AVS › AVS Forum › Audio › DIY Speakers and Subs › HOW TO: Measure using HOLM and ARTA
New Posts  All Forums:Forum Nav:

HOW TO: Measure using HOLM and ARTA - Page 4

post #91 of 155
4/6/10 at 1:21pm
Most dimmers are evil. They inject stuff into the power line and that can get into your amps, etc.
Reply
Reply

AVS Top Picks

post #92 of 155
4/28/10 at 8:06am
I've been playing around some with measurements, using the recipe from the first posts here..

But for some reason all my measurements end up having some crazy phase response like this..:



Does anyone know what is causing this ?
Reply
Reply
post #93 of 155
4/28/10 at 9:03am
I had similar problems with a SB Audigy sound card. Replacing the card fixed the problem. I think what happens is that when the mic picks up the tone from the speaker, it is remixed and sent out to the amp again. I tried changing many mixer settings, but the SB card would not cooperate.
Reply
Reply
post #94 of 155
4/28/10 at 9:12am
Quote:
Originally Posted by Myggpower View Post

I've been playing around some with measurements, using the recipe from the first posts here..

But for some reason all my measurements end up having some crazy phase response like this..
Does anyone know what is causing this ?

Use the response smoothing feature.
Reply
Reply
post #95 of 155
4/28/10 at 9:37am
Thread Starter 
That definitely looks like noise in the measurement. It could be the soundcard.
Reply
Reply
post #96 of 155
4/28/10 at 10:12am
Try turning all sliders in the system volume "advanced controls" to 50%.

You can turn the master volume, wave, and line-in up to ~70-75% max. I would not go over 50% with a mic input. Muting all unused I/Os can help some cards a lot.
Reply
Reply
post #97 of 155
4/28/10 at 10:16am
Thread Starter 
I wonder if he has run the calibration?
Reply
Reply
post #98 of 155
4/28/10 at 11:16am
I didn't run calibration, nor did I use smoothing..

So the phase issue has been taken care of.

It's starting to look like something now.. Only bass is lacking.. Got a really nasty peak at ~60-80hz.. but that is mostly because of my retarded room. Hoping a P1152 placed properly will take care of that, when I get it..
Reply
Reply
post #99 of 155
4/28/10 at 6:23pm
Quote:
Originally Posted by kbgl View Post

I had similar problems with a SB Audigy sound card. Replacing the card fixed the problem. I think what happens is that when the mic picks up the tone from the speaker, it is remixed and sent out to the amp again. I tried changing many mixer settings, but the SB card would not cooperate.

All the Audigy soundcards can be set to send the line-in to the speakers, or not...it's not the easiest thing to change until you've learned the mixer, but I've had no problems doing measurements on both Audigy 1 and Audigy 2 sound cards.

Of course, going to a laptop with a USB mixer that provides the mic with phantom power is a ton easier...
Reply
Reply
post #100 of 155
4/28/10 at 7:52pm
Quote:
Originally Posted by MBentz View Post

All the Audigy soundcards can be set to send the line-in to the speakers, or not...it's not the easiest thing to change until you've learned the mixer, but I've had no problems doing measurements on both Audigy 1 and Audigy 2 sound cards.

Of course, going to a laptop with a USB mixer that provides the mic with phantom power is a ton easier...

I'm not going to say that it would be impossible, but if I couldn't figure it out in a week, it's unlikely that the card wasn't the problem.
Mine was an Audigy SE using XP. Twenty minutes after installing a different card, I had everything working correctly.
Reply
Reply
post #101 of 155
5/18/10 at 1:10pm
This message is for penngray or anyone else that would like to answer if possible.

When you say: "This is our first measurements and its usually done on the tweeter alone so that we can set the zero time for the impulse. This is used to figure out the time delays between the tweeter and the woofer."

This is done by only changing the start frequency and you leave the woofer connected?
Reply
Reply
post #102 of 155
5/18/10 at 3:05pm
Thread Starter 
The idea behind that is to get the tweeter impulse to be set to zero.

You run the measurements separately,

1. connect the tweeter alone. Measure. Then click on time zero locked (click on the "Use") but.
2. Measure the tweeter again.
3. Disconnect the tweeter, connect the woofer, measure.
Reply
Reply
post #103 of 155
5/18/10 at 3:26pm
How do I maintain the proper load on my amplifier if I disconnect the woofer from my crossover and measure only the tweeter?
Reply
Reply
post #104 of 155
5/18/10 at 4:50pm
Thread Starter 
Quote:
Originally Posted by scitizen17 View Post

How do I maintain the proper load on my amplifier if I disconnect the woofer from my crossover and measure only the tweeter?

Why do you care about the load on your amp?

You only need to keep the same gain structure for all measurements. (ie do not touch any volume levels between all measurements)

There isnt any important reason to know the absolute SPL Value of either the tweeter or the woofer. The only importance is knowing the difference in dBs between the tweeter and the woofer.

Are you drivers very different loads? I guess that will need a Jig setup so that the load is constant.
Reply
Reply
post #105 of 155
5/18/10 at 5:15pm
Okay that makes sense, I'm with you, thanks. Pardon me for having a senior moment.

Quote:
Originally Posted by penngray View Post

Why do you care about the load on your amp?

You only need to keep the same gain structure for all measurements. (ie do not touch any volume levels between all measurements)

There isnt any important reason to know the absolute SPL Value of either the tweeter or the woofer. The only importance is knowing the difference in dBs between the tweeter and the woofer.

Are you drivers very different loads? I guess that will need a Jig setup so that the load is constant.
Reply
Reply
post #106 of 155
5/18/10 at 5:28pm
Thread Starter 
Quote:
Originally Posted by scitizen17 View Post

Okay that makes sense, I'm with you, thanks. Pardon me for having a senior moment.

No, you made a good point.

Measurement rigs are going to handle speaker loads accurate. I have posted nothing on those.

Im coming from an active design position and load issues are never a concern.

I do not have the answer on how XO design works when the load is 8 ohms on a woofer and 16 ohms on a tweeter. I will wait for others to explain that better.
Reply
Reply
post #107 of 155
5/18/10 at 5:59pm
My crossover design includes a Zobel compensation network for the woofer, so the parallel resistor in the Zobel should suffice as the temporary load at the low wattage power used for the measurement. Thanks.

Quote:
Originally Posted by penngray View Post

No, you made a good point.

Measurement rigs are going to handle speaker loads accurate. I have posted nothing on those.

Im coming from an active design position and load issues are never a concern.

I do not have the answer on how XO design works when the load is 8 ohms on a woofer and 16 ohms on a tweeter. I will wait for others to explain that better.
Reply
Reply
post #108 of 155
5/22/10 at 11:59am
I performed the measurements on my bass-reflex loudspeaker according to the steps outlined in pengray's writeup at the beginning of this thread. I may have a soundcard or calibration issue. I have a Realtek HD on my HP laptop. I performed the calibration and it seemed to go fine. I'm using a Dayton EMM-6 mic and loaded the calibration file. I checked the boxes for calibration. The bass response seems much better than the plot indicates, and I have no idea what the rise in the tweeter plot is at around 350Hz. I set the start frequency for the tweeter plot at 500Hz. As a test, I connected a signal generator and when I pump 350Hz into my tweeter, there is hardly anything at all generated. As I approach the crossover frequency for the tweeter, the response seems to come up where it should. My crossover is a passive 2-way 12dB Linkwitz-Riley alignment at 2100Hz. It also appears as though my calculated crossover frequency is wrong.

I meant to mention that I am using a Behringer phantom power supply in-line with the mic and a Shure impedance matching transformer to input the signal to my mic line-in.

Any ideas? I have attached the plot.

Thanks,

Scott
LL
Reply
Reply
post #109 of 155
5/23/10 at 11:30am
I read somewhere, in this thread I think, that the mic level may be on in the playback properties. Mine was, and signal was being routed from the mic to the speakers during my measurement. I'm sure now the the readings I took are flawed, I will plot these again.


Quote:
Originally Posted by scitizen17 View Post

I performed the measurements on my bass-reflex loudspeaker according to the steps outlined in pengray's writeup at the beginning of this thread. I may have a soundcard or calibration issue. I have a Realtek HD on my HP laptop. I performed the calibration and it seemed to go fine. I'm using a Dayton EMM-6 mic and loaded the calibration file. I checked the boxes for calibration. The bass response seems much better than the plot indicates, and I have no idea what the rise in the tweeter plot is at around 350Hz. I set the start frequency for the tweeter plot at 500Hz. As a test, I connected a signal generator and when I pump 350Hz into my tweeter, there is hardly anything at all generated. As I approach the crossover frequency for the tweeter, the response seems to come up where it should. My crossover is a passive 2-way 12dB Linkwitz-Riley alignment at 2100Hz. It also appears as though my calculated crossover frequency is wrong.

I meant to mention that I am using a Behringer phantom power supply in-line with the mic and a Shure impedance matching transformer to input the signal to my mic line-in.

Any ideas? I have attached the plot.

Thanks,

Scott
Reply
Reply
post #110 of 155
5/23/10 at 4:28pm
Thread Starter 
Please plot again but press zoom in on the impulse chart and set a gate on it also.
Reply
Reply
post #111 of 155
5/23/10 at 4:43pm
Quote:
Originally Posted by penngray View Post

Please plot again but press zoom in on the impulse chart and set a gate on it also.

I'm not sure I understand exactly how or where to set the gate. Will you please help me on that aspect?
Reply
Reply
post #112 of 155
5/23/10 at 4:44pm
Thread Starter 
HOLM is pretty easy with gating...there is a "dotted" line on the impulse that you click on and drag out further to figure out your gate.

if you zoom in on your impulse (button below the chart) you can see the small ripples that kind of stop then begin again. That is the point where your room starts to effect your response.
Reply
Reply
post #113 of 155
5/23/10 at 4:45pm
What exactly am I gating, and where do I set it?
Reply
Reply
post #114 of 155
5/23/10 at 4:49pm
Thread Starter 
I have family stuff to do right now, sorry.

I will explain gating tomorrow.
Reply
Reply
post #115 of 155
5/23/10 at 4:52pm
Quote:
Originally Posted by penngray View Post

I have family stuff to do right now, sorry.

I will explain gating tomorrow.

Absolutely. Thanks, no big rush.
Reply
Reply
post #116 of 155
5/24/10 at 12:24pm
I have been experimenting with the gating and I can see that moving it definitely changes the response curve. I am measuring the speaker in a large room with acoustical tile ceiling, and ceramic tile floor. The mic is placed 1 meter from the speaker and aimed between the woofer and tweeter. The speaker is about 40 inches above the floor. I assume ceramic tile floor is not the best for accurate measurements, but does the fact that the speaker is 40 inches off the floor and the mic is 1 meter away negate the floor effects?
Reply
Reply
post #117 of 155
5/24/10 at 3:47pm
Quote:
Originally Posted by scitizen17 View Post

I have been experimenting with the gating and I can see that moving it definitely changes the response curve. I am measuring the speaker in a large room with acoustical tile ceiling, and ceramic tile floor. The mic is placed 1 meter from the speaker and aimed between the woofer and tweeter. The speaker is about 40 inches above the floor. I assume ceramic tile floor is not the best for accurate measurements, but does the fact that the speaker is 40 inches off the floor and the mic is 1 meter away negate the floor effects?

The height above the floor is the limiting factor on the gating if your speakers are about 3 or more feet from any walls. You should be good at 700 hz or maybe a little lower. My speaker is on a 24" stand.


Try measuring from 2 meters and compare.

Reply
Reply
post #118 of 155
5/24/10 at 3:59pm
Quote:
Originally Posted by scitizen17 View Post

How do I maintain the proper load on my amplifier if I disconnect the woofer from my crossover and measure only the tweeter?

Disconnect the input to the woofer or tweeter section of the x-over. Do not disconnect at the driver.
Reply
Reply
post #119 of 155
5/24/10 at 4:23pm
Quote:
Originally Posted by kbgl View Post

Disconnect the input to the woofer or tweeter section of the x-over. Do not disconnect at the driver.

Ah ha, now that may make a difference although a little more difficult being as though my crossover is hardwired. I'll give it a try. I guess I'm learning that with a new design you need to build these crossover connection points so that they are easily tapped into or not.
Reply
Reply
post #120 of 155
5/25/10 at 3:31am
Thread Starter 
Quote:
Originally Posted by scitizen17 View Post

Ah ha, now that may make a difference although a little more difficult being as though my crossover is hardwired. I'll give it a try. I guess I'm learning that with a new design you need to build these crossover connection points so that they are easily tapped into or not.

Generally all the individual driver meaurements are done before any XOs are built so we usually just connect to the individual drivers. Im measuring the Tannoy V8 now to build new XOs and I had to remove the existing XO from it.
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs

AVS Top Picks

AVS › AVS Forum › Audio › DIY Speakers and Subs › HOW TO: Measure using HOLM and ARTA