diyAudio (
-   Multi-Way (
-   -   Speaker Workshop Distortion Question (

arc2v 17th July 2007 01:09 PM

Speaker Workshop Distortion Question
Paging all the SW power users:

I've been trying to follow the big manual, Claudio's page, and the help files to calibrate my sound card for use in SW. It's a M-Audio mobilePre.

I've gotten levels all over the place with and without clipping. In addition to the master volume, line in, and wave out, I also have gain pots on the front of the soundcard. But I've found some middle ground.

I figured I needed to get some distortion measurements to make sure I wasn't producing a lot of noise, even though the signal looked good (no clipping).

However, whenever I follow the instructions, I get a frequency/phase plot instead of a distortion plot. I'm not sure if I'm selecting the correct distortion to measure (Spectrum at 1 Power/1 Freq, vs freq at 1 Power, vs Power at 1 freq), or if I just need to reformat the output, or if it's a major bug.

But since it's shown up in several HowTo's, I imagine this works, so I think I'm doing something wrong.

Any help?

Also, I don't know if this is another bug or configuration issue, but I can never add anything to a folder I create. Everything I creats is dumped into the root folder and I have to move it manually. In all the tutorials, it looks like you can just add new items into whatever the selected folder is. Any advice here as well?

thanks in advance,

Dave Bullet 17th July 2007 08:55 PM

I've only learned how to use SW for impedance and freq response measurements so far. So can't help you there.... but will be trying soon so will battle on with you.


Also, I don't know if this is another bug or configuration issue, but I can never add anything to a folder I create. Everything I creats is dumped into the root folder and I have to move it manually. In all the tutorials, it looks like you can just add new items into whatever the selected folder is. Any advice here as well?
Does the same for me. Frustrating. Also does the same for file imports (dumps them under the root folder). I just thought that was a feature.

The source is at, so maybe if I get some time I might try and fix that one.


arc2v 17th July 2007 11:21 PM

Thanks for the response.

I'm convinced something is broken. It used to work, there's evidence of that. But it seems as though it was supposed to pop up a window with the calculations and that's it. Instead, I'm getting a chart that goes from 200Hz to 10kHz with a decreasing sine wave (it starts high, like at 1dB and slowly converges to the zero point). Weird.

Worst case, I'll configure SW to measure impedance, FR, and passive components and be done. I really wanted the distortion calculation to work, though. One of the reasons my current project is stalled is because I could not tell where the cone breakup started in the frequency response. I get a nice harsh sound, right at the crossover point now. DOH! I was hoping to use SW to actually measure it.

I'm also looking into SoundEasy, which seems like a great product and works well, and is still active and supported. However, it does not seem to play nice with my M-Audio MobilePre, so tack on a firewire sound card in addition to the price. Yikes! I have to remind myself that this is still a hobby. :)

Dave Bullet 18th July 2007 07:56 AM

Zaph uses Soundeasy. SW is limited in the types of distortion testing it can do. What is your woofer by chance? You might strike it lucky and find distortion testing results at either Mark K's site or Zaphs.

Failing that - can you measure the fundamental breakup? If so - breakups lower down can be calculated as to where they will be (but doubt the amplitude).

I think for anything more than crossover design, something more powerful than SW is required.


arc2v 18th July 2007 12:03 PM

It's a Dayton Ref Series 8", shielded:

Original design was based on their FR and Z measurements and the fact they said it had a usable frequency response through 2kHz. Turns out the breakup is around 1500 to 1800 Hz, I'm crossing at 1800, so it's noticable. Not awful, but noticable.

I'm using a 4th order L-R electrical crossover, so I figured it would hide all the high breakup effects, but to no avail. Every once in a while a voice or music note will hit it just right, though.

I must say, though, I'm getting some decent OB bass out of this thing, at least to 80Hz where I can crossover to a sub. Sadly though, I can't run the Neo3 tweeter any lower.

so with all that out of the way, I wanted to not repeat my mistake and be sure to measure drivers distortion before I built an enclosure and crossover. It looked like SW could do that, and still may for acoustic measurements. I can't seem to get it to measure the soundcard, so I'm still a bit wary.

Dave Bullet 18th July 2007 12:38 PM

When you say you can't measure the soundcard, are you talking about calibration (output impedance and capacitance) or distortion measurements?

If so - my soundcard distortion measurements (the ones Jay suggests you do when you calibrate everything) look like crap in SW too.

I presume you've searched the Audua SW forum to no avail.


arc2v 18th July 2007 12:49 PM

Yeah, I've spent some time over there. I can get the inter-channel calibration done and it actually looks good.

For the amp calibration itself, I have some confusion as to whether this should be done open or with a loop. I did both and they looked okay, I just wasn't sure which one was right.

Then I moved on to the distortion measurement (of the soundcard). I did this mostly to make sure my gain controls were set properly and that I was keeping the noise levels down.

I have not tried impedance or capacitance calibrations yet (still waiting on jig parts from Parts Express). Right now I'm just using two identical loop cables, 1/4" T-S for each in to out. Basically they're cables to connect guitar effects pedals (from my youth :D )

arc2v 18th July 2007 01:40 PM

Okay, I just spent some time reading the detailed description of speaker distortion measurements in the SW Manual (v2).

it seems I'm doing everything right, there's just a bug in the output.

First, it displays phase which is all over the place and useless for finding the peaks here. Easily disabled in the chart properties.

Second, the results are listed in dB, but the Y axis should be in %. I think I can interpret that in my head :)

With these in mind, I will have to recheck tonight and see if I get any better results.

Another option is the Right Mark Audio Analyzer, which seems to do some of these measurements and may be a better tool for determining if my soundcard is distorting.

I'm at work now, so this will have to wait until tonight, but I will report back tomorrow.

Jay Butterman really did a great job compiling all that data for the manual. It's not just a howto for the SW, it's got a lot of great references for speaker design and audio theory in general.

JMB 18th July 2007 02:09 PM

Thank you. I don't have much time to surf any more but I hope to resume soon. I compiled notes that I had taken for myself and felt that they might be helpful for others. A lot of credit should also go to Nelson and to Claudio who provided editorial support. In addition, there was input from those who wrote some of the programs.

Regarding the distortion measurements in Speaker Workshop a few comments should made. Some of the labeling is not accurate but the results were consistent in my testing. I mentioned this in the manual, I believe. I have not yet done cross reference testing with other programs to establish accuracy. Perhaps someone else out there might have.

Good luck,


arc2v 18th July 2007 02:22 PM

Yes, you did mention the mis-labeling in the writeup. I was assuming that's what was wrong, because my results for the soundcard were in the 0.01 to 0.05 range (which would make sense if it were %).

what was throwing me off was the phase measurements. I wasn't sure if it was really phase, or some other mis-labeling, or if I should just turn it off.

The last thing was the response. I wish I could post it here, but it sort of looks like a sine wave that starts up high at 200Hz and drops in both magnitude (offset from the zero point) and amplitude (the spread from the sinewave peak to peak). If I remember right, I think it was down near zero by 2 or 3 kHz.

I was expecting to see one of the "spike" charts with just a thin line at all the harmonic points.

Big thanks again to you and everyone who contributed. That truly is a great writeup. I find myself rereading sections all the time after I play around with some features. Eventually I may even build a speaker :)

I'm also working on a hybrid jig for SW. Cross between the Wallin2 and Clauio's cables. I'll post a writeup when I'm done.

thanks again to everyone for the help. I'm sure I'll get this figured out.

All times are GMT. The time now is 09:04 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio