on the left channel.

- Next I swapped the inputs. The test singnals now play the left channel sound on the right channel, so it's not the pre-amp or wires, but something in the amp itself, and whatever it is, is preventing the signal both from reaching the internals of the amp, and from reaching the subwoover.

My theory now is that somehow, something on the inside of the amp has become disconnected, possibly on the inside of the binding posts.

Does this sound reasonable?

I'm unplugging the amp right now to let the caps discharge over night, and then tomorrow I plan on opening it up to see if I can see anything obvious wrong with it.

Does anyone have any further suggestions? Any further tests I should do? Any cautions? I am not a stranger to electrical stuff or soldering, but I have never worked on an amp before.

Any help anyone can provide would be greatly appreciated.

Thanks,
Matt
 
TLS Guy

TLS Guy

Seriously, I have no life.
Hey all,

Almost 2 years ago, I picked up an Emotiva UPA-2 amp used on craigslist.

Up until today it had been working perfectly, until without warning the left channel just stopped working. The unit has not been dropped banged or mishandled in any other way, so the sudden loss of the left channel is perplexing me.

Currently I have it set up like this:
Schiit Modi Multibit DAC -> Sys Preamp (used as source switch) -> Schiit Jotunheim Headphone Amp (used both for headphones, and as pre-amp for the UPA-2).

I also have the line outs on my UPA-2 hooked up to my Subwoofer.

I just don't have sound out of the left channel, but the little blue LED's on the front of the UPA-2 that indicate the channel is operational are both on.

So, I have started troubleshooting:

- First I checked the connections to my speakers to see if they had come out (they hadn't).

- Then I decided to check if it was my source or the amp that was acting up, by listening to my headphones. The headphones have sound on both channels, so the problem is downstream from the Jotunheim headphone amp.

- Next I noticed that the Subwoofer also does not play any sound when there is input on the left channel.

- Next I swapped the inputs. The test singnals now play the left channel sound on the right channel, so it's not the pre-amp or wires, but something in the amp itself, and whatever it is, is preventing the signal both from reaching the internals of the amp, and from reaching the subwoover.

My theory now is that somehow, something on the inside of the amp has become disconnected, possibly on the inside of the binding posts.

Does this sound reasonable?

I'm unplugging the amp right now to let the caps discharge over night, and then tomorrow I plan on opening it up to see if I can see anything obvious wrong with it.

Does anyone have any further suggestions? Any further tests I should do? Any cautions? I am not a stranger to electrical stuff or soldering, but I have never worked on an amp before.

Any help anyone can provide would be greatly appreciated.

Thanks,
Matt
Sometimes in cases like this you do see damaged components. However, the damaged components are far more often then not the result of the fault and NOT the cause.

Now you will probably find line fuses to each power amp. Identify them first. See if they are blown. If so replace them only once with a fuse of the correct rating and blow rate. That means slow or fast.

More likely than not the replacement fuses will blow right away, as fuses usually blow for a reason.

Probably the commonest reason for a power amps to fail, is failure of one of the power transistors. The usual failure mode is rail to ground short circuit through the failed device and hence the blown fuse, or fuses.

However you will see a lot of components in that unit, anyone of which on that channel could fail and cause your problem.

Unless you are experienced in repair and have extensive test equipment, you are not going to repair it, unless you are lucky enough for it to be a board connecting lead coming off. Most of us don't get that dumb luck as a rule.

If you see nothing obvious, and or fuses are blown and blow again after replacement, you will have to pack the unit up and send it to Emotiva for repair and pay the piper. A unit like that would be worth the cost of shipping and repair I would think.

I think you should not be surprised the unit has failed, but that units of that complexity do not fail more often.

In the old days of tubes, we were chasing faults pretty frequently.
 
mattlach

mattlach

Junior Audioholic
Sometimes in cases like this you do see damaged components. However, the damaged components are far more often then not the result of the fault and NOT the cause.

Now you will probably find line fuses to each power amp. Identify them first. See if they are blown. If so replace them only once with a fuse of the correct rating and blow rate. That means slow or fast.

More likely than not the replacement fuses will blow right away, as fuses usually blow for a reason.

Probably the commonest reason for a power amps to fail, is failure of one of the power transistors. The usual failure mode is rail to ground short circuit through the failed device and hence the blown fuse, or fuses.

However you will see a lot of components in that unit, anyone of which on that channel could fail and cause your problem.

Unless you are experienced in repair and have extensive test equipment, you are not going to repair it, unless you are lucky enough for it to be a board connecting lead coming off. Most of us don't get that dumb luck as a rule.

If you see nothing obvious, and or fuses are blown and blow again after replacement, you will have to pack the unit up and send it to Emotiva for repair and pay the piper. A unit like that would be worth the cost of shipping and repair I would think.

I think you should not be surprised the unit has failed, but that units of that complexity do not fail more often.

In the old days of tubes, we were chasing faults pretty frequently.

Thank you for your reply.

I haven't taken the case off yet, but based on my experiences during my trouble shooting I suspect this is a matter of a simple intermittent connection problem.

The hint that is making me suspect this is that he left channel is not only not being amplified by the amplifier, but also the signal is not being carried through to the subwoofer which is connected to the line out on the amplifier.

My understanding of the design of the UPA-2 is that the RCA line ins and the RCA line outs are a simple bridged connection, so if the signal isn't getting to the sub, this suggests the connection is not being made to the amp either.

So, my suspicion is that if I examine the inside of the case where the RCA connector is, I'll find a loose wire or a broken connection or something like that.

It seems odd something like this would just happen on its own, but its winter, and my office gets pretty cold at night, so it could be up to thermal contraction/expansion.

Originally when I discovered this it made me suspect the RCA cable from my pre-amp was damaged, but swapping the channels left to right confirms that it works just fine.
 
mattlach

mattlach

Junior Audioholic
So,

I cracked it open and now I'm looking inside. I was hoping to see something obvious, but I don't

Some pictures (please excuse the dust, looks like I am going to have to blow it out before reassembling) You should be able to click on them for the full size.

Back of unit:
01-sml.jpg

Input PCB attached:
02-sml.jpg

Input PCB Removed, Top:
03-sml.jpg

Input PCB Removed. Bottom:
04-sml.jpg

I feel about 95% certain the issue is somehow with the input PCB. I guess it is time to crack out the multimeter and do some continuity testing.
 
Last edited:
mattlach

mattlach

Junior Audioholic
Hmm.

Went over the continuity on th einput board and it checked out, but after assembling everything together again, I have the same problem.

I tried with a different source and different cable, and the same. Then I jiggled the RCA cable on the back, and the left channel popped in and out for a fraction of a second.

So, it's definitely some sort of problem or intermittent contact on the input board, but where on the board? It's so weird that continuity checked out and then it didn't work when plugged in.

I wonder if I can buy a replacement input board. I'd have to solder that ground wire on the bottom... I hate PCB soldering (my hands arent the steadiest), but that one doesn't look TOO difficult.
 
mattlach

mattlach

Junior Audioholic
Well,

I have emailed Emotiva about ordering a spare part, but judging by other threads I have found while googling, it seems as if I am probably just going to be shot down with a statement that they "only do internal service".

If I have to pay Emotiva or an independent audio repair professional to repair this simple fault, it will probably wind up costing more than I paid for the unit used, so I might just have to scrap and replace it instead, which would be a shame.

03-sml-circled.jpg


Does anyone know if these double height dual RCA connections are a standard part I can find from an electronics catalogue? I can probably de-solder and resolder a new one. The through-hole mount looks fairly simple.

Appreciate any input.
 
mattlach

mattlach

Junior Audioholic
After some further research, it looks like what I might be dealing with is a custom all red version of a part with the part number FC684052.



It looks like the part I found is designed for stereo use, not for bridged use, but examining the Emotiva board, it appears it is electrically and physically the same, just different colors, and that the two connectors are bridged on the board...

Anyone have any thoughts?
 
mattlach

mattlach

Junior Audioholic
Alright,

The mystery deepens.

Both channels work if I remove the board and run it outside the case. (Side note, touching the wrong with with the case opened and powered on can kill you. Don't do this :p )

works_sml.jpg


Then, when I assemble the board back into the case, the left channel stops working again:

doesnt_work_sml.jpg


At least this explains why my continuity tests worked...
 
Last edited:
mattlach

mattlach

Junior Audioholic
My first thought was that maybe the solder joints at the base of the connector had cracked, and the pressure from being tightened to the case was pushing them in just the right way to break contact.

My next thought was maybe I should reflow those solder joints with my soldering iron in order to fix it.

Looking at the three joints that connect the RCA connector to the board though they look in perfect condition.

solder_joints_sml.jpg



This makes me think that there is some sort of break inside that dual board mount RCA connector, bringing me back to thinking about that replacement connector I linked above.

Anyone have any thoughts or suggestions?
 
TLS Guy

TLS Guy

Seriously, I have no life.
Hmm.

Went over the continuity on th einput board and it checked out, but after assembling everything together again, I have the same problem.

I tried with a different source and different cable, and the same. Then I jiggled the RCA cable on the back, and the left channel popped in and out for a fraction of a second.

So, it's definitely some sort of problem or intermittent contact on the input board, but where on the board? It's so weird that continuity checked out and then it didn't work when plugged in.

I wonder if I can buy a replacement input board. I'd have to solder that ground wire on the bottom... I hate PCB soldering (my hands arent the steadiest), but that one doesn't look TOO difficult.
I don't see a break on the board. From what you say, I think that it is possible that the RCA input block has failed.

Put an RCA plug into the input socket and see if it makes contact with the first solder point on the board. If it does not you have your problem. If it does trace all the point to point contacts on the board with your ohm meter. I don't see a break in the board, but you have to be alert for dry solder joints. Test all the solder joints on that channel.

If it all checks out then half of the dual IC op amp chip has failed most likely.

Can you post the number off that chip and I will look it up.

That board looks pretty serviceable to me and trouble shooting should be pretty simple.
 
mattlach

mattlach

Junior Audioholic
I don't see a break on the board. From what you say, I think that it is possible that the RCA input block has failed.

Put an RCA plug into the input socket and see if it makes contact with the first solder point on the board. If it does not you have your problem. If it does trace all the point to point contacts on the board with your ohm meter. I don't see a break in the board, but you have to be alert for dry solder joints. Test all the solder joints on that channel.

If it all checks out then half of the dual IC op amp chip has failed most likely.

Can you post the number off that chip and I will look it up.

That board looks pretty serviceable to me and trouble shooting should be pretty simple.

Greatly appreciate the offer. You are talking about th ecomponent labeled U01? That does look about the same size as I recall previous op-amps I ahve seen looking. (as mentioned, I'm a layman who (sortof) knows how to use a soldering iron and a multimeter, not an electrical engineer.) My discipline is Mechanical and Industrial engineering.

All that said, do you really think it is the OP amp?

The reason I ask is, if you refer back to my open box testing, in post #8 in this thread, the left channel magically starts working again, when I unmount the input board from the chassis.

This strongly suggests I am dealing with an issue of intermittent contact, not one of chip burnout, correct? And since the contacts look so good on the bottom, my thoughts immediately went to the internals of the RCA block.

My problem I have with any further continuity testing is that the damned thing works when I unscrew it from the chassis, and stops working again when I reassemble it to the chassis... So in order to confirm where the issue is, I need to do it while mounted, and I can't reach the bottom of the board with my multimeter when it is tightened down...
 
mattlach

mattlach

Junior Audioholic
I don't see a break on the board. From what you say, I think that it is possible that the RCA input block has failed.

Put an RCA plug into the input socket and see if it makes contact with the first solder point on the board. If it does not you have your problem. If it does trace all the point to point contacts on the board with your ohm meter. I don't see a break in the board, but you have to be alert for dry solder joints. Test all the solder joints on that channel.

If it all checks out then half of the dual IC op amp chip has failed most likely.

Can you post the number off that chip and I will look it up.

That board looks pretty serviceable to me and trouble shooting should be pretty simple.
As a workaround to not being able to reach any of the contacts on the bottom of th eboard, when it is assembled in place and not working, I tried this:

Plug in RCA connector with board out.

works2.jpg


Since the input and output are supposed to be electrically connected, we would expect the following:
Test continuity from ground to ground: Checks out fine.
Test continuity from center pin to center pin: Checks out fine.

This aligns with the fact that the left channel worked just fine in functional testing when the board was out of the chassis.

Now, lets do the same with the board assembled into the chassis:

doesnt_work2.jpg


Test continuity from ground to ground: Checks out fine.
Test continuity from center pin to center pin: Fails. No continuity.

This also aligns with the fact that the board works when out of the chassis , but not when installed, in functional testing.

So, it really sounds to me like there is an intermittent connection, and that the physical stresses of being tightened to the chassis is what is breaking the connection on the center pin.

Now th eonly question is where it is. My guess is inside the RCA block, but I can't prove it yet.
 
TLS Guy

TLS Guy

Seriously, I have no life.
I can't see how that unit could fail, unless the bar does touch the tip of the RCA plug when installed. Check and see if you get connection when pushing the contact bar to the RCA connector. If you do then bend the bar towards the tip of the plug.

Also make sure there are no dry joints to the pins of that connector block.

On the face of it this now seems a very simple problem.
 
mattlach

mattlach

Junior Audioholic
I can't see how that unit could fail, unless the bar does touch the tip of the RCA plug when installed. Check and see if you get connection when pushing the contact bar to the RCA connector. If you do then bend the bar towards the tip of the plug.

Also make sure there are no dry joints to the pins of that connector block.

On the face of it this now seems a very simple problem.

Appreciated. I ahve buttoned it up and put it away for the night, but I will try it again in a couple of days.

Is there an easy way to gain access to the bar? Looking at the pictures I took earlier, I can't quite see how I would access the portion that goes inside the connector to bend it, without breaking the whole thing open.

 
P

PENG

Audioholic Slumlord
So, it really sounds to me like there is an intermittent connection, and that the physical stresses of being tightened to the chassis is what is breaking the connection on the center pin.

Now th eonly question is where it is. My guess is inside the RCA block, but I can't prove it yet.
I think there is a good chance that you are correct about intermittent connection due to being tightened. Have you tried having the board back in place but not tightening it down? If it work just sitting there, then you can try tightening it gradually, see what happens, and go from there.

To me it is good news that it is just a connection/contact issue, it is a matter of time before you can find the point of trouble, by process of eliminations, metering, and applying pressure at different points to find the intermittent connection. I highly doubt the OPA is the issue, if it is, it won't work with the board outside consistently.
 

Latest posts

newsletter

  • RBHsound.com
  • BlueJeansCable.com
  • SVS Sound Subwoofers
  • Experience the Martin Logan Montis
Top