I have always used a USB mic and had good success but can see that an XLR mic and a USB 2 ch interface can be useful. Maybe time to buy an XLR microphone.
For anyone who thinks you can’t get accurate acoustic offsets using a single USB mic, here is recent XO I am working on for a PA speaker. The simulated response is blue and the actual measured response is the yellow curve, woofer is red, tweeter is green:
It’s not the simplest XO (here in prototype stage with Wago connectors):
Last edited:
Just sounds like basic clock sync.
seems like you could easily eliminate the clock issues by not using USB all together
its just another clock that has to be compensated for.
rather synchronous or asynchronous.
just curious how a basic internal soundcard works.
seems like it should sync directly with the codec clock.
so doesn't need to bother syncing with a serial port.
and would eliminate all the errors from having to divide
sample rate and codec rate with a phase lock loop.
far as I understand any digital to audio has to do this.
But using usb, you now have 2 clocks that need to divide
the sample rate. then sync together.
seems redundant and full of errors.
I was running full duplex in 1998
10 in 10 out.
no such thing as USB1 or 2
the card installed directly in the computer.
and then had a break out box with in and out
But that was for full duplex.
recording data you only need input.
which any basic soundcard should do???
which should be only one clock, not 2
seems like all you need is a basic analog pre amp
for the microphone. which the computer has.
then again if you using condenser mic
then you just need the phantom power.
no clocks no bullcrap, just plug into the
computer??
seems like you could easily eliminate the clock issues by not using USB all together
its just another clock that has to be compensated for.
rather synchronous or asynchronous.
just curious how a basic internal soundcard works.
seems like it should sync directly with the codec clock.
so doesn't need to bother syncing with a serial port.
and would eliminate all the errors from having to divide
sample rate and codec rate with a phase lock loop.
far as I understand any digital to audio has to do this.
But using usb, you now have 2 clocks that need to divide
the sample rate. then sync together.
seems redundant and full of errors.
I was running full duplex in 1998
10 in 10 out.
no such thing as USB1 or 2
the card installed directly in the computer.
and then had a break out box with in and out
But that was for full duplex.
recording data you only need input.
which any basic soundcard should do???
which should be only one clock, not 2
seems like all you need is a basic analog pre amp
for the microphone. which the computer has.
then again if you using condenser mic
then you just need the phantom power.
no clocks no bullcrap, just plug into the
computer??
As I want my .frd to represent how I will be using the driver, I make the file I will be loading into simulation @ 1 meter. I really don't care what it is a mm off the surface. I use Holm usually. I use my Woofer Tester II for the impedance.
So, my files only work for me, but they do work for me. I use my E8000 mic and a Focusrite IO box. At least until my mike developed a nasty 60 Hz spike. ( I had to edit it out by hand last time) . I then use Excel to do an offset so everything is a nice positive number.
Anyway, I do a prorotype box, mount everything and test. I "calibrate" @ 2V @ 2K. for tweeter and woofer. I may try my Anthem mic as it is not too far from flat and may be good enough for measuring. I have an old Gold Sound pulser I use to do the AC measurement.
Folks, you don't know how easy you have it. When I started, it was a signal generator, VTVM, and graph paper. No simulations. Before White published the tables, I had my own "close enough" QB4 table I generated by calculator for alignment. We did textbook crossovers, then measured them, tweaked by best guess, and did it again and again, ending with everyone sitting around with a glass of wine saying, "got it" As we had no MLS, we measured drivers sitting on a step-ladder outside.
So, my files only work for me, but they do work for me. I use my E8000 mic and a Focusrite IO box. At least until my mike developed a nasty 60 Hz spike. ( I had to edit it out by hand last time) . I then use Excel to do an offset so everything is a nice positive number.
Anyway, I do a prorotype box, mount everything and test. I "calibrate" @ 2V @ 2K. for tweeter and woofer. I may try my Anthem mic as it is not too far from flat and may be good enough for measuring. I have an old Gold Sound pulser I use to do the AC measurement.
Folks, you don't know how easy you have it. When I started, it was a signal generator, VTVM, and graph paper. No simulations. Before White published the tables, I had my own "close enough" QB4 table I generated by calculator for alignment. We did textbook crossovers, then measured them, tweaked by best guess, and did it again and again, ending with everyone sitting around with a glass of wine saying, "got it" As we had no MLS, we measured drivers sitting on a step-ladder outside.
Folks, you don't know how easy you have it.
Yes, I agree. I can measure, simulate and design a XO, test it out and listen to it all in one night in my lab. Tweaking values in simulator and by ear over next several days and have final XO done and settled within a week. It’s been so helpful to have Xsim, REW, and DATS as tools.
Yes, I agree. I can measure, simulate and design a XO, test it out and listen to it all in one night in my lab. Tweaking values in simulator and by ear over next several days and have final XO done and settled within a week. It’s been so helpful to have Xsim, REW, and DATS as tools.
More reasons to go for the analog mike and dual channel audio interface:
- one would want to measure impedance from time to time. Why buy a separate DATS when your dual channel interface fits the job perfectly?
- one can exchange microphones (just to check if your mike still does what it has to do after it's fallen on the ground...)
- one can record the occasional performance (ok, buy another microphone...)
- one acquires a pretty high standard DA for the music PC
- one could (when buying a 4 to 8 out interface) set up a PC based digital crossover solution for R&D or daily use