• knoll
  • ADAT user
  • Offline
  • Registered: 2012-07-30
  • Posts: 13

Topic: UFX: Defective Line and Mic Input Sections?

I think the analog input section of my UFX is somewhat defective. Here's my analysis of the inputs. I use extreme values for demonstration purposes but the problems exists also on normal values. During the analysis only power cable, headphones and USB cable were connected to the UFX. If I unplug the USB cable, I can still hear the problems.

1. Let's start with normal situation at a Line Input. Here we have all Line Inputs configured at -10dBV. That seems ok (is it?), only small bumps@50Hz and 150Hz.

https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/1.%20Line%20Input%20with%20All%20Lines%40-10dBV%20--%20ok.png

2. Continuing from the above situation, let's crank all Mic Input gains to max. Heres what the Line Input looks like now. Spike@382Hz and some distortion.

https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/2.%20Line%20Input%20with%20All%20Lines%40-10dBV%20%2B%20All%20Mic%20Gains%40Max%20--%20spike%40382Hz%2Bmild%20distortion.png

3. Let's put the Mic Input gains back to minumum and change all Line Input configurations to Lo Gain. Here's the line input. Spike@6.85kHz and some mild distortion.

https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/3.%20Line%20Input%20with%20All%20Lines%40%2B4dBu%20%28or%20LoGain%29%20--%20spike%406.85kHz%2Bmild%20distortion.png

4. Continuing from the previous, let's crank all Mic Input gains to max again. Here's what the Line Input looks like now. Spike@382 and lots of distortion.

https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/4.%20Line%20Input%20with%20All%20Lines%40LoGain%20%2B%20All%20Mic%20Gains%40Max%20--%20spike%40382Hz%2Blots%20of%20distortion.png

5. Enough about the Line Input. Here's what a Mic Input looks like with All Mic Gains set to zero. This looks really good, no problems here smile

https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/5.%20Mic%20Input%20with%20All%20Mic%20Gains%40Min%20--%20ok.png

6. Cranking all Mic Input gains to Max. Now the Mic Input looks like following. Spike@382Hz and moderate distortion.

https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/6.%20Mic%20Input%20with%20All%20Mic%20Gains%40Max%20--%20spike%40382Hz%2Bdistortion.png

7. Let's go crazy and enable Inst from all Mic Inputs on top of the previous step. This is terrible, see for yourself. Spike@382Hz and crazy amount of distortion.

https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/7.%20Mic%20Input%20with%20All%20Mic%20Gains%40Max%20%2B%20Inst%20Enabled%20--%20spike%40382Hz%2Bcrazy%20amount%20of%20distortion.png


On top of those, I have noticed that quite often when adjusting Mic Input gain, the unit does not respond to the gain change (display value changes but not the actual gain). It may change after a second but it may also take even up to a minute before the gain changes, and the change is then accompanied by a loud-ish pop. That happens whether I adjust the gain from TM or UFX.

Also, during the analysis I noticed a couple of times that changing the Line Input configuration (-10dBV, +4dBu, Lo Gain) did not actually change anything. After a while the change occured though, similarly as the gain change in the Mic Inputs. This I tried only from TM.

This is not the first set of problems for me with this unit. I've already shipped it twice back to the seller (1 month roundtrip each time). I'm guessing this will become trip #3. Quite annoying because we are trying to record an album at the same time.

  • Registered: 2011-10-27
  • Posts: 47

Re: UFX: Defective Line and Mic Input Sections?

I'm not sure but I think that you might be my ex-worker from IT company because symptoms are almost exactly same. Anyway: I think this hardware should be replaced with another without any fixing procedures. I think that if 3rd trip is coming it is already pointed that it can't be fixed and it's not problem in your computer.

3 Reply by knoll (edited by knoll 2012-10-15 21:14:19)

  • knoll
  • ADAT user
  • Offline
  • Registered: 2012-07-30
  • Posts: 13

Re: UFX: Defective Line and Mic Input Sections?

Yes, I probably am :-) and still working there.

1st trip didn't really fix anything. I wasn't pleased and sent the unit for a 2nd trip which fixed the "main" issue I was having back then (higher noise levels on certain gain settings).

The issue about "adjusting Mic Input gain and the unit not responding to the change" wasn't fixed either on 1st nor 2nd trip. Maybe my description about it wasn't clear enough as I was focusing more on the "main" issue I was having then. But now I want to get this issue fixed as well because it's getting quite annoying not knowing if a gain change has been registered or not.

About these new issues, spikes+distortion and Line Input configuration change not doing anything. I don't know if I just have missed these earlier or if they've emerged since the last trip.

I'm not looking forward to sending the unit for yet another trip but I guess that's what's coming. I sure wouldn't mind about replacing the whole unit.

4 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 24,922

Re: UFX: Defective Line and Mic Input Sections?

"adjusting Mic Input gain and the unit not responding to the change" is no issue, but your way of measuring - without input signal. The behaviour comes from the Mic preamp chip and has no meaning in real life.

Spike measurement I can not comment at this time. Line changes instantly, this is not a known problem of the device nor the driver/TM FX, IMHO you got confused and it works like it should.

Regards
Matthias Carstens
RME

MC's Website

  • knoll
  • ADAT user
  • Offline
  • Registered: 2012-07-30
  • Posts: 13

Re: UFX: Defective Line and Mic Input Sections?

MC wrote:

"adjusting Mic Input gain and the unit not responding to the change" is no issue, but your way of measuring - without input signal. The behaviour comes from the Mic preamp chip and has no meaning in real life.

Ok, I will test this with mic connected to the input to see if that changes the behavior.

MC wrote:

Spike measurement I can not comment at this time. Line changes instantly, this is not a known problem of the device nor the driver/TM FX, IMHO you got confused and it works like it should.

What do you mean by "Line changes instantly"? Those spikes are static for me, doesn't change at all. I can definitely hear them with headphones and also, if I record guitar though the inst input, I can hear those spikes+distortion in there.

Niko

6 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 24,922

Re: UFX: Defective Line and Mic Input Sections?

You wrote about Line Input configuration change not doing anything. Got it?

Regards
Matthias Carstens
RME

MC's Website

7 Reply by knoll (edited by knoll 2012-10-17 23:11:38)

  • knoll
  • ADAT user
  • Offline
  • Registered: 2012-07-30
  • Posts: 13

Re: UFX: Defective Line and Mic Input Sections?

knoll wrote:
MC wrote:

"adjusting Mic Input gain and the unit not responding to the change" is no issue, but your way of measuring - without input signal. The behaviour comes from the Mic preamp chip and has no meaning in real life.

Ok, I will test this with mic connected to the input to see if that changes the behavior.

Tested. You'll need mic (or instrument) connected and a input signal, just like you said, and the gain change works as expected.

But at least for me, this only seems to work for a while after the unit has been turned on. At some point, maybe an hour later or so, the gain changes didn't work anymore like they did at the beginning. During the hour I swapped mic and instruments cables to different inputs several times so that could be the reason(?). I then tested what happens if I turn the unit off for a minute and back on. The gain changes worked again like they should. I than swapped places of one instrument and one mic cable and enabled/disabled inst accordingly. The gain changes still worked ok for the mic inputs that I didn't change the cables from but the mic and inst inputs' gain change didn't work anymore. Turning the unit off and on again, fixes this and gain changes work again as expected. Firmware or hardware bug?

MC wrote:

Line changes instantly, this is not a known problem of the device nor the driver/TM FX, IMHO you got confused and it works like it should.

MC wrote:

You wrote about Line Input configuration change not doing anything. Got it?

Oh yeah, got it smile ...and I couldn't reproduce it this time so you may be correct about me getting confused. But I'll test this later when swapping cables between the line inputs. Could be similar issue as with the mic input gain change?

MC wrote:

Spike measurement I can not comment at this time.

I decided to do more measurements about these spikes. "Real life" measurements this time because that seems to be the key.

In the following screenshots I have mic connected to input 9, 10, 11 and 12. On measurement at a time, same mic, same cable, changing input between measurements. First screenshot is from input 9, gain maxed. Inputs 10 and 11 show similar behavior. The second screenshot is from input 12, gain maxed. There are spikes at 7kHz and 14kHz that aren't present in other mic inputs.
https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/8.%20Mic%20Input%209%2C%2010%20%26%2011%40Max%20Gain.png
https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/9.%20Mic%20Input%2012%40Max%20Gain%20--%20spikes%407%2614kHz.png

Continuing with Inputs 9, 10, 11 and 12 but this time with guitar connected to one, inst enabled, and gain maxed. 3 mics connected to the other inputs not occupied by the guitar, gain at zero for the mics. The first screenshot is from input 9 that has spikes at 7kHz and 14kHz. Input 10 shows similar behavior. The second screenshot is from input 11 which has two additional spikes at 3.4kHz and 17kHz. The third screenshot is from input 12 that has even more additional spikes than input 11. The fourth screenshot is from input 12 again but with the 3 other mic inputs' gain set to max (Other inputs also showed worsening spikes with maxing the gains from the mic inputs)
https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/10.%20Inst%20Input%209%40Max%20Gain%20--%20spikes%407%2614kHz.png
https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/11.%20Inst%20Input%2011%40Max%20Gain%20--%20spikes%403.4%267%2614%2617kHz.png
https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/12.%20Inst%20Input%2012%40Max%20Gain%20--%20spikes%20allover.png
https://dl.dropbox.com/u/326608/RME%20UFX%20Support/Noise%20Issues/13.%20Inst%20Input%2012%40Max%20Gain%20%2B%20Other%20Mic%20Inputs%40Max%20Gain%20--%20spikes%20allover.png

I don't have the time right now to go through the Line Inputs but they do also have some issues with "real life" measurements.

8 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 24,922

Re: UFX: Defective Line and Mic Input Sections?

What SNR do you see in TM FX on the line inputs? RMS and +3 dB active please, nothing going on else, nothing connected.

Is this a new unit? You have a serial number?

Regards
Matthias Carstens
RME

MC's Website

  • knoll
  • ADAT user
  • Offline
  • Registered: 2012-07-30
  • Posts: 13

Re: UFX: Defective Line and Mic Input Sections?

I bought the unit in June 2011. S/N 23200809.

Unplugged all cables except power and usb. Made Total Reset for TM FX. With RMS Level checked in View Options and RMS +3dB enabled in Preferences the Line Inputs through 1 to 8 show value of -105 most of the time sometimes quickly peaking higher.

-------- Continuing analysis ---------
If I switch all Line Inputs to Lo Gain the RMS values fluctuate between -104 and -103. The frequency response looks like the screenshot #3 in the first post.

Further more, if I at this point plug mics to Input 9-11 and guitar to Input 12 and crank their gains to max, the RMS values for Line Inputs are about between -93 and -91. The frequency response looks like the screenshot #4 in the first post. Plugging another unit to the Line Inputs doesn't change the situation except raising the overall noise floor about 2dB.

If I now unplug the other unit and change the Line Inputs back to +4dBu the RMS values for Line Inputs are about at -102. (Mic/Inst gains are still at max)

10 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 24,922

Re: UFX: Defective Line and Mic Input Sections?

If you connect nothing to the unit except USB or FW, have the line I/Os set to +4, a sample rate not higher than 48 kHz, then you must see the values that we give in the manual. Around 110 dB RMS unweighted, 113 dBA (see DIGICheck). 105 would be a defective unit.

Regards
Matthias Carstens
RME

MC's Website

  • knoll
  • ADAT user
  • Offline
  • Registered: 2012-07-30
  • Posts: 13

Re: UFX: Defective Line and Mic Input Sections?

I was previously using sample rate of 96kHz. Here are the values for the line inputs with 48kHz.

1. Line I/O at +4dBu:
This seems quite close to the values you provided but not totally there. Would you call this normal or abnormal?
DigiCheck RMS LB [dB+3] around -108.5 (input 1/2)
DigiCheck RMS [dBA+3] around -112.1 (input 1/2)

----------- more analysis ----------------

Continuing from the above situation.

2. Changing Line I/O to -10dBV:
yields the following values. This is what I expect, the noise floor gets a bit higher.
DigiCheck RMS LB [dB+3] around -106.6 (input 1/2)
DigiCheck RMS [dBA+3] around -110.0 (input 1/2)

3. Changing Line I/O to Lo Gain:
The results are the opposite of what I expect. The noise floor gets higher and being worse than at -10dBV. TM values seemed to fluctuate here quite a lot, not so much in DigiCheck.
This can't be normal, right? Shouldn't Lo Gain have lower noise floor because of being "lower gain".
TM between -107 and -103
DigiCheck RMS LB [dB+3] between -106.4 and -105.8 (input 5/6)
DigiCheck RMS [dBA+3] around -109.3 (input 5/6)

4. Keeping the Line I/O at Lo Gain. Enabling Inst and setting gain to 65 for Inputs 9-12:
This is what I don't understand, why the line inputs are affected so much in Lo Gain setting? Should they be? There is about 15dB difference compared to the previous case (3.). Values for line inputs are:
DigiCheck RMS LB [dB+3] around -90.2 (input 5/6)
DigiCheck RMS [dBA+3] around -94.2 (input 5/6)

5. Changing Line I/O back to -10dBV:
For Lo Gain line inputs there was huge difference depending on inputs 9-12 at min vs at max+inst enabled (about 15dB). For -10dBV there is almost no difference compared to case 2, only about 0.5dB. This difference seems quite normal to me.
DigiCheck RMS LB [dB+3] around -106.1 (input 1/2)
DigiCheck RMS [dBA+3] around -109.7 (input 1/2)

6. Changing Line I/O back to +4dBu:
Only minimal difference here also compared to case 1, only about 0.2dB. Difference seems normal to me.
DigiCheck RMS LB [dB+3] around -108.3 (input 1/2)
DigiCheck RMS [dBA+3] around -112.0 (input 1/2)

So could it be that the line inputs are otherwise ok except on lo gain settings? What about mic inputs (based on previous posts)?

12 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 24,922

Re: UFX: Defective Line and Mic Input Sections?

1. 109.0 dB / 112.5 dBA.

2. 107.5 dB / 111.0 dBA.

3. 109.1 dB / 112.6 dBA.

4. No influence (same as 3).

In other words: the internal voltage stabilization in that unit does not work correctly, and you get all kinds of noise spread throughout the unit depending on what you activate. Sorry, that is a defective unit.

Regards
Matthias Carstens
RME

MC's Website

  • knoll
  • ADAT user
  • Offline
  • Registered: 2012-07-30
  • Posts: 13

Re: UFX: Defective Line and Mic Input Sections?

Ok, I contacted my reseller. I'll be sending the unit for a repair within the next two weeks.