UA-5 Clipping above -6db in one channel - Need Help

[terrapinj]

Been having some problems with my bm2p+ ua-5 for the past couple months. I first noticed the problem by hearing some clipping/distortion in one channel. See waveform1 & waveform2 for screenshots from that show, I have included a link to an audio sample as well, sample1. Even as an mp3 you can hear the distortion. i was running mod 461 > bm2p+ ua-5 > jb3 onstage.
http://s30.yousendit.com/d.aspx?id=3Q4ZJCHZ7LI0B1XR83V0EC5MMB

The next two jpg files are from a different night. The ua-5 only sample is from the opener Big Chief Something running mod 461>bm2p+ ua-5 > jb3. The only thing changed during the set break was the addition of the mp-2 in front running 1/8" stereo out to the RCAs in the back of the ua-5. The mp2-ua5 file is from the headliner Galactic. I ran mod 461>mp2 > 1/8"-RCA > bm2p+ ua-5 > jb3. No audio samples attached, but I don't think they are necessary as you hear the same type of distortion as you do from the 1st mp3 sample, but if someone thinks its necessary I can post some samples.

I sent the ua-5 back to Chris (busman2) to test and see what he could do with it. Unfortunately Chris was unable to reproduce similar results when he tested at home. I don't believe he tested it in the field but not sure if that would make a difference.

I received my ua-5 back from Chris this week and only had a chance to do a quick test at home before having to go to the show. I ran mod 463 > bm2p+ ua-5 > jb3 for UM on Thursday night. You can't really hear the distortion as clearly as the 1st night I encountered the problem but you can definately hear it in some places if you listen carefully. More disturbing however is the waveform which I have included samples of as well, screenshot-set2 of the entire set and screenshot-set2b, zoomed in.

Chris and I are pretty much at a loss to figure out what the problem is. I am 99% sure that the problem is somewhere in the pre stage of the ua-5 as I have run at least a dozen or show shows using the ua-5 as ADC only behind the mp-2 and had no issues at all. I would really appreciate any feedback or help. Let me know if more samples or screenshots would help.

[jeromejello]

sorry, i am of no help, but damn, that looks fucked up...
+t for your troubles

[Brian Skalinder]

I had a similar issue with my Oade modSBM-1.  Sent it back to Doug and he replaced the dying op amps.  As long as my gain was below a certain threshold, the problem did not appear.  Hit the threshold, and the waveform looked (and sounded) similar to yours.

Try testing at home with the MP2 through the front XLRs on the UA5 - I bet the problem resurfaces.  But there may be a gain threshold at which the problem kicks in, so try different levels.  Then with the MP2 in the chain, try adding gain with just the MP2, and then lower the MP2 gain and try the UA5 gain.  I'm guessing that whether you apply the gain on the UA5 or on the MP2, you'll still experience the problem.  Just a guess... 

[terrapinj]

I had a similar issue with my Oade modSBM-1.  Sent it back to Doug and he replaced the dying op amps.  As long as my gain was below a certain threshold, the problem did not appear.  Hit the threshold, and the waveform looked (and sounded) similar to yours.

Try testing at home with the MP2 through the front XLRs on the UA5 - I bet the problem resurfaces.  But there may be a gain threshold at which the problem kicks in, so try different levels.  Then with the MP2 in the chain, try adding gain with just the MP2, and then lower the MP2 gain and try the UA5 gain.  I'm guessing that whether you apply the gain on the UA5 or on the MP2, you'll still experience the problem.  Just a guess... 

somehow i knew you would have an answer to this 


here are some of the screenshots chris got when he tested the unit, they seem to be well over the 6db point with no problems, which is where my waves flatten out.

i'll check it out and see what I can find. thanks Brian! +T

[Brian Skalinder]

somehow i knew you would have an answer to this

Well...not so much an answer as a shot in the dark.

here are some of the screenshots chris got when he tested the unit, they seem to be well over the 6db point with no problems, which is where my waves flatten out.

Odd that he couldn't reproduce the problem.  That always makes problems like this more frustrating.  Try the tests, anyway.  You may also want to try running a different set of mics in through the UA5 XLRs.  And different cables.  Basically, try to isolate each component in your testing, if you can: mic caps, mic bodies, cables, UA5 XLR inputs, UA5 pre, UA5 phantom, UA5 RCA inputs, etc.

[poorlyconditioned]

It sounds like you've got distortion with both RCA (line) and XLR (mic) inputs, right?

In that case, you could narrow it down to the ADC and/or a *single* opamp that drives the ADC.

You can debug this in your own home by:
- generating a sine wave in your audio program ("generate tone" or somesuch).
- do this at 0dB FS and transfer to your NJB3 via firewire
- plug a 1/8"<>2XRCA adapter into line out of your NJB3 and plug into RCA on your UA5

- plug the UA5 into USB port on your computer and record there
- play the wave (on your NJB3) and record the signal (on your computer), adjusting the gain on your (RCA inputs) of UA5.  Adjust gain so you're recording various levels, like 0dB, -3dB, -6dB.

See you if you can reproduce errors.

I've reproduced problems (different, but serious hardware problems) doing this.

Check solder connections on that opamp.  Maybe it is just connections and not the opamp itself.  I've had this problem during my hacking...

  Richard

[terrapinj]

It sounds like you've got distortion with both RCA (line) and XLR (mic) inputs, right?

In that case, you could narrow it down to the ADC and/or a *single* opamp that drives the ADC.

You can debug this in your own home by:
- generating a sine wave in your audio program ("generate tone" or somesuch).
- do this at 0dB FS and transfer to your NJB3 via firewire
- plug a 1/8"<>2XRCA adapter into line out of your NJB3 and plug into RCA on your UA5

- plug the UA5 into USB port on your computer and record there
- play on your NJB3 and record the signal, adjusting the gain on your (RCA inputs) of UA5

- adjust gain so you're recording various levels, like 0dB, -3dB, -6dB.

See you if you can reproduce errors.

I've reproduced problems (different, but serious hardware problems) doing this.

  Richard

no problem with the RCA line input - i have run that many times without an issue

the only issue seems to be with the XLRs which i've only used with phantom on, i will test using xlr line input as well to see if there's a problem as well

[poorlyconditioned]

It sounds like you've got distortion with both RCA (line) and XLR (mic) inputs, right?

In that case, you could narrow it down to the ADC and/or a *single* opamp that drives the ADC.

You can debug this in your own home by:
- generating a sine wave in your audio program ("generate tone" or somesuch).
- do this at 0dB FS and transfer to your NJB3 via firewire
- plug a 1/8"<>2XRCA adapter into line out of your NJB3 and plug into RCA on your UA5

- plug the UA5 into USB port on your computer and record there
- play on your NJB3 and record the signal, adjusting the gain on your (RCA inputs) of UA5

- adjust gain so you're recording various levels, like 0dB, -3dB, -6dB.

See you if you can reproduce errors.

I've reproduced problems (different, but serious hardware problems) doing this.

  Richard

no problem with the RCA line input - i have run that many times without an issue

the only issue seems to be with the XLRs which i've only used with phantom on, i will test using xlr line input as well to see if there's a problem as well

OK, just to be sure, run this *controlled* test on the RCA inputs.  That will completely rule out any problem, all the way up to 0dB.

Well, if it is only XLR inputs, you've reduced it to the other two opamps .  Or the phantom power and/or blocking capacitors, I suppose.  But I don't know what Chris has done with these, so it is harder for me to help there.

You can do the same systematic testing with the XLR inputs.  Generate a sinewave tone in your audio program.  Attenuate it by say, -20dB.  Then save to the NJB3.  Play on the NJB3 while recording on your computer.  Try it in both the XLR and the TRS lines, while cranking the gain up to as close to 0dB as possible.  You should get the same perfect sinewave out.  If not, record what level and what input type (TRS, XLR) causes the problem.  You could further isolate by trying balanced (TRS) and unbalanced (TS) inputs.  Oh yeah, try with and without phantom power.

My guess is either a bum opamp (or solder connection) or a problem with one or more blocking caps.

Good luck!
  Richard

[terrapinj]

i don't have anything terminting in TRS so I can't test that at all

im not exactly sure about creating a sinewave with soundforge so i just did a test in front of speakers at home

Test 1: 463 > dogstar > ua-5 (gain on ua-5 around 5 o'clock)
some distortion/clipping - waveform is symetrical but looks like it may be clipping around 3db
-one channel sounds like it intermittantly drops but this can't be seen in the waveform

Test 2: 463 > dogstar > mp-2 > canare XLRs > ua-5 (gain knob all the way left (no gain) on ua-5)
major distorted waveform - major distortion

Test 3: 463 > dogstar > mp-2 > canare XLRs > ua-5 (gain knob at 4 o'clock on ua-5 - knobs on mp-2 are all the way left (no gain)
similar results as Test 1

http://s33.yousendit.com/d.aspx?id=1JXZ06TOIBYAG3PFF17WJ0IKSX

[poorlyconditioned]

i don't have anything terminting in TRS so I can't test that at all

im not exactly sure about creating a sinewave with soundforge so i just did a test in front of speakers at home

Test 1: 463 > dogstar > ua-5 (gain on ua-5 around 5 o'clock)
some distortion/clipping - waveform is symetrical but looks like it may be clipping around 3db
-one channel sounds like it intermittantly drops but this can't be seen in the waveform

Test 2: 463 > dogstar > mp-2 > canare XLRs > ua-5 (gain knob all the way left (no gain) on ua-5)
major distorted waveform - major distortion

Test 3: 463 > dogstar > mp-2 > canare XLRs > ua-5 (gain knob at 4 o'clock on ua-5 - knobs on mp-2 are all the way left (no gain)
similar results as Test 1

http://s33.yousendit.com/d.aspx?id=1JXZ06TOIBYAG3PFF17WJ0IKSX

It would be a lot nicer if you could give a pure sinewave.  Much easier to diagnose.  Zoom in and see if output is a sinewave, or if not, where/how it distorts.  Maybe some software guru can say how to generate a sinewave.  It should be something like "generate tone" in your software.  Please try this as it will save us all time in the future.  If not, find another package that will do it.  *Hey* I just remembered.  I think I put a tone on my webpage (for testing ADC on the UA5):
  www.cs.uwaterloo.ca/~mannr/UA5/tone1600.wav
Attenuate this by, say, -20dB, and then save to the NJB3.  Then connect line out of the NJB3 to mic or line in on the UA5.  Record *to USB* from the UA5 and post the results here.

  Richard

[terrapinj]

i don't have anything terminting in TRS so I can't test that at all

im not exactly sure about creating a sinewave with soundforge so i just did a test in front of speakers at home

Test 1: 463 > dogstar > ua-5 (gain on ua-5 around 5 o'clock)
some distortion/clipping - waveform is symetrical but looks like it may be clipping around 3db
-one channel sounds like it intermittantly drops but this can't be seen in the waveform

Test 2: 463 > dogstar > mp-2 > canare XLRs > ua-5 (gain knob all the way left (no gain) on ua-5)
major distorted waveform - major distortion

Test 3: 463 > dogstar > mp-2 > canare XLRs > ua-5 (gain knob at 4 o'clock on ua-5 - knobs on mp-2 are all the way left (no gain)
similar results as Test 1

http://s33.yousendit.com/d.aspx?id=1JXZ06TOIBYAG3PFF17WJ0IKSX

It would be a lot nicer if you could give a pure sinewave.  Much easier to diagnose.  Zoom in and see if output is a sinewave, or if not, where/how it distorts.  Maybe some software guru can say how to generate a sinewave.  It should be something like "generate tone" in your software.  Please try this as it will save us all time in the future.  If not, find another package that will do it.  *Hey* I just remembered.  I think I put a tone on my webpage (for testing ADC on the UA5):
  www.cs.uwaterloo.ca/~mannr/UA5/tone1600.wav
Attenuate this by, say, -20dB, and then save to the NJB3.  Then connect line out of the NJB3 to mic or line in on the UA5.  Record *to USB* from the UA5 and post the results here.

  Richard

i don't have any way to go jb3 > xlr-in UA-5

would it have similar results in I played the tone out of my speakers and recorded it with my mics like I normally would?

[poorlyconditioned]

i don't have anything terminting in TRS so I can't test that at all

im not exactly sure about creating a sinewave with soundforge so i just did a test in front of speakers at home

Test 1: 463 > dogstar > ua-5 (gain on ua-5 around 5 o'clock)
some distortion/clipping - waveform is symetrical but looks like it may be clipping around 3db
-one channel sounds like it intermittantly drops but this can't be seen in the waveform

Test 2: 463 > dogstar > mp-2 > canare XLRs > ua-5 (gain knob all the way left (no gain) on ua-5)
major distorted waveform - major distortion

Test 3: 463 > dogstar > mp-2 > canare XLRs > ua-5 (gain knob at 4 o'clock on ua-5 - knobs on mp-2 are all the way left (no gain)
similar results as Test 1

http://s33.yousendit.com/d.aspx?id=1JXZ06TOIBYAG3PFF17WJ0IKSX

It would be a lot nicer if you could give a pure sinewave.  Much easier to diagnose.  Zoom in and see if output is a sinewave, or if not, where/how it distorts.  Maybe some software guru can say how to generate a sinewave.  It should be something like "generate tone" in your software.  Please try this as it will save us all time in the future.  If not, find another package that will do it.  *Hey* I just remembered.  I think I put a tone on my webpage (for testing ADC on the UA5):
  www.cs.uwaterloo.ca/~mannr/UA5/tone1600.wav
Attenuate this by, say, -20dB, and then save to the NJB3.  Then connect line out of the NJB3 to mic or line in on the UA5.  Record *to USB* from the UA5 and post the results here.

  Richard

i don't have any way to go jb3 > xlr-in UA-5

would it have similar results in I played the tone out of my speakers and recorded it with my mics like I normally would?

I don't know what the speakers would do.  We (tapirs) have $1000's worth of gear and can't solve a cable problem!  Hmm.  Could you take apart one side of an XLR cable and use aligator clips?  There must be a way...

OK, try the speakers and see what happens.  What I'm looking for is if you can zoom in on the output wave and compare to the input, maybe you can characterize the type of distortion (clipping/compression, or ?).  I guess this still doesn't isolate it though.  But perhaps if you can narrow it down more, Chris can try to fix it again.

I'm glad I'm not doing mods for money, though.  I wouldn't want one of my units to go bad, esp. if shows are lost due to it...

  Richard

PS: I'll be away from this board for about 48hrs.

[terrapinj]

i really appreciate your input richard

i understand exactly what you are trying to compare now. i'm gonna run down to ratshack and see what i can find as far as adapters. is it likely that the xlr and 1/4" inputs are running through the same opamps in case I can only find 1/8" stereo>1/4" cable?

another T in 12

[poorlyconditioned]

i really appreciate your input richard

i understand exactly what you are trying to compare now. i'm gonna run down to ratshack and see what i can find as far as adapters. is it likely that the xlr and 1/4" inputs are running through the same opamps in case I can only find 1/8" stereo>1/4" cable?

another T in 12

The XLR and 1/4" run thru *almost* the same circuit.  Gain stages are the same, but blocking caps/phantom may differ.  Furthermore, Busman or Oade mods might change this.  It is possible that 1/4" may not even work any more.  (That is the case on my mods.)

OK, as far as cables go, the simplest would be to take a standard XLR cable, plug it in the UA5, and open the other side of the cable to expose the wires.  Then somehow(aligator clips maybe), clip onto the ends of that so you're NJB3 line out goes there.

The other thing is how much is your time (+ frustration) worth?  Maybe it is time to send it back to Chris!

Anyway, good luck with this.  I hate it when gear does not work!  The only time I get *real* upset...
  Richard

[jeromejello]

The XLR and 1/4" run thru *almost* the same circuit.  Gain stages are the same, but blocking caps/phantom may differ.  Furthermore, Busman or Oade mods might change this.  It is possible that 1/4" may not even work any more.  (That is the case on my mods.)

i believe the oade mods also disable the 1/4" inputs