HOLMImpulse: Measurements in practice

[DQ828]

All Drivers

Well Iv'e measured all of the speakers with no crossover, I am working towards trying out a MiniDSP setup for these speakers as a trail run until I build my new speakers.

It looks to me that I should be crossing them over at about 250Hz & 6.2kHz

Given that the speakers are too heavy & difficult to take outside, I'll just have to live with the distorted room measurements.

Is there a better way to measure the bass driver or shouldn't I bother?

I couldn't load the zip file it was too big

 

[Pano]

Polarity of the mid will depend on your crossover and distance between drivers. You can check that with the phase plot. You'll have to lock the time, tho.

It looks like you are getting there. Just bring your measurement levels up some, they are very low. Maybe you should be doing your sweeps louder? "Torturing Aliens" my wife calls it. Either that or bring the mic gain up.

Otherwise you are on the right path, bravo!
.

 

[DQ828]

Yes I agree I am getting there still lots to learn though. I will raise the volume as the setup starts to clip if I raise the gain to much. I have been sick all week & doing the mowing/gardening today almost killed me, so I dont have any energy left for the fun stuff. Hopefully I be back in the saddle tomorrow.

Thanks for your help.

David

 

[Pano]

OK David, hope you feel better.

When you say it clips, where does it clip? At the mic preamp? Or in the playback side? Obviously the digital side isn't clipping, you're 40dB down. (is your record volume set to max?)

This time of year I mow the fallen leaves, not the grass. At least the mower does a better job than the leaf blower.
Pano

 

[DougSmith]

THD measurements from in-room response

Is there any value in the THD plots for evaluating in-room response? Mine go up quite high in the low bass region, but I assume that this is influenced strongly by reverberation form the walls, etc.

 

[Altec Best]

Is there any value in the THD plots for evaluating in-room response? Mine go up quite high in the low bass region, but I assume that this is influenced strongly by reverberation form the walls, etc.

Mine do as well ! With no improvement insight ! It seems nothing works the way it should.So I'm starting to suspect the computer,as the sound card,Mic,cables,have been checked out.

 

[DQ828]

OK David, hope you feel better.

When you say it clips, where does it clip? At the mic preamp? Or in the playback side? Obviously the digital side isn't clipping, you're 40dB down. (is your record volume set to max?)

This time of year I mow the fallen leaves, not the grass. At least the mower does a better job than the leaf blower.
Pano

Its been a combination of the M-Audio Pre clipping (flashing clipping light) & last tiime I did the measurements HolmImpulse told me it was clipping & I needed to turn the volume down. I use my daughters laptop to do the measuring so I have to also fit it in around her.

I wish the leaves where falling at my place, as half my problem is extreme reaction to spring, I hate spring. If I had the money I'd spend my life moving around to avoid spring.

 

[Pano]

Well if the M-Audio is near clipping, then the volume most be turned down somewhere. You should not be getting traces at -40dB.

Do you watch the VU meter when you do a sweep? That might tell you something.

Altec Best has been having a similar low end distortion problem. I found that it was software related. Make sure the M-Audio control panel is set to 16 bits. Use the ASIO drivers if you can. I've had to reboot to clear the distortion. I think some other software (or Windoze) has a handle on the card and can cause problems. Just don't know exactly what, yet.
.

 

[speakerdoctor]

Well Iv'e measured all of the speakers with no crossover, I am working towards trying out a MiniDSP setup for these speakers as a trail run until I build my new speakers.

It looks to me that I should be crossing them over at about 250Hz & 6.2kHz

Given that the speakers are too heavy & difficult to take outside, I'll just have to live with the distorted room measurements.

Is there a better way to measure the bass driver or shouldn't I bother?

I couldn't load the zip file it was too big

Bass drivers are usually measured in the near field. That is, move the measurement mic to within about 1/4 inch of the woofer's dust cap. Turn down the gain from that used for your far field measurements.
You can splice the near field with the far field afterward.

 

[speakerdoctor]

I did the measurements again after the recalibration & it now does not look as good inverted.

It is setting the gating itself, I have not been able to work out how to set it

Should I measure on Casual Impulse? or what?

Here's how you set the gating. First, it's done after the measurement.
Then click on the auto zoom button down in the impulse response window. The smooth impulse response curve will appear and also another, dashed line curve. Click on the dash line curve with the curser on it and you will be able to move it left or right. At the same time your moving it notice the gating lines moving left and right and the shape of the response curve changing accordingly. Ideally, try to get the gating located around 200 hz. Then, the data above 200 hz will have been gated and represent a quali-anechoic response.

 

[DQ828]

Bass drivers are usually measured in the near field. That is, move the measurement mic to within about 1/4 inch of the woofer's dust cap. Turn down the gain from that used for your far field measurements.
You can splice the near field with the far field afterward.

I was told the near field is for below 100Hz, is that correct?

 

[DQ828]

Here's how you set the gating. First, it's done after the measurement.
Then click on the auto zoom button down in the impulse response window. The smooth impulse response curve will appear and also another, dashed line curve. Click on the dash line curve with the curser on it and you will be able to move it left or right. At the same time your moving it notice the gating lines moving left and right and the shape of the response curve changing accordingly. Ideally, try to get the gating located around 200 hz. Then, the data above 200 hz will have been gated and represent a quali-anechoic response.

Thank you. the major bit I was missing was the "First, it's done after the measurement." Everyone, everywhere just says set the gating, so I spent lots of time trying to work out how to set it before a measurement was taken

 

[DQ828]

Well if the M-Audio is near clipping, then the volume most be turned down somewhere. You should not be getting traces at -40dB.

Do you watch the VU meter when you do a sweep? That might tell you something.

Altec Best has been having a similar low end distortion problem. I found that it was software related. Make sure the M-Audio control panel is set to 16 bits. Use the ASIO drivers if you can. I've had to reboot to clear the distortion. I think some other software (or Windoze) has a handle on the card and can cause problems. Just don't know exactly what, yet.
.

It's probably going to be a few days before I can have another go but I will check it all then, thanks

 

[Altec Best]

Here's how you set the gating. First, it's done after the measurement.
Then click on the auto zoom button down in the impulse response window. The smooth impulse response curve will appear and also another, dashed line curve. Click on the dash line curve with the curser on it and you will be able to move it left or right. At the same time your moving it notice the gating lines moving left and right and the shape of the response curve changing accordingly. Ideally, try to get the gating located around 200 hz. Then, the data above 200 hz will have been gated and represent a quali-anechoic response.

SpeakerDoc Great Info !! Thanks, I was wondering how to set that gating as I'm fairly new (Noob) WRT measuring.As I learn more it will help me make more accurate measurements Thank You ! Now if you can help me (us) with this distortion issue.I have Napster,iTunes,Windows media player all active on my computer would they interfere with my soundcard/measuring,Holmimpulse software ?

 

[Pano]

They might, especially if you've had them open recently. That has happened to me even with other measurement software, like ARTA. Sometimes HOLM will be strange is ARTA was just used.
I suggest a reboot before you start your measurements.

(I sure wish the word measurement didn't have that stupid letter "a" in it. I have to correct my typing almost every time. "mesurement")

 

[speakerdoctor]

I was told the near field is for below 100Hz, is that correct?

Room mode effects on far field measurements kick in below 200 hz which is why the gating is done.
Nearfield measurements on the woofer simulate 2 pi measurement conditions (as if the woofer was installed flush with the ground outdoors facing up) where the room modes can't become part of the measurement because the volume is so low for the NF measurement.

 

[DQ828]

while I don't have the opportunity to do any measuring I thought I'd try & get my brain around how it works, hopefully somebody will be able to answer my questions. If you cannot answer them maybe you could point me to some reading I could do about the subject.

1. I don't understand the relationship in gating between time & frequency, in HolmImpulse, when you increase the gating in ms (more ms) it reduces the frequency level?
2. I would have thought that to get a clean measurement (no reflections) the gating would be set to just before the first reflection arrives, which can be worked out for each set-up, i.e. Reflected path minus direct path divided by 0.34m = ms
3. There must be a period of time that it takes the individual driver to get a note/frequency out to the mic. Any idea how that is determined?
4. Because my Bass driver is so close to the floor the reflections theoretically will be arriving 0.538ms after the original signal, which I assume is too early get get any worthwhile measurement.

I guess once the wavelength gets longer than the reflected path you wont get any worthwhile measurements

 

[DQ828]

while I don't have the opportunity to do any measuring I thought I'd try & get my brain around how it works, hopefully somebody will be able to answer my questions. If you cannot answer them maybe you could point me to some reading I could do about the subject.

1. I don't understand the relationship in gating between time & frequency, in HolmImpulse, when you increase the gating in ms (more ms) it reduces the frequency level?
2. I would have thought that to get a clean measurement (no reflections) the gating would be set to just before the first reflection arrives, which can be worked out for each set-up, i.e. Reflected path minus direct path divided by 0.34m = ms
3. There must be a period of time that it takes the individual driver to get a note/frequency out to the mic. Any idea how that is determined?
4. Because my Bass driver is so close to the floor the reflections theoretically will be arriving 0.538ms after the original signal, which I assume is too early get get any worthwhile measurement.

I guess once the wavelength gets longer than the reflected path you wont get any worthwhile measurements

I thought I'd try and answer my own questions

1. I don't understand the relationship in gating between time & frequency, in HolmImpulse, when you increase the gating in ms (more ms) it reduces the frequency level? Answer, The relationship is that the time in ms is the length of the particular frequency.
2. I would have thought that to get a clean measurement (no reflections) the gating would be set to just before the first reflection arrives, which can be worked out for each set-up, i.e. Reflected path minus direct path divided by 0.34m = ms. Answer, well this wasn't so much a question as a statement & I figure the statement is correct.
3. There must be a period of time that it takes the individual driver to get a note/frequency out to the mic. Any idea how that is determined? Answer, it depends on the frequency if the frequency is 2000Hz it will take the driver 0.5 of a second to produce it.
4. Because my Bass driver is so close to the floor the reflections theoretically will be arriving 0.538ms after the original signal, which I assume is too early get get any worthwhile measurement. Answer, Correct
5. I guess once the wavelength gets longer than the reflected path you wont get any worthwhile measurements. Answer, Correct

How did I go guy's?

 

[Charoenchai]

Hi everyone, I used Holmimpulse for a while and everything was OK untill yesterday. I ran the program. It showed HOLMImpulse v1.4.2.0 .... (checking for update online) .... and so on .. like everytime I did. After that, everything disappeared. I 've tried to run the program again many time but the result is the same, nothing appear. I have tried to uninstall and install the program but could not solve the problem. May anybody help me to fix this? .. Thanks.

 

[Pano]

How did I go guy's?

Well, you are about 1/2 way there.

I don't understand the relationship in gating between time & frequency, in HolmImpulse, when you increase the gating in ms (more ms) it reduces the frequency level?

No, the other way around. More time (a wider gate) gives you a lower cutoff frequency. Try it and see.

I would have thought that to get a clean measurement (no reflections) the gating would be set to just before the first reflection arrives, which can be worked out for each set-up, i.e. Reflected path minus direct path divided by 0.34m = ms

Correct! You can often see the first reflection in the impulse, if you look closely enough. Knowing about where it is can help in finding it. Just figure out how long after the direct sound the 1st reflection should arrive. Then look for it in the impulse. Don't be surprised if you are a little off.

There must be a period of time that it takes the individual driver to get a note/frequency out to the mic. Any idea how that is determined?

By the speed of sound. 1 meter = about 343Hz. But don't worry too much about that. Even a 17 meter, 20Hz wave starts changing pressure right away. That's what is important.

Because my Bass driver is so close to the floor the reflections theoretically will be arriving 0.538ms after the original signal, which I assume is too early get get any worthwhile measurement.

Yes and no. There will be some reflection, you just have to bear that in mind. But in some ways the woofer close to the floor will simply appear to be two woofers - or a big oval one. Also remember that the woofer is not a point source located 185mm above the floor. So not all of the sound energy will be coming from that one point.

I guess once the wavelength gets longer than the reflected path you wont get any worthwhile measurements

Kinda, sorta. There will be much argument about that. The trick is often to look at the driver very close, just a few cm away, then again out in the room and use your best judgment as to what is going on.