Just stumbled upon this very informative thread, thanks Scott for the concise description of the tools needed for a good and cost effective setup. I do find it appealing that its capabilities won't be the limiting factor as you progress into more advanced techniques.
As for the "accurate acoustic phase measurements", is there a tutorial or write-up somewhere on how to perform these using the gear and software you recommended on post #2 of this thread? Any help will be appreciated as I want to avoid spinning my wheels too much when I start doing this.
Look at these two, they should be all you need for measuring. The acoustic offset information is based on using the Passive Crossover Designer program, but the same technique can be used in XSim or other crossover design programs.
Thank you for the write-ups - it's very much appreciated! I will assume that the methods described should work just as well with large z-offsets i.e. midrange horn and direct radiator woofer.
One other question; when measuring for the horizontal off-axis FRs, where is the vertical axis of rotation located? Would it be on the horn mouth, the acoustic center of the woofer, on the baffle or some other point? Given that the acoustic centers of the two drivers can be as much as 12 inches apart I would like to know what the correct procedure would be in these cases.
Thanks.
Some of you mentioned a USB mic not being able to .. do what ?
I just bought my miniDSP-UMIK1 and learing REW nowadays.. first just playing around speaker measurements in free air and preparing the impedance calibration + thiele-small measurement method.
I found a lot of videos with this mic online and doesn't seem to be all that bad for a calibrated measurement tool. At least for my hobby needs it's alone amazing, I managed to quickly correct the 15dB bump in my old little crappy Logitech 2.1 set and they sound after correction like "wow". (REW + Equalizer APO under newest W10 with the on-board Realtek soundcard). I really wanted to measure and correct a mediocre setup first, with the biggest possible flaws in it .. and it seems to work to a degree far better than just acceptable.
You're talking about feedback, like USB mics don't have it or something like that. Not sure if it's feedback, but I think yes. Anyway... for timing reference I can choose electronic loopback (where I lead one side of the stereo signal back to another input on the sound card) and acoustic timing reference too (here, a tweeter is required for the reference signal emission). And I can set under Properties on the MIC if feedback should be taken from L or R side .... with regards to the mic itself (on the stereo amp, only L is connected to soundcard output, so 1 driver). Quite interesting approaches for a beginner with this tool.
Please feel free to extend or correct me, I'm interested in your opinion.. still learning and enjoying it at the same time.
Maybe one question right now: how can I make woofer near-field SPL measurement which translates into a 1m/2.83V/8Ohm standard SPL graph ? I have no clue at which volume mixer setting (Linux) and amp pot setting I'm outputting exactly 2.83V on the speaker terminals. (Woofer is nominal 4 Ohms but that's just 3dB more than the same with 8 Ohms, right ? So a measured 4Ohms woofer's graph is 3dB less converted into 8 Ohms term). Can I measure output voltage with a multimeter ? (Either in DC or AC Voltage measurement - I assume AC. Will it be an average or peak-to-peak showing up on the voltmeter ..? At which frequency ? Pink noise ?)
REW often reminds me when I try to calibrate something that my mic is calibrated already, no need to do that, but when I measure a driver's SPL curve, the system doesn't have any info about the ouput voltage nor does it fix the distance to 1m, most tutorials tell I can place the mic anywhere. This allows me to make relative SPL to see an overall behaviour of a speaker (approximately), but does not tell me anything on an absolute scale about the driver's sensitivity. I'm into a medium sized 3-way with intentionally old, low quality drivers, to see if I can do 'wonders' with them
but I'd give it a try with a passive XO first if I manage to get avarage SPLs for my drivers, in order to design a passive XO efficiently. Then I'll correct them in an active way too, just looking for a bunch of acceptable-quality class D modules on aliexpress. (No ICE/Hypex now) 
I just bought my miniDSP-UMIK1 and learing REW nowadays.. first just playing around speaker measurements in free air and preparing the impedance calibration + thiele-small measurement method.
I found a lot of videos with this mic online and doesn't seem to be all that bad for a calibrated measurement tool. At least for my hobby needs it's alone amazing, I managed to quickly correct the 15dB bump in my old little crappy Logitech 2.1 set and they sound after correction like "wow". (REW + Equalizer APO under newest W10 with the on-board Realtek soundcard). I really wanted to measure and correct a mediocre setup first, with the biggest possible flaws in it .. and it seems to work to a degree far better than just acceptable.
You're talking about feedback, like USB mics don't have it or something like that. Not sure if it's feedback, but I think yes. Anyway... for timing reference I can choose electronic loopback (where I lead one side of the stereo signal back to another input on the sound card) and acoustic timing reference too (here, a tweeter is required for the reference signal emission). And I can set under Properties on the MIC if feedback should be taken from L or R side .... with regards to the mic itself (on the stereo amp, only L is connected to soundcard output, so 1 driver). Quite interesting approaches for a beginner with this tool.
Please feel free to extend or correct me, I'm interested in your opinion.. still learning and enjoying it at the same time.
Maybe one question right now: how can I make woofer near-field SPL measurement which translates into a 1m/2.83V/8Ohm standard SPL graph ? I have no clue at which volume mixer setting (Linux) and amp pot setting I'm outputting exactly 2.83V on the speaker terminals. (Woofer is nominal 4 Ohms but that's just 3dB more than the same with 8 Ohms, right ? So a measured 4Ohms woofer's graph is 3dB less converted into 8 Ohms term). Can I measure output voltage with a multimeter ? (Either in DC or AC Voltage measurement - I assume AC. Will it be an average or peak-to-peak showing up on the voltmeter ..? At which frequency ? Pink noise ?)
REW often reminds me when I try to calibrate something that my mic is calibrated already, no need to do that, but when I measure a driver's SPL curve, the system doesn't have any info about the ouput voltage nor does it fix the distance to 1m, most tutorials tell I can place the mic anywhere. This allows me to make relative SPL to see an overall behaviour of a speaker (approximately), but does not tell me anything on an absolute scale about the driver's sensitivity. I'm into a medium sized 3-way with intentionally old, low quality drivers, to see if I can do 'wonders' with them
but I'd give it a try with a passive XO first if I manage to get avarage SPLs for my drivers, in order to design a passive XO efficiently. Then I'll correct them in an active way too, just looking for a bunch of acceptable-quality class D modules on aliexpress. (No ICE/Hypex now)
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.