hi,
Is there a way i can measure the freq response of my amp if i dont have any oscilloscope?
If i need to build a sine-wave generator, is there any good design I can copy? One of those low distortion types....
Is there a way i can measure the freq response of my amp if i dont have any oscilloscope?
If i need to build a sine-wave generator, is there any good design I can copy? One of those low distortion types....
yes
My soundcard is built in-audio97 codec.
Do you mean to say that I can use an oscillator program?
My soundcard is built in-audio97 codec.
Do you mean to say that I can use an oscillator program?
built-in ac97 probably won't do.
but to measure freq response, thd, etc. you can use the free http://audio.rightmark.org/
it's pretty simple but the quality of the results depend on your soundcard.
you would also need a shunt and/or selfmade preamp thing to measure high voltage of a power amp. soundcards usually only measure up to 2V.
but to measure freq response, thd, etc. you can use the free http://audio.rightmark.org/
it's pretty simple but the quality of the results depend on your soundcard.
you would also need a shunt and/or selfmade preamp thing to measure high voltage of a power amp. soundcards usually only measure up to 2V.
http://www.dazyweblabs.com/shannonsoft/page3.html
especially TMS-1.
Run a very short cable from output to input on your soundcard
to check first how good it is.
especially TMS-1.
Run a very short cable from output to input on your soundcard
to check first how good it is.
To get a reasonable idea of freq resp using a soundcard, I would imagine the key thing is freq resp of the card itself. Unless the error is gross, freq accuracy wouldn't be that key and THD+N almost non-germaine. I would just take any free oscilator program and run it with a DMM reading AC volts at the soundcard output. Then two things: A set the level below clipping* on the amp and B: record the output at each frequency you are going to check. Next drive the amp with the same signal but attach the DMM to the amp outputs (I'm not sure if you need a nominal 8-ohm load on the output but don't use the speakers as test tones may harm them). If the the AC volt reading at the soundcard had dips or spikes, you can caculate correction factors for each freqwhich you multiply by the amplifier output. Plot these results against freqency. Results may not be as precise as using hundreds or thousands of dollars of test equipment, but it should pick up any serious anomomolies.
(* If you can determine this directly, assume your amp has a gain of 20, multiply the DMM AC volts reading by this, square it and divide by 8. If this is below the rated watts for your amp, you should be OK.)
If I've missed something, I trust another poster to add it or correct me.
(* If you can determine this directly, assume your amp has a gain of 20, multiply the DMM AC volts reading by this, square it and divide by 8. If this is below the rated watts for your amp, you should be OK.)
If I've missed something, I trust another poster to add it or correct me.
thanks, chris, ice, sam
sam, correct me if im wrong. your idea is that, first, get my souncard to generate signals so that I can measure my souncard's response itself.
then, i feed the tones from my souncard to my amp, and then I measure the amp output.
after that, I compare the amp output with my computer output, is it?
sam, correct me if im wrong. your idea is that, first, get my souncard to generate signals so that I can measure my souncard's response itself.
then, i feed the tones from my souncard to my amp, and then I measure the amp output.
after that, I compare the amp output with my computer output, is it?
I am downloading those programs now, gonna try them tomorrow.
one question i missed out: how do i feed the signals from my amp into my computer safely?
one question i missed out: how do i feed the signals from my amp into my computer safely?
1) to make sure there are no err...what u call it...... holes, and/ or spikes/resonance at certain frequencies. when i get a flat response, i hear music exactly as recorded in studio/ live
2) making sure i have my amp is working all the way to 20kHz, even though I may not be hearing them-)
3)checking for oscillations
correct me if im wrong
2) making sure i have my amp is working all the way to 20kHz, even though I may not be hearing them-)
3)checking for oscillations
correct me if im wrong
Hi Skyraider,
Are we talking about a self made amplifier here?
I hope I can express my answers to your questions in a way to help you:
You're right, your amp shouldn't have anything like holes, spikes and resonances in it's 'transfer function', the way it 'processes' it's input signal should be absolutely linear and independent of the audio material.
But why do you expect errors? Is it damaged, or did you DIY it and want to test it?
And another point: your amp having a flat response doesn't mean that you hear *exactly* what happened in the studio! You'll actually never hear what happened in the studio, unless you visit the engineer in the studio 😀
Your speakers or headphones have an influence on the sound that's way above the influence of your amp. You should probably care for measuring and optimizing your speakers first 😉
Hope this helps 😉
Are we talking about a self made amplifier here?
I hope I can express my answers to your questions in a way to help you:
You're right, your amp shouldn't have anything like holes, spikes and resonances in it's 'transfer function', the way it 'processes' it's input signal should be absolutely linear and independent of the audio material.
But why do you expect errors? Is it damaged, or did you DIY it and want to test it?
And another point: your amp having a flat response doesn't mean that you hear *exactly* what happened in the studio! You'll actually never hear what happened in the studio, unless you visit the engineer in the studio 😀
Your speakers or headphones have an influence on the sound that's way above the influence of your amp. You should probably care for measuring and optimizing your speakers first 😉
Hope this helps 😉
"sam, correct me if im wrong. your idea is that, first, get my souncard to generate signals so that I can measure my souncard's response itself.
then, i feed . . . . "
Yup. If you have faith in the soundcard you could assume that it's response is flat to begin with, but it's a good idea to be certain. I expect that most amps manufactured in recent history are dead flat, so if you get minor wobbles chances are they are from the the soundcard, the software, the DMM, or from your procedures. All these are more likely than a recently bult solid state amp.
If you find major anomolies it would be a sign that the amp neads to be looked at by a service tech.
then, i feed . . . . "
Yup. If you have faith in the soundcard you could assume that it's response is flat to begin with, but it's a good idea to be certain. I expect that most amps manufactured in recent history are dead flat, so if you get minor wobbles chances are they are from the the soundcard, the software, the DMM, or from your procedures. All these are more likely than a recently bult solid state amp.
If you find major anomolies it would be a sign that the amp neads to be looked at by a service tech.
If you have no reason to think there is a problem with the amp you will be going to a lot of trouble for nothing, power amplifiers are particularly uneventful in this respect.
There would have to be a serious problem with a power amplifier for it to be anything other than ruler flat within the audio band.
Incidently you could always use an audio test CD and suitable DMM to read of the amplitude at different frequencies, absolute accuracy isn't important.
There would have to be a serious problem with a power amplifier for it to be anything other than ruler flat within the audio band.
Incidently you could always use an audio test CD and suitable DMM to read of the amplitude at different frequencies, absolute accuracy isn't important.
WARNING
If you intend to use a soundcard, be aware that with many amps input gnd and speaker gnd is not always the same. In many amps speaker ground is connected through something like a 100 ohms resistor to the actual gnd.
If you connect input gnd and spk gnd together at your soundcard with such an amp you can run into problems and even into oscillations. In that case you need a balanced input for your soundcard.
Cheers 😉
If you intend to use a soundcard, be aware that with many amps input gnd and speaker gnd is not always the same. In many amps speaker ground is connected through something like a 100 ohms resistor to the actual gnd.
If you connect input gnd and spk gnd together at your soundcard with such an amp you can run into problems and even into oscillations. In that case you need a balanced input for your soundcard.
Cheers 😉
A bit off topic...and then again, maybe not.
Anybody made a cd with known frequencies to play and check freq. response?
Cheers
Magura
Anybody made a cd with known frequencies to play and check freq. response?
Cheers
Magura
Magura said:
No, I only have C++ hack that I threw together to get
it done. 🙂
I could post the source code for it if you wish, but be warned,
it IS a quick hack.
I think others have succesfully used various programs to do
the same thing.
actually, I have no background in electronics, just picked up bits of info along the experience of building a couple of amplifiers. Just that I am wondering why some of the amps Ive built sounded better than others, some amplifies better vocals, some is just all muddy colored. I thought i can check my design for errors. recently Ive built a preamp with tone controls, and I just want to check its response just to be certain.
just for thought... someone discussed that a good amp gives a sense of depth, like a 3-d effect. In my opinion, 3-d effect is caused by the stereo sound which we hear, that means two separate chanels. Now, a question is that does crosstalk reduces this stereo effect? I meant a bit of signal leaked into the other channel, at a large degree it almost cancels out the stereo sound.
If sharing the same power supply for two chanels, how much is the signal leak? how about using a stereo potentialmeters?
any comments please?
just for thought... someone discussed that a good amp gives a sense of depth, like a 3-d effect. In my opinion, 3-d effect is caused by the stereo sound which we hear, that means two separate chanels. Now, a question is that does crosstalk reduces this stereo effect? I meant a bit of signal leaked into the other channel, at a large degree it almost cancels out the stereo sound.
If sharing the same power supply for two chanels, how much is the signal leak? how about using a stereo potentialmeters?
any comments please?
- Status
- Not open for further replies.