Topic: MicroTrack gain measurements

[it-goes-to-eleven]

I just made some gain measurements of the MT's line stage. I also tried to measure how much gain is added by the push buttons.

My tests focused on the 1/4" inputs and the 'H' setting (Line in mode). I collected some M mode data but the file doesn't seem to be here.. My CF card may have filled up.


Summary

On the lowest gain setting, the MT adds a minimum of 14.81 dB of gain.

The clip light seems to come on at -1.0 dB.

In Line-in mode, each press of the level button adds about 0.5 dB.

The meters of the MT are divided by 10 bars.  Generally, each bar represents 1dB.  But each button press does Not move the same amount across the bar scale.

When the signal level hits the 10th bar, you're probably clipping.


In the table below, bar 1.0 is having the level setting arrow tip aligned with the first bar.  It is not possible to align the tip on every bar (3.0).

Line in mode
Gain setting vs. dB

Bars   dB
0    14.81
1.0  15.25
2.0  16.27
2.8  17.23
3.5  18.25
4.0  18.76
4.25 19.21
4.75 19.67
6.0  21.17
6.9  22.17
8.2  23.66
8.9  24.69
9.1  25.14
9.5  25.64


Now that we have a reference for how much gain the line stage adds, it will be much easier for others to test the Medium gain setting.


Test detail

I played a 440 hz sine.wav from my PC via USB to UA5's RCA jacks and into my 722 via an XLR adapter.  I set the 722 gain at 0.0 dB and adjusted the UA5 output signal level until the -20dB on the 722 (while in "Peak" mode) lit.

Analyzing that file in Wavelab, the peak level was -19.83 dB.

I fed this signal into the 1/4" inputs of the MT while in line-in mode.  I noted the gain setting and time offset of the recording. I then analyzed each gain level in wavelab.

A second test file was created where I set the UA5 signal strength so that the VU level on the MT would be at 5 bars when the MT gain was set to the minimum.  That peak signal level from the UA5 was calibrated with the 722 and wavelab at -27.13 dB.


Weird bug

As I continued to increase the gain once the clip light came on, the level portion of the meter began to flash.  Once further gain was added, the flashing stopped and the bar stayed unshaded, as if no signal was present.

[Colin Liston]

Weird bug

As I continued to increase the gain once the clip light came on, the level portion of the meter began to flash.  Once further gain was added, the flashing stopped and the bar stayed unshaded, as if no signal was present.

I've noticed this too.  When getting too hot a signal the meter disappeared, " as if no signal present"

What's with that?

[prof_peabody]

+T

[udovdh]

Line in mode
Gain setting vs. dB

Bars   dB
0    14.81
1.0  15.25
2.0  16.27
2.8  17.23
3.5  18.25
4.0  18.76
4.25 19.21
4.75 19.67
6.0  21.17
6.9  22.17
8.2  23.66
8.9  24.69
9.1  25.14
9.5  25.64

This looks like the 12 dB of analog gain in the ADC being used.
So there is gain outside of the ADC or there is already digital gain involved..
I could live without the 14.81 dB of gain. Maybe a mod is in sight?

The clip bug you mention is indeed very strange! Others can reproduce? (My MT is in the country but not yet on my doorstep)

[MattD]

A testing suggestion: generate noise for this test instead of a single frequency. This will eliminate any frequency-specific artifacts.

[udovdh]

Did anoyone here have a look at the signal path for the line ins?
Figure out part of the schematic?
The ADC has max of 12 dB internal analog gain.
The ADC also has from 4:1 tot 1:1 attenuation depending on external resistors. (via the L/M/H switch?)
The ADC has the possibility of digital gain (which we don't want to use).
The ADC is at 0dbFS with 1Vrms. You do the math for your situation what amplification you want/need.

It looks like there is extra gain outside of the ADC. Does anyone here know how that is implemented?

[it-goes-to-eleven]

I just did some tests with the new 1.05 firmware.  Sadly, the 1/8 still does not support line mode.

Good suggestion on the noise Matt, I used it this time.


                  gain
              min      max
              ---      ---
1/4 line in   +14      +26
1/8 Line in   +29      +41
1/8 M         +41      +49


Someone asked about sending a +4dB line signal into this thing.  With it adding +14, I'm not sure that would work well.   But I haven't tried it.

[udovdh]

Someone asked about sending a +4dB line signal into this thing.  With it adding +14, I'm not sure that would work well.   But I haven't tried it.

Hmm. Time for a mod in that area?
MSOP opamps maybe, but they need resistors to set the gain?
Anyhone have a look at the internals of this device?

[udovdh]

Someone asked about sending a +4dB line signal into this thing.  With it adding +14, I'm not sure that would work well.   But I haven't tried it.

The ADC has an attenuator whth takes 4Vrms and with a resistor divider made of 47K and 12K makes it +/- 1Vrms.
If there's an extra 14 dB and maybe different resistors this could work.

I'd rather have no 14 dB and less attenuation, of course. Please post info on what part of the PCB has the 14 dB gain.

[Do.com]

This may be a stupid or obvious question, but I haven't found a comparison that is what I am looking for.  I currently run my rig optical in on my JB3 with the gain set to 0db.  I do my levels on the SBM1.  Seems to work quite well.  I now want to use my Micro Track for a stealth show Sunday night but am unsure how the levels on the MAFT match with the 0db gain level on the JB3 when going digital in.  I will be going SPDIF in on the MAFT so do the levels even matter or is it just for the analog inputs?  Help?

Clint

[Brian Skalinder]

I will be going SPDIF in on the MAFT so do the levels even matter or is it just for the analog inputs?

When running digital-in to the recorder, the levels are controlled upstream - at either the ADC or preamp (or both).  In your case, on the SBM-1.  So...set levels with theSBM-1, just as you've done previously, and you're all set.

[Do.com]

+T   

Thanks Brian - as always your help is appreciated. 

Clint

[it-goes-to-eleven]

As to a +4dB board signal, NFW!

I ran a board patch into the 1/4" in line mode last night.  As you'd expect  from a device juicing things with +14, it was too hot.  Unfortunately, an inexperienced board operator didn't know how to reduce the levels until I figured it out for the second set.   No big deal, it was just a test and wasn't my primary rig. But it was still a lost board..

[udovdh]

As to a +4dB board signal, NFW!

I ran a board patch into the 1/4" in line mode last night.  As you'd expect  from a device juicing things with +14, it was too hot.  Unfortunately, an inexperienced board operator didn't know how to reduce the levels until I figured it out for the second set.   No big deal, it was just a test and wasn't my primary rig. But it was still a lost board..

I guess we put the +14 dB (or no +4dB signal capability) in as a bug?