US-144 Calibration Question...

**facesnorth** · **01-10-09, 02:53 AM**

I don't really know how this works or what makes a good device for use with REW, but that ASUS line sure looks good. Why did you get the US-144 if you have that ASUS? why not just add the xenyx802 for cheaper? Just so you could use your laptop and not have to move your desktop over by the a/v equipment? (that was my rational)

are you using vista 32 or 64? what other uses do you plan for the us-144? and why did you choose the us-144 over similar devices like the EMU 0404 or m-audio fastrack pro? thanks

**etc6849** · **01-10-09, 12:29 PM**

Well, my computer fans are a little noisy on the HTPC. I was hoping my laptop would work so I could have more portability and quieter operation. The US-144 is a fine card as the calibration will account for any deviations from 0dB that you see on my plots. You can always run a plot after calibration with a loopback cable to see; just clear the mic calibration file first though.

From Bruce's plots even if you have a very nice card (his looks even flatter than the HDAV1.3), the pre-amp will drop it down a little anyways. There are better and cheaper cards than the HDAV1.3 in terms of analog performance, I bought it because it's one of the only HDMI cards to bitstream TrueHD. If you look at some reviews, there are several cards with even better S/N ratios.

My laptop issues are probably my laptop. It's pretty beat up and the dvd drive doesn't even work. Personally, I don't think I would go with the ASUS line. I've had too many HDMI handshaking issues. If you are looking for a soundcard for an HTPC, wait a few more months until the new HDAV version comes out and drivers are not in beta.

Quote:

facesnorth wrote:

I don't really know how this works or what makes a good device for use with REW, but that ASUS line sure looks good. Why did you get the US-144 if you have that ASUS? why not just add the xenyx802 for cheaper? Just so you could use your laptop and not have to move your desktop over by the a/v equipment? (that was my rational)

are you using vista 32 or 64? what other uses do you plan for the us-144? and why did you choose the us-144 over similar devices like the EMU 0404 or m-audio fastrack pro? thanks

**BigPines** · **01-10-09, 06:11 PM**

I didn't realize this discussion was going on and I posted regarding a similar issue with the M-Audio FastTrack Pro: http://www.hometheatershack.com/foru...tml#post139665.

I am still trying to figure out if I have accounted for my mic pre-amp in my calibration procedure. I have been trying to understand everything that has been said here but I am at a disadvantage because I am not an electrical engineer. Here is the scoop:

The FastTrack Pro has two XLR and 1/4" TRS combo jacks as the only inputs. Page three of the manual says:

"Microphone/Instrument Inputs (Mic/Inst) – These Neutrik hybrid connectors will each accept a low-impedance mic level signal on a standard three-pin balanced XLR or TRS plug, or a high-impedance instrument level signal on an unbalanced 1/4” TS plug."

Page fourteen of the manual has the detailed specs:

Mic Inputs (A/D):
Input Impedance: 2.7k Ohms unbalanced, 5.4k Ohms balanced
Maximum Input Level: from +24dBu @ min gain, pad on to -40dBu @ max gain, no pad
Preamp Gain: >40dB
Pad: -20dB pad

Line Inputs (A/D):
Input Impedance: 20k Ohms balanced and unbalanced
Maximum Input Level: +4.1dBu balanced/+1.9dBV unbalanced
Pad -20dB pad

Line Outputs (D/A)
Output Impedance: 150 Ohms unbalanced, 300 Ohms balanced
Maximum Output Level: +1.8dBV, unbalanced/+10.1dBu, balanced

So I am still confused. Given that I am going from the balanced 1/4" TRS outputs to the balanced 1/4" TRS inputs in the loopback without using the available pad attenuator or instrument switch, it would seem as though I am including the mic preamp but after reading this thread I just don't know.

Since it is the same physical input, I am confused about how I can know if I am using the input as a 1/4" balanced Mic input or a 1/4" balanced line input. I thought as long as I was not using the instrument level switch on the input, I was using the low-impedance mic level input.

Any help would be much appreciated.

Mike

**DrWho** · **01-10-09, 09:27 PM**

Do you have any other USB devices plugged into your laptop? Every USB device needs to authenticate itself at a certain power level, which in the USB2.0 spec I believe is only one of two power levels. Your laptop's USB bus can only provide so much power, so each device is allocated only so much power. If you plug in an additional device asking for more than what isn't authenticated, then that device will not authenticate.

As far as the noise, that can only happen if your gain structure is incorrect (not sure what you're doing when you hear the noise). If the device has the 100 whatever dB dynamic range it claims, then it just means you're operating towards the bottom of the dynamic range of some device in the chain. You could also be imparting noise by forcing the laptop to attempt to provide power that it can't.

Btw, I measured -3dB at 3Hz with the unbalanced out to unbalanced in....am I reading correctly that you're showing -3dB at like 20Hz? Or is your measurement showing the XLR input?

**DrWho** · **01-10-09, 09:33 PM**

Quote:

facesnorth wrote:

What needed to be soldered? I wonder if this is something I can do. I don't understand all the talk about padding and what to do with this thing once you make it. It sounds very unelegant which is exactly what I liked about this unit. Plus it sounds imprecise... Is there a way to be confident that I can get this thing properly calibrated and be able to trust my measurement graphs?

You do not need to pad or solder anything. I don't know why the discussion went down that path. Just attenuate the output so that you don't clip your input.

Technically you lose dynamic range when you do this, but the extra noise is in no way going to affect acoustic measurements in a normal listening environment. I hope my mentioning this doesn't turn this into an issue too - I'm just trying to be complete.

**etc6849** · **01-11-09, 06:07 PM**

Thanks Dr. Who,

You have found an issue with my US-144 set up in vista, but I don't know what's wrong. I really think I am using the software correctly, but yes the unbalanced line in/out seemed to fall near 20hz at -3dB as did the XLR padded loopback. This was on a vista pc. However, when I did the test on an XP dell laptop, -3dB fell around -4.5hz with the XLR padded loopback cable.

I'm quite confused now. REW said my levels were fine when I did the US-144 plot. The ASUS card has a response close to -3db @ 3hz so I think I am using REW correctly. Is there some set level I should have the US-144 set to when I do a calibration measurement? IE, should I have the line in @ 50% and the output at 100%?

Quote:

DrWho wrote:

Do you have any other USB devices plugged into your laptop? Every USB device needs to authenticate itself at a certain power level, which in the USB2.0 spec I believe is only one of two power levels. Your laptop's USB bus can only provide so much power, so each device is allocated only so much power. If you plug in an additional device asking for more than what isn't authenticated, then that device will not authenticate.

As far as the noise, that can only happen if your gain structure is incorrect (not sure what you're doing when you hear the noise). If the device has the 100 whatever dB dynamic range it claims, then it just means you're operating towards the bottom of the dynamic range of some device in the chain. You could also be imparting noise by forcing the laptop to attempt to provide power that it can't.

Btw, I measured -3dB at 3Hz with the unbalanced out to unbalanced in....am I reading correctly that you're showing -3dB at like 20Hz? Or is your measurement showing the XLR input?

**etc6849** · **01-11-09, 07:44 PM**

I'm by no means an audio person, so I don't know how much I can help.

I can say that I have also tried an m-audio buddy (little preamp) with the my vista computers soundcard. For the audio buddy I decided to try an XLR to rca cable without padding to see what would happen and I put this cable between mic input on the preamp and the soundcards right channel output. I also put a cable from the output of the preamp to the input of the soundcard.

I had to set the computers line in to 15% and the mic level to about 20% to keep the necessary headroom when I used a non-padded cable (ie cable without resistors). The calibration data was close to when I used the padded cable Bruce helped me with, so I'm not sure if you need to do it or not. I had to adjust level a lot lower to avoid using the non-padded cable and almost couldn't do it, even with REW output set to -20dB.

In contrast with the padded cable, I set the computer line in to 100%, REW soundcard calibration to -14dB, and the m-audio buddy to 60% mic level. The only thing I noticed about the pre-amp versus the ASUS HDAV1.3 soundcard with a standard loopback going from in to out is that there is -15dB difference at very low frequencies under 10Hz, so I'm not sure it matters unless you are trying to measure frequencies under 15Hz.

Quote:

BigPines wrote:

I didn't realize this discussion was going on and I posted regarding a similar issue with the M-Audio FastTrack Pro: http://www.hometheatershack.com/foru...tml#post139665.

I am still trying to figure out if I have accounted for my mic pre-amp in my calibration procedure. I have been trying to understand everything that has been said here but I am at a disadvantage because I am not an electrical engineer. Here is the scoop:

The FastTrack Pro has two XLR and 1/4" TRS combo jacks as the only inputs. Page three of the manual says:

"Microphone/Instrument Inputs (Mic/Inst) – These Neutrik hybrid connectors will each accept a low-impedance mic level signal on a standard three-pin balanced XLR or TRS plug, or a high-impedance instrument level signal on an unbalanced 1/4” TS plug."

Page fourteen of the manual has the detailed specs:

Mic Inputs (A/D):
Input Impedance: 2.7k Ohms unbalanced, 5.4k Ohms balanced
Maximum Input Level: from +24dBu @ min gain, pad on to -40dBu @ max gain, no pad
Preamp Gain: >40dB
Pad: -20dB pad

Line Inputs (A/D):
Input Impedance: 20k Ohms balanced and unbalanced
Maximum Input Level: +4.1dBu balanced/+1.9dBV unbalanced
Pad -20dB pad

Line Outputs (D/A)
Output Impedance: 150 Ohms unbalanced, 300 Ohms balanced
Maximum Output Level: +1.8dBV, unbalanced/+10.1dBu, balanced

So I am still confused. Given that I am going from the balanced 1/4" TRS outputs to the balanced 1/4" TRS inputs in the loopback without using the available pad attenuator or instrument switch, it would seem as though I am including the mic preamp but after reading this thread I just don't know.

Since it is the same physical input, I am confused about how I can know if I am using the input as a 1/4" balanced Mic input or a 1/4" balanced line input. I thought as long as I was not using the instrument level switch on the input, I was using the low-impedance mic level input.

Any help would be much appreciated.

Mike

**DrWho** · **01-11-09, 11:39 PM**

Quote:

etc6849 wrote:

Thanks Dr. Who,

You have found an issue with my US-144 set up in vista, but I don't know what's wrong. I really think I am using the software correctly, but yes the unbalanced line in/out seemed to fall near 20hz at -3dB as did the XLR padded loopback. This was on a vista pc. However, when I did the test on an XP dell laptop, -3dB fell around -4.5hz with the XLR padded loopback cable.

I'm quite confused now. REW said my levels were fine when I did the US-144 plot. The ASUS card has a response close to -3db @ 3hz so I think I am using REW correctly. Is there some set level I should have the US-144 set to when I do a calibration measurement? IE, should I have the line in @ 50% and the output at 100%?

That almost sounds like a latency issue maybe? Did you have REW set to a sample rate of 44.1kHz? I don't think the Tascam does 48kHz (but that wouldn't explain a 2 octave shift).

If I'm remembering correctly, I run the line out at 12-o'clock and both line-ins at 3-o'clock. But really, the actual volume doesn't matter so much as long as you're above the noise floor. I run it at the settings I do because I get a full volume from the microphone at 3-o'clock as well. In other words, I don't think that is the problem.

I didn't get the chance to measure the XLR input last week, but I'm hoping to get around to it sometime this week. Maybe I will detail the process and throw it up on my website - dunno if that will clear things up or not.

**brucek** · **01-12-09, 07:33 AM**

Quote:

Is there some set level I should have the US-144 set to when I do a calibration measurement?

Take at look at the REW scope results after the measurement and be sure it looks OK.. (i.e. proper input output levels, no clipping, etc)

bruce

**etc6849** · **01-14-09, 12:56 PM**

Quote:

brucek wrote:

Take at look at the REW scope results after the measurement and be sure it looks OK.. (i.e. proper input output levels, no clipping, etc)

bruce

What a practical way to check output/input levels. I guess I need to become more familar with the software because I didn't even think about looking at this!

Thanks,

Ellery

**etc6849** · **01-14-09, 01:12 PM**

Thanks Dr. Who. I've been busy these last few days, but I'll check to see what I have the sampling frequency set to. I remember the US-144 being set to 44.1khz in vista, is 48khz better? What about using 96khz or 192khz (the asus card will do this)?

If I understand the theory correctly, a sample rate of 44.1khz should be fine for frequencies upto 21.55khz right (nyquist theorum, about all I remember from a linear systems)? Are you saying that 44.1khz is not as good as 48khz if the DACs sample at 96khz? I'm pretty confused on this issue. However, I always wondered why receivers have 192khz DACs though when DVD's and most blu-rays are only sampled at 48khz/channel. I'll have to read up on this when I find some time... I'm assuming the added bandwidth leads to a better S/N ratio?

Cheers,

Ellery

Quote:

DrWho wrote:

That almost sounds like a latency issue maybe? Did you have REW set to a sample rate of 44.1kHz? I don't think the Tascam does 48kHz (but that wouldn't explain a 2 octave shift).

If I'm remembering correctly, I run the line out at 12-o'clock and both line-ins at 3-o'clock. But really, the actual volume doesn't matter so much as long as you're above the noise floor. I run it at the settings I do because I get a full volume from the microphone at 3-o'clock as well. In other words, I don't think that is the problem.

I didn't get the chance to measure the XLR input last week, but I'm hoping to get around to it sometime this week. Maybe I will detail the process and throw it up on my website - dunno if that will clear things up or not.

**DrWho** · **01-14-09, 11:15 PM**

I mentioned the 44.1kHz issue because I was getting weird results with the 48kHz setting, but that was on the very first day I plugged it in and I was still messing with the driver. I should see if I can get 48kHz to work.

The only advantage to higher sampling rates is that the lowpass filters become easier to implement. You don't usually see a significant reduction in the noise floor because most ADCs and DACs are way oversampling at rates like 300kHz+ internally. You really don't want to go above 96kHz for the data though because you lose bit accuracy and it's harder to maintain clock accuracy as you go higher in frequency (since your PLL loops have gotta have way higher bandwidths to keep the jitter low).

As a measurement tool, 44.1kHz is totally sufficient as the passband ripple is less than 0.1dB (or very close). I doubt very much that the microphone is that accurate, but even if it were, there are no speakers that will be anywhere close....and it's not like the big problems that need fixing are going to be addressed to that accuracy either.

I've been too busy to even pull my measurement rig out of my car, but eventually I'll get around to testing everything you've been doing so we can compare notes.

**mswlogo** · **01-16-09, 07:04 AM**

I use the TASCAM US-144 with the ECM-8000 Mic and it works fine.

You have to fiddle with levels a little so that you get good dynamic range on the inputs to the US-144 (i.e. just under the lights going Red Yellow) and getting REW happy with the levels. And not driving the speakers to low or to hard. You have to set input sensitivity different for the Mic vs the feedback.

I have used it with the Left Channel feedback loop for sound card calibration.

Also when you do the feedback you use the 1/4" plug and I believe that will connect to a line level input. When you connect the Mic you use a Balanced Cable which will connect to the Mic Preamp. You also need to turn on Phantom Power.

You won't be able to calibrate out the difference between the Mic input and Line Level input but I don't think that's a huge deal.

I too had to fiddle with 44.1Khz vs 48Khz and I forget which worked better.

Because REW can't support ASIO (Yet

) you are still going through the computers Mixer which often resamples to 48Khz. I think this is the reason why 48 vs 44 makes things work well or not. Don't forget the Levels set it the PC's Mixer are also in effect.

**mswlogo** · **01-16-09, 07:31 AM**

Quote:

facesnorth wrote:

are you using vista 32 or 64? what other uses do you plan for the us-144? and why did you choose the us-144 over similar devices like the EMU 0404 or m-audio fastrack pro? thanks

I own both the EMU 0404 and the Tascam US-144. I like the TASCAM US-144 better. Under Vista-32 EMU 0404 is buggy. Under Vista-32 Tascam also has issues with 1.12/1.122 Driver/Firmware but 1.11/1.11 works good. The 1.12/1.12 Firmware/Driver was released to support Vista-64 and don't know how stable it is under 64bit OS. When I say not stable with 1.12/1.12 on US-122 it will hang playing music once every few hours. EMU hangs once every few songs. Both are easy to recover and just annoying.

I use TASCAM as Analog Phantom Powered Mic Preamp for Merdian Room Calibration with ECM-8000.
I use it for REW.
I use it as an SPDIF output into Meridian Digital Speakers (BitPerfect using ASIO) using Winamp.
I use it as a HeadPhone amp using Winamp (again with ASIO and plays 24/88.2 files just fine with no resampling).
I use it to record 24/88.2 SACD out of a PlayStation3 with a Modified CLUX11SA (taps SPDIF off HDMI).
I use it with an FM Tuner to feed Meridian SPDIF input only speakers.

I really like the TASCAM being USB powered.

I also bought the Creative Audigy and the M-Audio units and returned both. Real junk in my opinion. Might be fine for REW but not for a lot of other duties.

**DrWho** · **01-17-09, 10:46 PM**

Oh cool, I didn't even realize the 144 could do ASIO. Where did you get your winamp driver?
Posted via Mobile Device

**mswlogo** · **01-19-09, 12:17 PM**

Quote:

DrWho wrote:

Oh cool, I didn't even realize the 144 could do ASIO. Where did you get your winamp driver?
Posted via Mobile Device

From here http://otachan.com/out_asio(exe).html

It inlcuded source and I downloaded the ASIO SDK and rebuilt it and fixed a few small bugs. Turns out most of my problems were the latest Vista driver from TASCAM. The 1.11 driver firmware is much more stable.

I'll put it up the cleaned up ASIO winamp plugin on my own site when I get some time. The main bug I fixed was it would crash when switching sample rates (in gapless mode).

http://www.meridianunplugged.com/wik...DSP%20Speakers

**The Bogg** · **06-09-09, 05:45 PM**

I just got the US-144 to use on a laptop with Win XP SP3. When I try to calibrate the soundcard it doesn't give anything close to a flat frequency response. Above about 600hz it starts to oscillate and is off by 3db or more at higher frequencies. I'm only using it for 400hz and under but still wonder why this is the case. I'm probably doing something wrong but I've followed the instructions in the help file to a T.
I upgraded to the latest 1.12 driver and firmware. Anyone have any suggestions? I tried initially doing the loopback with a cable with rca connector on one end and xlr on the other and I used the mic input. Since the result was off I suspected it was my use of a hybrid cable. I got the rca to trs adaptor and use the regular line in but the same result...ie. not even close to flat from 20hz to 20khz. Any suggestions would be appreciated!

**brucek** · **06-09-09, 06:54 PM**

Quote:

I tried initially doing the loopback with a cable with rca connector on one end and xlr on the other and I used the mic input. Since the result was off I suspected it was my use of a hybrid cable. I got the rca to trs adaptor and use the regular line in but the same result...ie. not even close to flat from 20hz to 20khz. Any suggestions would be appreciated!

The cable should be an RCA to TS adapter, not TRS.

brucek

**The Bogg** · **06-09-09, 11:25 PM**

Quote:

brucek wrote:

The cable should be an RCA to TS adapter, not TRS.

brucek

I meant it was a phono to 1/4" mono jack - the same one that was listed in the help file.

**brucek** · **06-10-09, 07:51 AM**

You do need to connect from line-out to line-in on a single channel only and be in stereo mode with any effects or mixers in the application software turned off.

If you've done all that, then it should work. Try adjusting the input level down a bit when setting up the soundcard calibration routine to see if that helps the oscillation.

brucek

**The Bogg** · **06-12-09, 08:55 PM**

Quote:

brucek wrote:

You do need to connect from line-out to line-in on a single channel only and be in stereo mode with any effects or mixers in the application software turned off.

If you've done all that, then it should work. Try adjusting the input level down a bit when setting up the soundcard calibration routine to see if that helps the oscillation.

brucek

Thanks for your reply. I do have only the left line-out to line-in. I'll recheck the other settings.
In any case, it seems linear enough from a few hz to 500hz which is really what I'm interested in
but it is a bit puzzling...

Advertisement