UMIK-1 or any USB MIC unsuitable for timing measurements?

[gnarly]

Under construction means one mic ahead of one speaker.

Thx Salt, still don't get it.
What does ahead of one speaker mean?

Are we talking DIY speaker building where drivers need to be measured individually, or room correction?
I've found they are vastly different, and their differences often seem to be at the root of trying to understand measurement viewpoints

 

[Salt]

Thx Salt, still don't get it.
What does ahead of one speaker mean?

Are we talking DIY speaker building where drivers need to be measured individually, or room correction?
I've found they are vastly different, and their differences often seem to be at the root of trying to understand measurement viewpoints

I mean measurement of a speaker to tune the xover. Conservatively 1 m in front of the tweeter/coax/what-ever-makes-the-highs-on-level-of-ears-in-listening-position.
No room messing ...

 

[gnarly]

I mean measurement of a speaker to tune the xover. Conservatively 1 m in front of the tweeter/coax/what-ever-makes-the-highs-on-level-of-ears-in-listening-position.
No room messing ...

Gotcha. That's my quick and easy too. Works.
I do outdoors at 3m off a deck whenever possible, for maybe an 20 ft reflection free window...(but moving that's to different discussion).....

How have you found timing repeatability, with USB mic and ARTA?
By frequency range?

 

[fluid]

For anyone who refuses to or cannot use an analogue mic to make their life a lot easier, there is an option for relative offsets between drivers which is to use the Time Lock function of Holm Impulse. This keeps the sound card driver open and counts samples. It does work and is how Earl Geddes designed and measured his speakers. I have tested it when I was making some measurements for him. It does work, it can be a bit flaky to get it working properly and Holm impulse is not as nice as REW or ARTA.

https://techtalk.parts-express.com/forum/tech-talk-forum/55182-holm-and-time-locking

 

[dcibel]

Don't do that with a USB mic...it's not reliable. Acoustic reference is a better idea. But real answer is get the right tool for the job.

 

[fluid]

Why...

 

[dcibel]

Try it and see.. run the same measurement 10 times and see if timing remains constant. Walk away for 10 min and try again.. close application and reopen and try again. If it works for you then enjoy, as I'm sure you speak from experience. My experience resulting in unreliable timing with low repeatability. Ask Earl if he uses a USB mic.

 

[fluid]

But real answer is get the right tool for the job.

Of course which is why I don't own a USB mic. I said for those who cannot or will not, and they exist which is why this keeps coming back up.

Try it and see..

All valid reasons to do use an analogue mic, there is no argument from me that a USB mic is a good idea for speaker design. The Holm method only works for relative offsets while the driver is open, that's it. A way to measure directly instead of using interference. For some that is enough, when it isn't put the USB mic down.

 

[Keith_W]

What measurement software do USB mics work with, besides REW?
Do any of them NOT use an acoustic timing reference, like REW's 5kHz up sweep?

I only have direct experience with REW and Acourate, and indirect experience with Audiolense and Dirac. Audiolense lets you use a USB mic, but recommends against it. Dirac actually recommends a USB mic. Acourate doesn't let you use a USB mic at all. Both Acourate and Audiolense have an acoustic timing reference, but done differently. With Audiolense, individual driver sweeps are done sequentially with the mic not moved between sweeps. With Acourate, you have to create your own sweep with acoustic timing reference by taking a full range sweep and muting the drivers you are not testing.

 

[sam_adams]

I made a timing reference speaker to use with REW—an old RS SD-61 tweeter in a box. A 22µF blocking cap gives an F6 of about 1400Hz. Below are ten (10) sweeps of the tweeter from 20Hz to 24KHz using the tweeter as both the DUT and timing reference. The sweeps were made over a period of twenty-five minutes, ≈ 2.5 minutes between sweeps. The mic is a Umik-2, set to 24-bit, 48KHz. Output is from the analog headphone port of an iMac.

The frequency response, all ten (10) sweeps, 0.5m, 2.83V/8Ω, unsmoothed:

The impulse responses of the ten (10) sweeps (Horizontal scale, 2ms):

The numbers in the legend at the bottom of both illustrations are the system delay (SD) in milliseconds followed by the clock adjustment (CA) in parts-per-million, so SDms/CAppm.

The differences in distance work out to be about 0.040in and the differences in samples is between 0.1 and 0.2 samples at 48KHz.

 

[Vladimir Filevski]

You have no proof...

You don't have it either.
Your conspiracy theory has the same defect as any other conspiracy theory - complicated and time-consuming cover-up... instead of simple measurement of individual calibration files (tied to the mic serial number, of course).

 

[Salt]

Gotcha. That's my quick and easy too. Works.
I do outdoors at 3m off a deck whenever possible, for maybe an 20 ft reflection free window...(but moving that's to different discussion).....

How have you found timing repeatability, with USB mic and ARTA?
By frequency range?

I mainly use sweeps with the Hypex software for active speakers, no longer building passive ones.
Arta I use only to see the impulse and step response of whole system.

 

[Emlin]

You don't have it either.
Your conspiracy theory has the same defect as any other conspiracy theory - complicated and time-consuming cover-up... instead of simple measurement of individual calibration files (tied to the mic serial number, of course).

I have no co sppiracy theory. I've already stated that I think the mics are probably individually measured. Ok?

 

[JohnPM]

For $79 retail for the UMIK-1 individually calibrated sounds like a tall order but I don't know for sure.

UMIK-1 and UMIK-2 are individually measured and calibrated. They are not batch measured, posts claiming that are baseless speculation. I visited miniDSP's facilities on several occasions back in the days I used to travel to China regularly.

 

[Blumlein 88]

UMIK-1 and UMIK-2 are individually measured and calibrated. They are not batch measured, posts claiming that are baseless speculation. I visited miniDSP's facilities on several occasions back in the days I used to travel to China regularly.

Thank you. I've no skin in the game either way. This idea of batch testing has been around more than a decade, and I've never seen anything other than rumor behind it. We don't need bad information even for a sub $100 measuring microphone.

 

[levimax]

UMIK-1 and UMIK-2 are individually measured and calibrated. They are not batch measured, posts claiming that are baseless speculation. I visited miniDSP's facilities on several occasions back in the days I used to travel to China regularly.

Thank you for clarifying. Also thank you for REW, it is a great resource especially because of your work to keep constantly improving it. Do you have a comment on the issues with making timing measurements with a USB Mic vs a 2 channel loopback analog Mic in REW ?

 

[192kbps]

Depending on your interfaces you can start with something like this as @staticV3 has suggested lots of times:

Primo EM258 Mono module, with 3.5mm plug, 1.5m

A tiny (6 mm dia) high quality Primo EM258 capsule, wired as a complete working microphone module, ready to build into a housing or enclosure. Requires capsule and cable to be supported. These rubber holders fit the capsule: Primo 6mm holder and 6 mm Microphone Holder Good ultrasonic response...

micbooster.com

View attachment 358455

Then use an adapter? to use on an audio interface?

 

[Dave Bullet]

Convenience? All those steps just to capture relative delay and major downside mentioned at the end sounds rather inconvenient to me. Not to mention crossover simulation with this method is locked to measurement distance, not actual listening distance. Maybe okay for small bookshelf speakers, but anything large will be problematic. I admit that I started into this hobby years back with an Omnimic as well following the Jeff Bagby mantra, but have since moved on years ago and never looked back. The Omnimic is just an SPL meter now, mostly collecting dust in the closet.

In any case, if you must stick with USB for convenience...you can skip a few steps. Re-calculating min-phase through HBT process simply isn't necessary, and often the primary source of error in this process. Use the measurements as-is. Just go to VituixCAD Tools Menu -> Auxiliary -> Time Align. Load your 3 measurements, un-check MP (min-phase), solve delay. Enter delay value in driver tab of main window, leave driver coordinates at 0,0,0 in the crossover.

But, with all the work that goes into a DIY speaker, why take a shortcut on arguably the most important part? Take a bit more time and gather the full spatial set of measurement data with 2-channel system properly.

Yes it is convenient (which can cover cheap in my book).

I had a ECM8000 and an Event pre-amp that died. It was a rather nice preamp by all accounts and I lacked the technical skills to diagnose and fix it. In NZ we don't have a lot of electronics support (or shipping to where it can be done would end up being the 2nd hand cost of the device itself). I was up for buying another pre-amp (and shipping) so upward of NZD $400+.

I'm aware of the downsides. That one relies on moving coil speakers being minimum phase devices AND only fixing Z in one axis (as most software doesn't allow for per axis Z). A per axis Z measurement would be far less convenient if one was to try and replicate what a 2 channel actual phase suite of measurements could achieve. For single channel measurement users, you are limited to "pick your listening axis" for Z setting. For 2 channel users, you have all this information, but like any crossover design, you have to pick your axis you optimise for (as you won't be able to solve all Z related phase issues with a single crossover).

However, I have found extremely good correlation with reality as long as you have the apply the method required.

For me, although not the topic of this thread but to put into perspective, we have the following bigger issues to contend with (and these impact both single and dual channel measurements):
1. A reflection free measurement environment - especially more difficult with 3+ way designs (and where ground plane is not possible)
2. Proper Mic mounting. A short range clip mounted mic with stand in the gated window will show high frequency aberrations that are not there.
3. Being able to to drive speakers above the ambient noise floor so enough information is collected to show HD products. Sometimes our listening spaces (family / neighbours) limit what we can do here.

 

[gnarly]

Hi again dcibel,

I just took the time (pun intended ) to make an acoustic timing ref measurement with REW. Haven't done so in nearly 10 years.
And it was taken with dual channel.
I see REW backs out all time-of-flight to the acoustic ref signal. And there's no way i can tell of knowing what the measurement distance was/is.
So thinking i see the point you were making.

My point about the 5kHz and up sweep, is yes it makes for a sharp impulse peak.... for more repeatable timing measurements,
but if and only if the speaker driver has real content 5kHz and above. I do a lot of measurement work on drivers operating below 1kHz.

 

[dcibel]

For acoustic reference, the impulse of the acoustic reference determines t=0, so the location of measurement impulse is the difference in time of flight between the acoustic reference and DUT, not the distance of DUT to mic.

For electrical loopback, impulse of the electrical loopback is same point in time as the sound leaves the speaker, so location of impulse will be always the physical time of flight of DUT to the mic, always. For this reason, instruction for VituixCAD measurement with REW suggest to place the mic 1 meter from the baffle surface, and use 2.907ms of timing offset, or 1m time of flight so measurements are captured with positioning relative to the baffle surface.

Acoustic reference speaker should be a high frequency driver like a tweeter, do not a large woofer for an acoustic reference. You can measure whatever you want, but the acoustic reference output gets set to a different output channel, connected to a high frequency capable driver.