Actually, it turns out i am not sure which channels i am supposed to use in the audio device setup. The way my sound card drivers are set up is extremely confusing to me.. i think i should use 1/2 for input and 3/4 for output, but when i do, the levels are suspiciously low and can't calibrate. Picture:
What do you mean exactly by shorting? Remove it or run a wire between the two ends of the resistor? Sorry for the dumb question, but i am really bad at electronics.
p.s. check my last post, i can't really figure out which channels i am supposed to use in the driver setup -_-
Last edited:
You may already know this, so forgive me if I am stating the obvious. When the speaker specification states a ohmic value, they are stating total impedance (Z) at a given audio frequency, not a DC resistance measured with a DMM. A DC resistance should be less than total impedance, not more. A digital LCR meter that has a resistance function measured at the 10Khz setting will give you a more realistic representation of the speaker impedance. Also, I may be corrected by one of the more experienced audiophiles, I would think you may consider an inductor that has an X of L (aka reactance) slightly less than the impedance of the speaker you are testing at a given frequency for some type of reference? If you do use a resistor for the reference, it needs to be constructed in a manner that provides minimal to zero reactance to a audio signal -- meaning a wire wound resistor while more stable with change of temperature, would not necessarily have the same resistance to an audio signal; it's value would need to be determined at the test frequency and wattage.
Thanks for the input. That's a bit too technical for me, tho, i am looking to reproduce what others have already done successfully. I am certain, that a normal 1% resistor is what is used to measure with. As for the stated ohmic value of speakers, well, i don't know what to say. As i have neither an accurate DMM nor any LCR meter, all i can do (for now) is measure with a normal DMM and use that. I know there are other ways to measure T/S parameters (like using an accurate low frequency millivolt meter and a frequency generator).
The driver i am attempting to measure isn't a new one and i used a really cheap chinese made DMM, so the 5 ohm measurement might be off for a number of different reasons. I don't think that's really very important in the case, since i apparently have other issues with the software.
Besides, i am merely trying to learn how to use this software / DIY hardware combo, not aiming at super precise results. After i get it working and become familiar with it, i will certainly look more closely at the user-input data.
cheers
Last edited:
If I can somehow be of some help feel free to contact me. To be honest, I haven't had the money or the time to be an audiophile. Most of my practical experience is in the radio frequency bands of the spectrum. I am not an electrical engineer, either. Having said all that, I think I gave accurate general electronic theory. As with any type of engineering, what one works out with your favorite computing device does not aways work, with the exception of NASA, maybe.... ;-) That or they are way too good at hiding their mistakes. Anyway, thanks for the reply. Are you from the UK? I spent 4 years there in Suffolk working at the airbases around Woodbridge and I lived in Shotley Gate.
One of the reasons I wanted you to use bare wire is due to making it easier for you to troubleshoot any issues by unravelling wire connections with your fingers. As you have everything soldered and shielded, no one can really tell what is taking place. I see you are using thick RCA cable which appears to resemble coaxial.
Can you not just use basic RCA cables?
Using the below configuration offers a Channel Difference of 0.05 dB with the on-board soundcard.
Why not just use the on-board soundcard on your desktop computer using a bare wire scheme?
Can you not just use basic RCA cables?
An externally hosted image should be here but it was not working when we last tested it.
Using the below configuration offers a Channel Difference of 0.05 dB with the on-board soundcard.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Why not just use the on-board soundcard on your desktop computer using a bare wire scheme?
Are you sure you have properly connected the soundcard ???
I've seen that kind of "oscillating" signals and they have been due to aa input (or ouput) without a connection or with a poor one. Some time ago there was an article in a forum speaking about to reset the sound card settings to factory ones and then, restart the PC and configure the sound device again (I have never tried that... but perhaps it is useful...)
And as you have been said, tray to use pink-noise (FFT mode) and activate the averaging to smooth the impedance curve (about 20 o more averages will do fine).
I've seen that kind of "oscillating" signals and they have been due to aa input (or ouput) without a connection or with a poor one. Some time ago there was an article in a forum speaking about to reset the sound card settings to factory ones and then, restart the PC and configure the sound device again (I have never tried that... but perhaps it is useful...)
And as you have been said, tray to use pink-noise (FFT mode) and activate the averaging to smooth the impedance curve (about 20 o more averages will do fine).
With the M-Audio cards you should be using outputs 1,2 and inputs 1,2. Ins and outs 3,4 are the SPDIF ports.
Appreciate it, thanks 🙂 I hear you about the audiophile stuff, it sure requires a solid investment in both time and money, that's why i am trying to not go too crazy myself.
I am actually from Bulgaria, but i also used to live in the UK; spent some time in Coventry while i was studying. That's where i picked up the habit to use "cheers" instead of "thanks".
I get the idea to start with the simplest version in order to spot problems easier. Too bad i started with the fanciest box, but i am pretty sure the simple jig i made is wired correctly. Also, i am pretty sure i have a problem with the audio device setup and/or connection, something like that.
I can use basic cables and replicate the setup in your picture with my on board sound card. I probably will if it comes to that. But, i'd really like to put the M-audio card i bought years ago (and rarely used) to some use, now that i can think of one. Don't you think the better card will lead to higher accuracy measurements? I am not sure about that, just shooting in the dark here.
Thanks
Hey. Thanks for the suggestion, i am definitely going to check the card. Gonna take it out and clean it / change the slot i use. I suppose by restarting the factory settings you mean uninstalling all drivers and re-installing them. Could be some windows **** up, that won't surprise me.
That's unexpected 🙂. So, in audio device setup i have to use the 1/2 for both input and output, like this:
and the cables connected like this:
It actually calibrates like that, but the difference i get is -0.01dB:
And the measurement result is as usual:
An externally hosted image should be here but it was not working when we last tested it.
tried to measure a resistor, getting outrageous values as well.
Last edited:
The picture of your jig looks correct, but your drawings have the right channel as the reference.
The last measurement you've shown looks good with the exception that the scaling is off by a factor of ~10.
Verify the value your 27 ohm resistor. If it was actually 2.7 ohms the measurement would be close.
You might also try swapping the right and left sense (input to the sound card) cables to see what happens.
The last measurement you've shown looks good with the exception that the scaling is off by a factor of ~10.
Verify the value your 27 ohm resistor. If it was actually 2.7 ohms the measurement would be close.
You might also try swapping the right and left sense (input to the sound card) cables to see what happens.
Hey. Which picture is that? I am a bit confused. I also thought the last measurement looked great, because that's what i expect a subwoofer impedance curve to look like and the Fs seems about right.
As to the resistor i used, it clearly states 27ohm 10w on the casing. It is a big ceramic one and i don't think it is 1%. Before i soldered it, tho, i measured it with a DMM and it read 27 ohms. Now i took that reading again (while it is soldered in the simple jig) and the reading is a bit dodgy, but reads 04.3. Is that normal?
I tried swapping the left and right input cables to the sound card and there's literally no difference. Both the calibration and measurement are the same:
I prefer to use MFR with at least 1% tol. Wirewound resistors might have inductance. else use NI type wirewound resistors if available.
So, basically, i have to try to swap the ceramic resistor for a MFR one (1%, same value), right?
cheers
Hi there, any progress with solving the problem?
I have the very same problem with my Behringer UCA222 sound card connected to my laptop. Imedance results are about 100 times larger than they should be. Don't know what to do, seems like it has to do something with input and output impedances of soundcard line in and out (UCA222 has 400 ohm imp. on line out and 27K on line in). perhaps if I use 100 times higher value for reference resistor (27k)? I don' have one with me now to try it.
Vedran
I have the very same problem with my Behringer UCA222 sound card connected to my laptop. Imedance results are about 100 times larger than they should be. Don't know what to do, seems like it has to do something with input and output impedances of soundcard line in and out (UCA222 has 400 ohm imp. on line out and 27K on line in). perhaps if I use 100 times higher value for reference resistor (27k)? I don' have one with me now to try it.
Vedran
- Status
- Not open for further replies.