Hi!
I searched these forums for about an hour with many different search terms but I didn't find an answer. So, I feel justified in starting a new thread.
Ever since I read about T/S parameters a few years back I have always wanted to measure them for myself because it seems so cool! I found all of these great resources on how to do just this but I am confused over one main point. Well, actually two maybe.
At this website:
http://www.diysubwoofers.org/measure.htm
It says to determine the resonance frequency of a driver, vary the frequency until the voltage across the resistor reaches a minimum. What exactly is meant by this? Is it zero volts that I need or is it 1.0 volt? Also, I think my multimeter may be too limited! There are only two settings for the VAC -- 750v and 200v. Is it sufficient to use the 200v setting? Do I even use AC at all? Or, do I use DC? How many digits does the interface need? I have three zeroes -- 00.0
Questions, comments, theories, or revelations? Anything would be greatly appreciated! Thank you.
I searched these forums for about an hour with many different search terms but I didn't find an answer. So, I feel justified in starting a new thread.
Ever since I read about T/S parameters a few years back I have always wanted to measure them for myself because it seems so cool! I found all of these great resources on how to do just this but I am confused over one main point. Well, actually two maybe.
At this website:
http://www.diysubwoofers.org/measure.htm
It says to determine the resonance frequency of a driver, vary the frequency until the voltage across the resistor reaches a minimum. What exactly is meant by this? Is it zero volts that I need or is it 1.0 volt? Also, I think my multimeter may be too limited! There are only two settings for the VAC -- 750v and 200v. Is it sufficient to use the 200v setting? Do I even use AC at all? Or, do I use DC? How many digits does the interface need? I have three zeroes -- 00.0
Questions, comments, theories, or revelations? Anything would be greatly appreciated! Thank you.
As you know, the impedance of a loudspeaker's voice coil changes with frequency. By putting a resistor (whose impedance doesn't change with frequency) in series with the voice coil and measuring the voltage across that resistor at various frequencies, you can make an impedance plot for your driver.
When the voltage across the fixed resistor reaches a minimum the voltage across the voice coil will be at a maximum because the two are in series (that's why the rms output voltage of the amp should be constant regardless of frequency), which just so happens to be the driver's resonance frequency.
Just set your meter to 200VAC and you should be fine.
When the voltage across the fixed resistor reaches a minimum the voltage across the voice coil will be at a maximum because the two are in series (that's why the rms output voltage of the amp should be constant regardless of frequency), which just so happens to be the driver's resonance frequency.
Just set your meter to 200VAC and you should be fine.
I'm still not sure that I understand. What exactly is meant by minimum voltage? Is it zero volts?
What I am doing is testing a speaker for which I already have the manufacturer's specs. I am going to compare them against my own to see if I am doing the procedure correctly. The manufacturer says the Fs is 48 Hz and so I set my freq. gen. to 48 and the multimeter just displays 00.0! It actually reaches 00.0 way before 48Hz, somewhere around 65Hz or so.
What I am doing is testing a speaker for which I already have the manufacturer's specs. I am going to compare them against my own to see if I am doing the procedure correctly. The manufacturer says the Fs is 48 Hz and so I set my freq. gen. to 48 and the multimeter just displays 00.0! It actually reaches 00.0 way before 48Hz, somewhere around 65Hz or so.
Do you just have the frequency generator connected to the resistor and driver? If you do, adjust its output to 1Vrms. You should be able to measure that with your multimeter and the voltage across both the resistor and driver should be 1Vrms (or whatever you set the output to) at all times regardless of frequency.
What "minimum voltage" means is that you will see the voltage across the resistor vary as you change the frequency. When you find the frequency where the voltage across the resistor reaches its lowest value the impedance of the voice coil will be at its maximum and that corresponds to the driver's resonance frequency.
What "minimum voltage" means is that you will see the voltage across the resistor vary as you change the frequency. When you find the frequency where the voltage across the resistor reaches its lowest value the impedance of the voice coil will be at its maximum and that corresponds to the driver's resonance frequency.
My setup is this:
1. PC based frequency generator
2. External amplifier
3. Multimeter
4. 9.7ohm resistor
5. Driver (Seas L12RCY/P h1207)
Ok, so the voltage across the driver and the resistor should stay the same regardless of frequency. And the voltage across the resistor should change with variations in frequency. Sounds impossible to me.
1. PC based frequency generator
2. External amplifier
3. Multimeter
4. 9.7ohm resistor
5. Driver (Seas L12RCY/P h1207)
Ok, so the voltage across the driver and the resistor should stay the same regardless of frequency. And the voltage across the resistor should change with variations in frequency. Sounds impossible to me.
jaberwocky6669 said:
The impedance of the driver changes with frequency, so the voltage drop across the resistor and driver change with frequency, but they always add up to the same amount.
I prefer a different method of measuring, seen at:
http://www.epanorama.net/documents/audio/speaker_parameters.html
Attached is an impedance plot of an 8ohm nominal 4" driver in free air and in a small sealed enclosure. You can see the impedance peak is about 58ohms at about 106Hz - this is the driver's resonance frequency. Most other impedance curves will look very similar, including the one for the driver you are trying to measure. You can easily see how the impedance of the voice coil changes with the input frequency.
Your multimeter won't display millivolts so it probably isn't accurate enough to measure with.
The number you see on the display is in volts so 0.1 is 100mV.
Your multimeter won't display millivolts so it probably isn't accurate enough to measure with.
The number you see on the display is in volts so 0.1 is 100mV.
jaberwocky6669 said:
When your meter reaches 00.0 switch scales to 750 mv.
If this still don't work, raise the level of the input signal.
I know I have 2 meters and for some reason 1 of them
(the cheaper?) will measure low level ac fine but the other
won't.........Also try the other method (as mentioned) where
you look for a peak reading rather than a null. (1K ohm in
series with driver and put meter accross driver)...........
I would try using a larger value of resistor. Maybe you could double or triple the value to start.
BTW, I typically use a 1k resistor, but in a slightly different setup where I measure the voltage across the driver (still the same information). I can usually hear the test tone quietly. The amp typically needs to be turned up.
BTW, I typically use a 1k resistor, but in a slightly different setup where I measure the voltage across the driver (still the same information). I can usually hear the test tone quietly. The amp typically needs to be turned up.
The problems that I am having is that I read all kinds of methods for measuring thiele small parameters but then all of these extra things that have to be done or not done or whatever. It's all too overwhelming! Ok, I think that I will just spend a lot of time calibrating SW and hope for the best. thanks everyone...
Build a Wallin jig. It makes life so much easier.
For calibration, use Rightmark to find the best combination of sample frequency and size for your card, and get the volumes set up properly, then save these with Quickmix. You can then go through the SW calibration without having to worry about anything other than the main volume control.
The tutorial hops around a bit, and is a little inconsistent, but if you read through it a few times, you will get the feel of the processes required, and can then work at your own rate.
For calibration, use Rightmark to find the best combination of sample frequency and size for your card, and get the volumes set up properly, then save these with Quickmix. You can then go through the SW calibration without having to worry about anything other than the main volume control.
The tutorial hops around a bit, and is a little inconsistent, but if you read through it a few times, you will get the feel of the processes required, and can then work at your own rate.
Hi,
I did my Thiel Small measurements in about 30 minutes and then about 30 minutes to calculate the values. One driver done, now the rest. Or spend hours setting up a spreadsheet and the sums are done as soon as the return key is released.
That was before I knew about Wallin and sound cards. Well I still don't know, just that they exist. Thanks Pinky.
I did my Thiel Small measurements in about 30 minutes and then about 30 minutes to calculate the values. One driver done, now the rest. Or spend hours setting up a spreadsheet and the sums are done as soon as the return key is released.
That was before I knew about Wallin and sound cards. Well I still don't know, just that they exist. Thanks Pinky.
Well, if you look at this post, you can see the results that are achievable. Look closely at the time stamp, I had installed the card and drivers the night before, then got up, had some breakfast, and then it took me about two hours to calibrate and get the results you see as the first test.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.