Speaker workshop measurement jig help

[GrahamT]

*Problem is fixed now, thanks for the help guys!


Hi fellas,

I've done the calibrations, and impedance measurements. Everything works beautiful.

Now I want to do some acoustic measurements but before I plug everything in could you take a look at my setup. I'm not great at setting this stuff up and dont want to burn my Audigy2.

Here is what I need to build:

https://web.archive.org/web/20040926133948/http://planeta.terra.com.br/educacao/claudionegro/english/swacoustic/cables/cables.html

Image not available

This is what I did build:



Are they the same? Will this work?

Here is the description of the circuit:

https://web.archive.org/web/20050415133929/http://planeta.terra.com.br/educacao/claudionegro/english/

Thanks in advance.

[cjd]

Looks right to me!

Just make sure you don't have bogus connectors that short to ground. I had to do some custom work on mine!

C

[GrahamT]

Thanks Chris

[GrahamT]

I just hooked it up but as soon as I turn the amp on, the driver starts moving at around 20 Hz. Barely audible but the driver is moving quite a bit. What could this be. I obviously wont be able to run any sweeps until I get rid of this.

I'm stumped.

[GrahamT]

I ran a near field test.

I dont think it went very well. This is a Paradigm 8" woofer in a sealed box.

It's still making the noise.

This is the result of the near field test.


Any suggestions are welcome.

[GrahamT]

Okay, I rechecked all the cables, didn't find any problems.

What I am getting is a high pitch constant tone, my guess is it is pretty close to a constant 1000 Hz signal. Then, if I unplug any cable and then plug it in again I get a signal pretty close to 20 Hz. These tones are quite loud. They alternate occasionally but one of them is always on.

The noise dissappears if I disconnect the cable jumper from the left Line out to the left Line in. When I only use the Line in left, no noise, when I only use the Line out no noise, but when they touch they produce the noise.

One more thing. When I do the MLS and sine tests, I get volumes on the VU meters between 24k-28k on both channels. WHen I do the near field or far field tests the annoying tones go away for the duration of the test, but return after. The VU meter reads +-28k on the left channel but only 3-4k on the right channel.

Anybody have any ideas? I'm open to any suggestions.

Chris,

Can you desribe what your problem was? Maybe it is similar to mine.

[GrahamT]

I rechecked all the cables and connections again. All good. Pretty simple layout.

I tried plugging everything into the outlets with cheaters, no change.

I tried plugging everything in grounded, no change.

It must be a problem with the JIG layout, cables, SC, or how the sound card is set up. It is a CB Audigy2 24 bit. I also have a disabled AC97 on board card that I will try tomorrow.

I'm away on an engineering tour tomorrow, I'll be back late in the afternoon. Please leave any suggestions and I'll try them Sat afternoon.

Thanks in advance for any help.

[cjd]

Just thought of one thing - you're using the line in, not the mic in right?

I had a few problems. And I think I have mine wired wrong. :P d0h! (I believe I'm not using the left out only, but am using both L&R out - this should not matter WRT your problems)

1. One of the wires was touching the body of one of the plugs, shorting the signal straight to ground. Check continuity and make sure you only have it where you want it.

2. I had the input and output plugs hooked into my computer backwards so I was getting tons of feedback! Mic response was going straight in and back out through the speaker! oops.

3. I had a nifty ground loop. If your amp has a 3-prong grounded plug, lift the ground when you're using it with a computer! Computers are always grounded and MUST be. Your amp will pick up the ground from there so it'll still be grounded.

My suggestion is to first check the cable - make sure there is no short from signal to ground anywhere. Then, make sure the correct plug is on the correct port on your computer. It may be worth connecting only the signal from the L out to the L in and not ground. Could be a ground loop there though I can't really figure why that would be...

Not sure where to go from there. Hopefully you're not using a good speaker yet. 8)

C

[Dennis H]

The Audigy is one that gives measurement systems fits because it has hardwired record monitoring (playback of all input audio that is being recorded, or recording of all signals that are being played) causing a feedback loop. You may be able to disable that and make it work. See the Liberty Instruments page for more info -- no personal experience.

https://web.archive.org/web/20050315011513/http://www.libinst.com/compatibility_issues_with_praxis.htm

audigy-setup.pdf

[GrahamT]

Chris,

Yep, line in.

1. Yep, checked continuity between all the wires. Everythings kosher, no shorts to ground on the cables, maybe internally in the SC? I use the same cables in a different config for impedance and it works great. See the graphs below.

3. I have all the computer stuff grounded and the amp lifted.

I'll try lifting the grounds for the lines as you suggest. Dont know why there would be a loop there either but I'll try that.

Dennis,

Thanks for bringing that to my attention. I knew about the problem with the first Audigys but apparently the Audigy2 will work with SW. I purposely did not install that CL mixer for that reason but I will double check. In the windows mixer I disabled the "monitor when recording" option.

Here is the impedance plot Jon did for the Extremis:



Here is the impedance plot using SW:





I use most of the same cables and the same input and output and levels for impedance and acoustic measurement.

Thanks for the ideas guys. Keep em coming. I appreciate the help very much.

[GrahamT]

SON OF A VONDROOK!!!!! :B

That problem is gone.

The high and low pitch noises were because of a stupid mistake of mine. In the windows mixer, there are two Line In levels, one in Play Control and one in Recording Control. I forgot to lower the Line In level in the Play Control when switching tests. If any of the levels are too high (mine was like 1mm too high) you get a noise all the time.

I quickly threw the mic in front of the test speaker and measure the near field. The driver is an 8" poly so this response looks reasonable to me. The increasing high end is probably due to inaccuracy of the mic, or my calibration. I'll figure it out.

How does this look to you guys? Is that phase response normal?
BTW, closed box, resonant frequency 60 Hz.

Thanks again for the suggestions.:T

[GrahamT]

I played around with a few things and set everything up better.

I tried a new nearfield measurement. I think it turned out rather well this time. Well, the response looks crappy but that is what I expected, so in terms of the setup I think it is working.

Here is my 1/8th octave smoothed nearfield response with a 8" Paradigm poly cone.

[cjd]

So I'm sitting here moving my mouse wondering WHY DID MY COMPUTER JUST LOCK UP?!?!

oh yeah... that's in your screen capture! :rofl: :rofl: ops: :rofl:

How low are they crossing this beast? I hope this is in a 3-way.

C

[GrahamT]

In the design I took apart they are crossing over at 1.8k, 3rd order. It is used in 3 ways as well. My measurements may not be accurate yet though.

[GrahamT]

How does this look guys?

Far field (0.5m) ungated. The response slopes off, I'm sure some of it is baffle step loss, but maybe I am doing something wrong? Any suggestions?

[JonMarsh]

Looks like you're getting it sorted out pretty well. I was out with my daughter yesterday- car problems, looks like she's going to need a new clutch (throw out bearing is rattling and whining).

That looks like a combination of normal baffle step and a driver with several cone modes- note how diagnostic your nearfield measurement is. In comparison, look at this nearfield measurement on M8a 8" midwoofer, at center, cone edge, and mid between. This is how a pistonic driver behaves.

[GrahamT]

Thanks Jon,

I'm going to do some more measurements today. I'll try the Extremis and compare that to results from you and others. That way I can do some direct comparisions to see if I am doing things correctly.