Feature Request: Distortion vs Power

[nyt]

Hi,

A great feature REW seems to be missing compared to some other software is graphing distortion against a stepped signal level. It would be nice to have something like the stepped sine measurements with start/stop signal level fields as well as a stop THD%. Signal duration in ms or cycles may also be uesful. Fields for dbfs -> voltage and load resistance could be used to calculate wattage.

 

[John Mulcahy]

I've added that to the list, but will not get to it until some time next year.

 

[nyt]

Great news. If you need any testing lmk

 

[nyt]

Also in the same realm, would be nice to have an option to display signal level as watts or volts if db->voltage and/or resistance is specified...

 

[nyt]

I've added that to the list, but will not get to it until some time next year.

Any thought on displaying signal level in volts / watts in RTA view?

 

[John Mulcahy]

It's on my list to look at

 

[nyt]

Any chance on implementing this in the near future? Thanks

 

[dc2bluelight]

It would be useful for some test situations to have the level you set on the generator also set the measurement chirp level. The reverse is already true.

And if that were an option, then a "Measure Now" button that triggers a measurement without going through the setup/options screen first would be nice. My clicker finger would be glad to avoid as my clicks as possible.

A one-click SPL meter cal would also be a nice option, bypassing the cal result/confirmation windows. It takes 3 clicks to cal the meter now, one would be excellent.

You see where this is going. It's not me being lazy, I'm really hurting my clicker finger with so many clicks!

Possible bug: I noticed that sometimes the SPL meter won't indicate an input even though it has one and the meter is turned on. Closing and re-opening the meter fixes that. The level meter conflicts with the SPL meter of course, and the level meter wins. I don't use the level meter much, but use the SPL meter and re-cal it quite a bit, usually with the Generator window up.

Combining the SPL meter and the generator in a single "test bench" window would be really slick...and avoid clicking to focus windows.

The next one is pretty out there...have an option to generate a group of closely-spaced HF tones, more than two, more like 20-30 or more. The technique is covered in Deane Jensen's "Spectral Contamination" AES paper. He used a programmable arbitrary waveform generator and an ancient hardware FFT plus some filtering to reveal distortion behavior that is not generally detectable with THD and IMD. I'd like to experiment with his technique using modern hardware and REW, but the limitation is being able to generate the multi-tone test signal. But I realize that one may not even make it to the list, but I thought I'd toss it out there.

Thanks for all your hard work!

 

[John Mulcahy]

Possible bug: I noticed that sometimes the SPL meter won't indicate an input even though it has one and the meter is turned on. Closing and re-opening the meter fixes that. The level meter conflicts with the SPL meter of course, and the level meter wins.

Can you explain more explicitly what you mean by "won't indicate an input"? What behaviour do you see and with what audio source? The level and SPL meters are not in conflict, it is fine to run both.

You see where this is going. It's not me being lazy, I'm really hurting my clicker finger with so many clicks!

The keyboard shortcuts can help with reducing clicking, can also use the space bar to activate the currently selected button - the one you are likely to want is usually set as the default. I'll look at the possibilities for fewer steps though.

Multitone generation and analysis are on my list to look at at some point.

 

[John Mulcahy]

Any chance on implementing this in the near future? Thanks

Meaning displaying signal level in volts / watts in the RTA? Possibly, when I added the dBc option I did some preparatory work for further axis options. Can you elaborate on exactly what you would like to see and where you think the scaling from dBFS to V and the reference impedance for watts should be specified? Do you mean V/W or dBV/dBW?

 

[dc2bluelight]

Can you explain more explicitly what you mean by "won't indicate an input"? What behaviour do you see and with what audio source? The level and SPL meters are not in conflict, it is fine to run both.

The sequence that messing things up is, Generator on, Level Meter on, then start a measurement. When it's done the generator is back on at the previous settings, the level meter is on, but the SPL meter is out of cal, sometimes by a lot. Closing the level meter, turning the SPL off then back on clears the issue.

This image shows what happens after a measurement. Everything was spot-on perfectly calibrated (In Level meter, SPL dBFS and dB were all calibrated to the same reading) and running fine before the measurement. This is what I get post-measurement. The dB meter reading seems to be random, but always a much lower reading than what it was calibrated for. I've seen it in the -100+dB range. Again, closing the Level Meter, stopping and starting the SPL meter cures the issue.

(Windows 7, latest updates, Motu 4Pre, line out to line in)

The keyboard shortcuts can help with reducing clicking, can also use the space bar to activate the currently selected button - the one you are likely to want is usually set as the default. I'll look at the possibilities for fewer steps though.

Thanks! I also might consider one of those little programmable USB button stacks. But fewer steps is always nice.

Multitone generation and analysis are on my list to look at at some point.

Thanks!

 

[John Mulcahy]

The sequence that messing things up is, Generator on, Level Meter on, then start a measurement. When it's done the generator is back on at the previous settings, the level meter is on, but the SPL meter is out of cal, sometimes by a lot. Closing the level meter, turning the SPL off then back on clears the issue.

Thanks. It isn't that the meter is out of cal, but the data feeding it stops being updated. I've fixed it for the next beta.

 

[nyt]

Meaning displaying signal level in volts / watts in the RTA? Possibly, when I added the dBc option I did some preparatory work for further axis options. Can you elaborate on exactly what you would like to see and where you think the scaling from dBFS to V and the reference impedance for watts should be specified? Do you mean V/W or dBV/dBW?

Sorry for the delay in response, didn't get a notification or anything.

You could add the conversion options under the settings in the RTA window. It'd be nice to be able to display as volts or watts instead of db in the RTA view, and maybe even the graph view. You'd obviously need a db->v conversion variable, and for wattage, a resistance value.

Ideally, I'd like to be able to use REW to create graphs like in the first post, amplitude sweeps of a specified frequency up to a specified THD+N limit. Currently, I have to use different software to do this, but would love to be able to do it all in REW. If it could be implemented like the stepped sine sweeps under RTA, except stepped amplitude sweeps, that's be excellent.

 

[John Mulcahy]

I've added dBu, dBv and dBW for the RTA and dBu, dBV and V for the signal generator, that is in the upcoming 5.20 beta 2. Understand about the stepped amplitudes, but haven't got to that yet.

 

[nyt]

I've added dBu, dBv and dBW for the RTA and dBu, dBV and V for the signal generator, that is in the upcoming 5.20 beta 2. Understand about the stepped amplitudes, but haven't got to that yet.

Awesome, can you add a standard V and W (non decibel) option?

tyvm

It's very useful when performing tests with the rtx-6001 that was developed over at diyaudio. 0dbfs = .1v up to 100v depending on the setting, so having that display actual voltage or wattage saves a step.

 

[John Mulcahy]

You will still need to recalibrate REW whenever you change the input sensitivity on the RTX-6001, REW has no visibility of the sensitivity changes being made in the RTX.

 

[nyt]

You will still need to recalibrate REW whenever you change the input sensitivity on the RTX-6001, REW has no visibility of the sensitivity changes being made in the RTX.

Of course. Is it a cal procedure or can you just enter a value somewhere?

 

[John Mulcahy]

Apply an input and enter a value to tell REW what voltage is being applied.

 

[nyt]

Apply an input and enter a value to tell REW what voltage is being applied.

Is it possible to just be able to enter the voltage somewhere for a 0dbfs signal?

 

[John Mulcahy]

I'll support both.

 

[nyt]

I'll support both.

Looking good so far. I notice if I change the value voltage in the tone generator and navigate away, the value stays, but it is not updated until I hit enter in the text input field for it.

 

[John Mulcahy]

I'll add some code in the next beta to update it when focus is lost.

 

[nyt]

I'll add some code in the next beta to update it when focus is lost.

Thanks... also, not sure there's a way to work around it with how the audio interfaces are handled, but if you change the output channel in the generator, it stops the rta.

 

[John Mulcahy]

For ASIO the audio streams have to be stopped, but I can automatically restart them afterwards, I'll do that in the next beta.

 

[bmx]

. Currently, I have to use different software to do this, but would love to be able to do it all in REW. If it could be implemented like the stepped sine sweeps under RTA, except stepped amplitude sweeps, that's be excellent.

Hello nyt and John,
I would also love the stepped sine sweep option. I'm using REW to not only test room acoustics, but to balance my Left and Right channels of my Analog "Two Bus". Referring to stereo linked Eq's and compressors.

nyt: What is the software you currently use to measure stepped sin distortion?

Many thanks,
bmx

Edit: Sorry for the noob post! I didn't realize stepped sin measurements were included. Found them in the RTA. Thank you !