DiyMobileAudio.com Car Stereo Forum banner

1 - 20 of 20 Posts

·
Registered
Joined
·
1,743 Posts
Discussion Starter #1
I have been trying to set time delay properly on my setup for what seems like an eternity. I have tried several different methods and honestly I do think I am very close to perfect. One of my original trials was to use Arta to do time based measurements (impulse) but I was way over my head then and kind of scrapped it for other methods. Fast forward to today and alot more knowledge from both self trial and a ton of reading on here. I want to attempt those Arta readings to verify if I have my time alignment set correctly or not. I guess I have a few questions that I have not found answers to on here. The big one is what kind of a signal do you use for the measurement? I have always used pink noise for all measurements. Is there some other recording that should be used. 2nd. I am using external exciteation for my measurements using my Mobile Pre USB/Behringer 8000 mic combo. When I attempted this yesterday, I couldn't make out any large initial peaks. It really looked like quite a mess actually. I didn't take any screenshots but I could if needed. Anyone have any thoughts on this?

I have also been looking at spending yet more money on a new setup for this. So far I have what I started out with as listed above and since gotten the Binaural mics and Andrea USB which appears to have been a waste of money. The latter setup was pretty much useless for me on anything above around 4 kHz it seems and had no accuracy on the low end. I don't know if I used that right either tho honestly. Now, that the background has been set, I am now looking at the Dayton Omnimic setup. Does this give much better results than my current MObile Pre/ECM 8000 setup for getting time based measurements and frequency response? I would love any info that people that have used it could give. I have read the thread on it here and the reviews on PE.

I thank anyone that has the info I am looking for.

James.
 

·
Premium Member
Joined
·
4,235 Posts
First, there's noting wrong with the mobile pre and the 8000. Second, if you use ARTA, the program must provide the stimulus signal--the program has to know what it sends and when in order to have a baseline to determine the transfer function (what hapens to the signal when it passes through the system and then through the ir to the mic). read on the ARTA site. It's explained there.
 

·
Registered
Joined
·
162 Posts
Just picked up the Dayton Audio OmniMic Precision Measurement System from parts express.

Here is a link to the Daytons website: Dayton Audio OmniMic Precision Measurement System - Test & Measurement

In there How to section they describe an easy way to use impulse response to help set Signal delay. I think it may work for a starting point before tuning by ear.

Here is the info:
Align two speakers in "time of arrival" at a listening position
Arrange your system so that the Test Track sound plays out of both speakers you are working with, but with any other speakers silent. Choose the Frequency Response page and play the indicated track with the microphone set at the desired position. Set the control that is near the top to "blended" and observe the Impulse Response display. If there is a time of arrival difference between the two speakers, you should be able to see two impulse peaks. The one to the left (near the 0 msec point) is from the speaker with the earlier delay time, the other is for the later one. If you have an electronic delay unit, adjust the delay applied for the earlier speaker so the peaks become superimposed. If you have more than two speakers you wish to align together, choose the latest (the one making the right-most peak) to be the reference and delay the others to align to its peak
 

·
Registered
Joined
·
1,743 Posts
Discussion Starter #4
First, there's noting wrong with the mobile pre and the 8000. Second, if you use ARTA, the program must provide the stimulus signal--the program has to know what it sends and when in order to have a baseline to determine the transfer function (what hapens to the signal when it passes through the system and then through the ir to the mic). read on the ARTA site. It's explained there.
First, thanks for the feedback and response Andy. I must ask, why does Arta offer for external excitation for Impulse stimulus in this instance then? Ok, so I need the program to send the signal, that means calibrating the Mobile PRE then?How does one do that if that is needed then (ie. what cables are needed exactly)? Also, how does one get the Arta signal into the stereo system so as to be able to test the response/time domain?

I have only used ARTA only for frequency response. I have found two different ways as well in Arta. The way I was doing it before was with the 1/3 octave RTA feature. I could never get it to sound right tho. I went back to the main FR1 screen which looks the same but seems to give quite a different response reading. What are the differences between these two screens and which works better for what we are trying to do in the car environment?
 

·
Registered
Joined
·
1,743 Posts
Discussion Starter #5
Just picked up the Dayton Audio OmniMic Precision Measurement System from parts express.

Here is a link to the Daytons website: Dayton Audio OmniMic Precision Measurement System - Test & Measurement

In there How to section they describe an easy way to use impulse response to help set Signal delay. I think it may work for a starting point before tuning by ear.

Here is the info:
Align two speakers in "time of arrival" at a listening position
Arrange your system so that the Test Track sound plays out of both speakers you are working with, but with any other speakers silent. Choose the Frequency Response page and play the indicated track with the microphone set at the desired position. Set the control that is near the top to "blended" and observe the Impulse Response display. If there is a time of arrival difference between the two speakers, you should be able to see two impulse peaks. The one to the left (near the 0 msec point) is from the speaker with the earlier delay time, the other is for the later one. If you have an electronic delay unit, adjust the delay applied for the earlier speaker so the peaks become superimposed. If you have more than two speakers you wish to align together, choose the latest (the one making the right-most peak) to be the reference and delay the others to align to its peak
Yes, that is what I read as well. It sounds like there is no calibration needed of sound cards and the system being measured is the one that plays the signal. This is what had me most interested in grabbing this setup. I just wish it wasn't so expensive or I wish it came out about 1.5 years ago so I wouldn't have already wasted more than that on setups that are apparently not giving me the results I am looking for. If it is possible to set this up to give me what I want it will likely be alot cheaper. I am just unsure how to get the program to send the signal through my stereo and don't know if I need to do some form of calibration on the Mobile PRE USB.
 

·
Premium Member
Joined
·
4,235 Posts
Oh, with some analyzers, you can use an external signal, but it would also have to be fed to the second channel of the analyzer so the transfer function is the difference (well, not exactly the difference in the mathematical sense of difference if difference is one minus the other), but you know what I mean.
 

·
Registered
Joined
·
1,743 Posts
Discussion Starter #7
Oh, with some analyzers, you can use an external signal, but it would also have to be fed to the second channel of the analyzer so the transfer function is the difference (well, not exactly the difference in the mathematical sense of difference if difference is one minus the other), but you know what I mean.
Arta can actually be set to one or two channel readings. I have it set to one channel as I am only using one mic. When I tried this yesterday I had two speakers playing at same time and that might have been the issue. What excitation signal should be used for such a reading anyways? I have been using recorded pink noise from my dvd player but is there a different recording I shoud be using. I read something about a pistol shot or balloon recording to be used while scanning through the Arta manual? I don't even know what these are - please don't think I am ignorant here:blush: If there is some other signal I should use can you (or anyone) give me a link to it? I guess I could buy a stereo mini to rca and plug the Mobile PRE into one of my Bit One's aux input rcas. Would that be the best idea here and use the generator in Arta for the sweeps? Is there anything I should know here too as to not blow up any speakers?
 

·
Registered
Joined
·
19 Posts
I use HolmImpulse instead of ARTA, but have a few suggestions.

1. Try it out in a large room first. All the early reflections in a car make it much more difficult. It's much easier to get your procedure down in a more controlled environment first. Then try attacking the nasty car environment.

2. Here's a helpful link: HOW TO: Measure using HOLM and ARTA - AVS Forum

3. You can get good T/A results with just a tape measure and a stand placed where the center of your head is positioned using these steps:

a. Measure and record the distance of each speaker you want to T/A. The speaker farthest away is your "baseline" speaker. All of the other speakers need to be delayed to match the distance of the "baseline" speaker.

b. For each speaker: calculate the distance difference between this speaker and your "baseline" speaker. Example: The "baseline" (farthest away) speaker is 65 inches and "speaker A" is 50 inches. Difference = 15 inches.

c. Convert the distance into time. The speed of sound is approx. 13,500 in/sec. So T/A time = distance difference/ 13,500. Example: 15 inches / 13,500 inches/sec = 1.11 msec. This is your T/A time for this speaker.

d. If at all possible, measure the tweeter distance to its faceplate and other speakers distance to their dust cap. This is a rough approximation to the speaker's "acoustic center" (i.e. more accurate than measuring to the outside grill).

I've used the above procedure and compared it to what my Alpine PXE-H650 calculates during its "Multi-EQ". I'm almost always within 0.1 - 0.2 msec. (1.35 - 2.7 inches) of the H650. If nothing else it will provide values to do a "sanity check" when using ARTA or HolmImpulse.

Happy Tuning....
 

·
Registered
Joined
·
1,743 Posts
Discussion Starter #9
I use HolmImpulse instead of ARTA, but have a few suggestions.

1. Try it out in a large room first. All the early reflections in a car make it much more difficult. It's much easier to get your procedure down in a more controlled environment first. Then try attacking the nasty car environment.

2. Here's a helpful link: HOW TO: Measure using HOLM and ARTA - AVS Forum

3. You can get good T/A results with just a tape measure and a stand placed where the center of your head is positioned using these steps:

a. Measure and record the distance of each speaker you want to T/A. The speaker farthest away is your "baseline" speaker. All of the other speakers need to be delayed to match the distance of the "baseline" speaker.

b. For each speaker: calculate the distance difference between this speaker and your "baseline" speaker. Example: The "baseline" (farthest away) speaker is 65 inches and "speaker A" is 50 inches. Difference = 15 inches.

c. Convert the distance into time. The speed of sound is approx. 13,500 in/sec. So T/A time = distance difference/ 13,500. Example: 15 inches / 13,500 inches/sec = 1.11 msec. This is your T/A time for this speaker.

d. If at all possible, measure the tweeter distance to its faceplate and other speakers distance to their dust cap. This is a rough approximation to the speaker's "acoustic center" (i.e. more accurate than measuring to the outside grill).

I've used the above procedure and compared it to what my Alpine PXE-H650 calculates during its "Multi-EQ". I'm almost always within 0.1 - 0.2 msec. (1.35 - 2.7 inches) of the H650. If nothing else it will provide values to do a "sanity check" when using ARTA or HolmImpulse.

Happy Tuning....
I actually do exactly that right now for time alignment and I think I am very close but I am looking for perfection and I want a program (like Arta - just what I am familiar with) to give me numbers to see how off I am if at all. I use a Bit One so all I have to do is measure for the mids and tweets. It is the sub that is the tricky one or rather the group delay of the separates. For the longest time I think I had way too much time delay on everything. I might still too and what is why I want something to tell me if it is right as is or not. If only the Bit One had an auto tune feature...

Thanks for the link but I was very disappointed to not see anything about Arta even tho the guy said he would do up a tutorial for it after Holm. I tried Holm a couple years ago and it frustrated me too much. That was when I went to Arta. It seemed to give me the best results and was free to use.
 

·
Registered
Joined
·
1,090 Posts
I finally got set up to do these measurements with my external Tascam USB interface, connecting the line out into RCA inputs on my home receiver (and soon my car receiver), and Behringer EMC8000 mic connected to the input port.

I still need to mess around, but I've downloaded and installed HOLM. What do the output tones sound like that are used to determine the arrival times? Frequency sweep?
 

·
Registered
Joined
·
1,743 Posts
Discussion Starter #11
I finally got set up to do these measurements with my external Tascam USB interface, connecting the line out into RCA inputs on my home receiver (and soon my car receiver), and Behringer EMC8000 mic connected to the input port.

I still need to mess around, but I've downloaded and installed HOLM. What do the output tones sound like that are used to determine the arrival times? Frequency sweep?
This is one of the largest things I would like answered as well. What type of excitation is to be used for this?-- pink noise, white noise, a constant specific frequency, a sine sweep, a chirp, a constant pinging ( I think my home setup uses this). Also if not pink or white noise a place to obtain those signals would be great too. I might try using the generator in the program too if it really does give better result.
 

·
Moderator
Joined
·
18,605 Posts
Man, where to start...

First off, the issue with Binaural being junk above 4khz is user error. Look up HRTF (head related transfer function) and you'll probably then have a better understanding of why you think that method sucks. I like it. But I don't use it anymore since I used to simply because I lost my andrea usb.


This is one of the largest things I would like answered as well. What type of excitation is to be used for this?-- pink noise, white noise, a constant specific frequency, a sine sweep, a chirp, a constant pinging ( I think my home setup uses this). Also if not pink or white noise a place to obtain those signals would be great too. I might try using the generator in the program too if it really does give better result.
What you want is an impulse response. Like Andy said, read aRTA's manual and you'll understand what the purpose is.
You'll also have to have a loopback system, like Andy said, which is essentially a means of sending signal out and getting it back in a manner that allows the software to tell you how long it takes to receive the acoustical signal (not the electrical signal from the feedback loop).
You'll use that to line up the first peaks and that correlates to arrival times being the same. The lower the frequency, however, the longer the wavelength which makes setting alignment for subs and midbass a bit more complex. A simple way to get around this is to increase the LPF to something that gives you higher frequency output (ie: 800hz-1000hz). It's not the best way, but it helps and is close enough for car audio. I put more emphasis on level matching for subwoofers than I do phase.

The omnimic is what I use for most of my testing. It's simple and quick. However, it also costs about $150-200 more than alternatives out there (ie: free software + usb/xlr adapter + calibrated mic). I still use aRTA now and again. I don't really think there's one great all-in-one solution.
 

·
Registered
Joined
·
1,743 Posts
Discussion Starter #14
Thanks Erin. I have read that and a number of others but it has been a while too. If I remember that guide was for TrueRTA which I won't pay for. I still have not found anywhere what the preferred stimulus is for timed measurements tho. Yes, I know it must be an impulse measurement but what signal type should be used or does it matter? I am strongly considering the Omnimic still. In your opinion will this be a smart purchase for me? I will have to trial the bi wheals again. I do enjoy this stuff. I did try an experiment today too. I put a couple towels on my dash and WHOA! What a difference. Things centered perfectly and my tweeters seemed to vanish. I only had time for about 1song tho so I will have to test more. Initial thoughts are I NEED a dash mat. I will probably make a new thread later for info on advice for type and pros and cons for material types.
 

·
Moderator
Joined
·
18,605 Posts
that tutorial doesn't use TrueRTA. I believe he uses winMLS. You can't do impulse measurements with true (though, they do have an "impulse" sweep, IIRC... not the same thing as an impulse measurement, though).

Regarding the stimulus, again, read the manual. It's a sine sweep.
Here's another good tutorial specifically using aRTA:
http://audio.claub.net/tutorials/FR measurement using ARTA.pdf
I found that using google: arta impulse. You may find other useful links the same way.

Omnimic: depends on what you're trying to do and how much you want to be involved and how easy you need the setup to be. I like it, but I bought it to use for home audio measurements and the kicker for me was the extra features it gives you over some measurement suites plus the fact that it's incredibly quick and easy to use. I tune friends' setups (both car and home) and like the ease of use. It's nice to take a mic, usb cable and my netbook and be up and going in a minute vs. making the correct loopback connections. But, then again, it depends on what I need to do. Some days it's easy for me to recommend it. Other days it's not. It depends on the user and what they're goals are.

I suggest you play around with aRTA a bit more to understand just what measurements types are available and what they tell you before you spring $250 for the omnimic measurement rig.
 

·
Premium Member
Joined
·
6,034 Posts
I have been trying to set time delay properly on my setup for what seems like an eternity. I have tried several different methods and honestly I do think I am very close to perfect. One of my original trials was to use Arta to do time based measurements (impulse) but I was way over my head then and kind of scrapped it for other methods. Fast forward to today and alot more knowledge from both self trial and a ton of reading on here. I want to attempt those Arta readings to verify if I have my time alignment set correctly or not. I guess I have a few questions that I have not found answers to on here. The big one is what kind of a signal do you use for the measurement? I have always used pink noise for all measurements. Is there some other recording that should be used. 2nd. I am using external exciteation for my measurements using my Mobile Pre USB/Behringer 8000 mic combo. When I attempted this yesterday, I couldn't make out any large initial peaks. It really looked like quite a mess actually. I didn't take any screenshots but I could if needed. Anyone have any thoughts on this?

I have also been looking at spending yet more money on a new setup for this. So far I have what I started out with as listed above and since gotten the Binaural mics and Andrea USB which appears to have been a waste of money. The latter setup was pretty much useless for me on anything above around 4 kHz it seems and had no accuracy on the low end. I don't know if I used that right either tho honestly. Now, that the background has been set, I am now looking at the Dayton Omnimic setup. Does this give much better results than my current MObile Pre/ECM 8000 setup for getting time based measurements and frequency response? I would love any info that people that have used it could give. I have read the thread on it here and the reviews on PE.

I thank anyone that has the info I am looking for.

James.
Couldn't you just use a ruler?

In other words -

#1 - Crossover filters insert delay outside of their passband
#2 - Unless you have a really byzantine setup, you're probably using your speakers in their passband

Therefore:

A plain ol' ruler should suffice.

Sound travels 13.5" in a millsecond. Delay accordingly.
 

·
Registered
Joined
·
1,486 Posts
Couldn't you just use a ruler?
that will get you close but for whatever reason I have ALWAYS had more success actually measuring the impulse response. I assume this has something to do with the reflective nature of the car environment but I can't prove it. Try both methods and see what you like best.
 

·
Registered
Joined
·
632 Posts
I finally got set up to do these measurements with my external Tascam USB interface, connecting the line out into RCA inputs on my home receiver (and soon my car receiver), and Behringer EMC8000 mic connected to the input port.

I still need to mess around, but I've downloaded and installed HOLM. What do the output tones sound like that are used to determine the arrival times? Frequency sweep?
great combo for the money!
I work with them with amazing results..
 

·
Registered
Joined
·
1,486 Posts
And I thought I was the only one running the tascam
 
1 - 20 of 20 Posts
Top