REW now includes a wired timing reference. So if you generate your sweep file, with the timing chirp on the left channel, you can loop this back whilst measuring a USB mic. now I haven't quite got this working, as I need to support dual device inputs - one for left channel and another (USB mic) for right (something the pro edition may support).
In regards to dual channel measurement, I have found this more cumbersome and only useful on a PC with a hard wired line in / out card. That is - one PCI or motherboard etc... attached. USB devices tend to have variable timing, making a single loopback measurement useful, for finding "flight time" between left and right channels, but useless when it comes to expecting the same impulse marker between measurements.
The other issue is measuring a reflection free woofer (and if 3 way) midrange response. This usually requires quasi-anechoic techniques which splice measurements, meaning measured phase via 2 channel is useless (since when splicing, one must derive phase).
So - I use the combined driver measurement technique to find precice acoustic offset on the listening axis and use derived phase on all driver measurements. When compared to my dual channel measurement, the method comes up with the exactly the same acoustic response curves for the same crossover.
So... I reverted to a single USB channel measurement.
In regards to dual channel measurement, I have found this more cumbersome and only useful on a PC with a hard wired line in / out card. That is - one PCI or motherboard etc... attached. USB devices tend to have variable timing, making a single loopback measurement useful, for finding "flight time" between left and right channels, but useless when it comes to expecting the same impulse marker between measurements.
The other issue is measuring a reflection free woofer (and if 3 way) midrange response. This usually requires quasi-anechoic techniques which splice measurements, meaning measured phase via 2 channel is useless (since when splicing, one must derive phase).
So - I use the combined driver measurement technique to find precice acoustic offset on the listening axis and use derived phase on all driver measurements. When compared to my dual channel measurement, the method comes up with the exactly the same acoustic response curves for the same crossover.
So... I reverted to a single USB channel measurement.
That's interesting, I'm assuming this is with respect to "automated" (or hand-adjusted loudspeaker axis) polar measurements in a series.
That's something I haven't done before (..though I also have a pci Echo Layla to work with that I normally just use for crossover work).
Oh well, another thing for my "to-do" list.
Yeah, if you go low enough for the crossover then it's something you are likely to do outside (..if possible). (..and even then lifted up a bit with pillows surrounding the speaker.)
I have tried to purchase the Calibrated EMM-6 Mic from the site that Scott recommended, but it is not available! 🙁
https://www.cross-spectrum.com/measurement/calibrated_dayton.html
Can anyone recommend a suitable alternative?
https://www.cross-spectrum.com/measurement/calibrated_dayton.html
Can anyone recommend a suitable alternative?
A couple of options, but they cost more than the cross-spectrum calibrated EMM-6:
I use an Audix TM1 for about $250. It is half the price of an Earthworks M23, and is probably just as fine.
Another option is to buy a calibrated USB mic, such as the Minidsp Umik-1, and a Dayton EMM-6. You can use the USB mic to calibrate the EMM-6, and honestly, it is not a bad thing to have two sources of information to compare to each other.
I use an Audix TM1 for about $250. It is half the price of an Earthworks M23, and is probably just as fine.
Another option is to buy a calibrated USB mic, such as the Minidsp Umik-1, and a Dayton EMM-6. You can use the USB mic to calibrate the EMM-6, and honestly, it is not a bad thing to have two sources of information to compare to each other.
If you go this route and are willing to calibrate other mics off of the UMIK, you could also consider getting a bag of ten decent 'naked' mic capsules for a lot less than one single EMM-6.
So I am still going around in circles a little regarding the mic.
An Audix TM1 would be around £250 which is quite a lot. Are there reasonably well calibrated alternatives for less or do I need to just accept that a half decent mic is in this ballpark?
Unfortunately, the site Ariendj recommended ( https://shop.hifi-selbstbau.de/produkt-kategorie/messtechnik/ ) doesn't deliver to the UK. I guess Brexit has something to do with that. 🙁
Should I just start with something cheap like an ECM8000 while I am learning or is this a false economy?
As you can see, I am still not sure what a sensible choice is at the minute.
An Audix TM1 would be around £250 which is quite a lot. Are there reasonably well calibrated alternatives for less or do I need to just accept that a half decent mic is in this ballpark?
Unfortunately, the site Ariendj recommended ( https://shop.hifi-selbstbau.de/produkt-kategorie/messtechnik/ ) doesn't deliver to the UK. I guess Brexit has something to do with that. 🙁
Should I just start with something cheap like an ECM8000 while I am learning or is this a false economy?
As you can see, I am still not sure what a sensible choice is at the minute.
I recommend the MiniDSP UMIK-1 (I have it). It comes with a calibration file (per mic). I can't comment on accuracy, but the speakers I've designed with it with a flat frequency response sound pleasing to me. I haven't wanted for a more accurate microphone. I don't know how inaccurate the UMIK-1 is compared to a reference standard. However, like anything, you end up paying up to 80% more, for that last 20% performance... and if you are new to the hobby, then I think that money may be better spent on other tools / setup.
It's certainly a lot more than I'd spend for my first mic.
A UMIK with the calibration file from minidsp would qualify as 'reasonably well calibrated' in my book.
For £250 I'd expect a lot more than just 'half decent'. At that price I'd expect it to be more than adequate for hobbyist use.
That's disappointing.
Personally, I'd start with something cheap that is calibrated. I'm very content with my UMIK and the calibration file from the manufacturer, just like Dave. I have not yet found a scenario where I had the feeling that the UMIK is holding me back. Every time I used it the results lived up to my expectations. Should that ever not be the case I'd send it in for calibration. I'm sure there's someone on here who would send your mic in for calibration in Germany and have it shipped back to you, should you so desire.
I'd build you a mic calibrated off of my UMIK and send it across the pond if my stuff wasn't all in boxes scattered around the house, garage and cellar after my last move... I can't seem to find anything at the moment.
The Dayton EMM is calibrated, but the accuracy of the calibration is questionable (thus the recommendation to get the calibrated version from Cross Spectrum).
The Behringer ECM8000 does not come with a calibration file.
A possibility is to have a diyaudio member with a good mic calibrate your mic for you.
I recently bought an ECM8000 just because they were on sale for $29.99. if you can find someone local to calibrate the mic, either the Dayton or Behringer if probably fine.
I'd also suggest, though others may completely disagree, that since you are focused on learning and understanding right now, get something inexpensive (calibrated or not) for now and worry about more precision later.
The Behringer ECM8000 does not come with a calibration file.
A possibility is to have a diyaudio member with a good mic calibrate your mic for you.
I recently bought an ECM8000 just because they were on sale for $29.99. if you can find someone local to calibrate the mic, either the Dayton or Behringer if probably fine.
I'd also suggest, though others may completely disagree, that since you are focused on learning and understanding right now, get something inexpensive (calibrated or not) for now and worry about more precision later.
Calibration doesn't determine your success. Designing a crossover requires "relative" measurements. The mic will be the same for all measurements so there's no error in that respect. I have designed crossovers with an uncalibrated mic and all it takes is a little global EQ (tone control) after the crossover is done.
I'm saying this so this issue can be put into perspective.
Someone recommended the Beyer MM1 in an older thread
https://vintageking.com/beyerdynami...7irXQz0A2LuLCethvddnPV6pSwU8-gA3NjBPWngaVEyzI
https://vintageking.com/beyerdynami...7irXQz0A2LuLCethvddnPV6pSwU8-gA3NjBPWngaVEyzI
If it alleviates your concern... here is my UMIK-1 calibration file. Although the curve looks wild - please bear in mind the SPL axis:
Here is my ECM8000 response (self calibrated) - which I did before I got the UMIK-1 and results compared nicely with the UMIK-1
Most of these capsule based mics, probably use the same capsule and have the same rising response above 2KHz.
Since measurement software like REW allows you to load the microphone calibration file and apply it automatically to each measurement, you get a "flat" response. What I am not guaranteeing is the accuracy of the supplied calibration files, but I believe they are good enough for our DIY efforts.
Here is my ECM8000 response (self calibrated) - which I did before I got the UMIK-1 and results compared nicely with the UMIK-1
Most of these capsule based mics, probably use the same capsule and have the same rising response above 2KHz.
Since measurement software like REW allows you to load the microphone calibration file and apply it automatically to each measurement, you get a "flat" response. What I am not guaranteeing is the accuracy of the supplied calibration files, but I believe they are good enough for our DIY efforts.
You could also do it the cheeky way and get both a UMIK and an ECM 8000 from amazon, calibrate the ECM off of the UMIK, and then return the UMIK for a refund.
I have a Beyerdynamic MM1 and it has a rising HF response on axis (similar to the UMIK and ECM 8000) but it is designed to be used at 90 degrees where it's response is essentially +/-0db to 2K, +/-1dB to 15K and then falls to -2db at 20k. Beyerdynamic will provide 0deg and 90deg calibration files base on the s/n if you ask for them (...at least they did in 2017).
Thanks for the suggestion Longspeak. The MM1 is coming in at around 200 euros (uncalibrated?) on the beyerdynamic website, so still a bit expensive.
At the moment I am thinking it may be a good idea to get a ECM 8000 (around £30) to learn with and then start looking for a calibrated mic later. If I could get a nicely calibrated mike for under £150, I would probably go for that straight away, but other wise buying the whole set up in one go with a £200+ mic is a little tricky at the moment.
If anyone has any suggestions for a reasonably priced, decently calibrated mic that I can get in the UK then please feel free to post the details.
I will take the plunge with buying some equipment at the end of this month.
Thanks for everyone's help.
At the moment I am thinking it may be a good idea to get a ECM 8000 (around £30) to learn with and then start looking for a calibrated mic later. If I could get a nicely calibrated mike for under £150, I would probably go for that straight away, but other wise buying the whole set up in one go with a £200+ mic is a little tricky at the moment.
If anyone has any suggestions for a reasonably priced, decently calibrated mic that I can get in the UK then please feel free to post the details.
I will take the plunge with buying some equipment at the end of this month.
Thanks for everyone's help.
My MiniDSP UMIK-1 cost USD$97 including shipping to New Zealand. That works out to ~ GBP 80. I can't vouch for the accuracy of the calibration, but for DIY speaker design, I'd recommend that as entry level. Otherwise if you design using the uncalibrated ECM8000, you'll have a nagging worry about whether it is the mic calibration meaning you overcompensated the treble for a dull speaker... or you got your measurements wrong... or crossover wrong etc...
Remove one variable - get a basic calibrated mic.
Remove one variable - get a basic calibrated mic.
The Behringer mic is £31 from here in the UK , Possibly consider buying this with a cheap interface for timing measurements, and buy the calibrated UMIK for frequency response stuff. (that's what I've ended up doing, but haven't set up the loopback rig yet so still use the UMIK for everything at the moment)
Rob.
Rob.
The MM1 is calibrated, they send a response chart with it but if you want the 0 and 90deg files you need to contact them. Here a the link about it: https://support.beyerdynamic.com/hc...et-a-calibration-file-for-my-MM-1-microphone-
I looked at both the UMIK and ECM8000 before deciding on the MM1 since I didn't want to be stuck wishing I'd bought a better mic in the beginning and I couldn't spring for an Earthworks. Nothing wrong with the Dayton or Behringer for the purpose, and I think as AllenB mentioned most measurements are relative and not absolute so as long as the mic produces the same response consistently I wouldn't worry.
I looked at both the UMIK and ECM8000 before deciding on the MM1 since I didn't want to be stuck wishing I'd bought a better mic in the beginning and I couldn't spring for an Earthworks. Nothing wrong with the Dayton or Behringer for the purpose, and I think as AllenB mentioned most measurements are relative and not absolute so as long as the mic produces the same response consistently I wouldn't worry.