So I have been searching all over for info on doing Thiele/Small parameter measurements for a driver that I have. I have found a bunch of sites that talk about how to do it, but they all have different opinions, and they all seems to have missing or incomplete descriptions on how to take the measurements. I would love to know what has worked for any of my fellow DIYers.
For reference these are the best sites that I have found so far, but like I said i'm not totally clear on any of them. If you have tried one of these methods I would also love to hear about it.
http://www.mh-audio.nl/downloads/read-measuring-tsp.pdf
http://sound.westhost.com/tsp.htm
The Subwoofer DIY Page - Measurements
Loudspeaker Design Cookbook section 8.31
I should mention that I don't want to buy any software or products specifically to perform this task. I know that there are products out there that basically do it for you, but I also know that it can be done with limited equipment, so I would like to do it like that.
Note: limited equipment includes things like a multimeter, an amplifier, a program to produce waves (there are free ones I think), basically things most DIYers already have, or are free.
Also I started this thread earlier (http://www.diyaudio.com/forums/mult...alibration-when-measuring-t-s-parameters.html), but it turns out that I need more information that just the info on voltage calibration, so that is the reason I started this second one. I hope that many of you will be able to post on here so that those looking for this info in the future will have an easy time finding good resources.
For reference these are the best sites that I have found so far, but like I said i'm not totally clear on any of them. If you have tried one of these methods I would also love to hear about it.
http://www.mh-audio.nl/downloads/read-measuring-tsp.pdf
http://sound.westhost.com/tsp.htm
The Subwoofer DIY Page - Measurements
Loudspeaker Design Cookbook section 8.31
I should mention that I don't want to buy any software or products specifically to perform this task. I know that there are products out there that basically do it for you, but I also know that it can be done with limited equipment, so I would like to do it like that.
Note: limited equipment includes things like a multimeter, an amplifier, a program to produce waves (there are free ones I think), basically things most DIYers already have, or are free.
Also I started this thread earlier (http://www.diyaudio.com/forums/mult...alibration-when-measuring-t-s-parameters.html), but it turns out that I need more information that just the info on voltage calibration, so that is the reason I started this second one. I hope that many of you will be able to post on here so that those looking for this info in the future will have an easy time finding good resources.
I took a look at the directions on your site (Read Research - Articles - Measuring Thiele-Small Parameters - Page 1). I fully understand your method now. I am still curious about one thing though. When you do your calibration you use a constant value resistor. I really like the idea of calibrating, but wouldn't the impedance that the amp sees make a bigger difference in it's ability to drive than frequency. In other words, you calibrate with a 10 ohm resistor, but when the impedance of the driver reaches values like 60 ohms wouldn't the calibration be useless at that point? Please forgive my ignorance if I'm way out to lunch.
Does anyone have any experience with, comments, or opinions on the method that is mentioned briefly in the Speaker Design Cookbook? If you are not sure what this is basically you use a device to measure the voltages across a series resister and the driver as you sweep though the frequencies. In the end you are left with two graphs: one that can be used to find the current at any point during the sweep, and one that can be used to find the impedance of the driver at any point. Instead of using a special device I'm sure that you could simply use two multimeters. To me this seems like the most logical method because there are no unknowns. At every point you know the voltage that the amp is supplying by adding the two voltages, you can calculate the current though the circuit, and you can therefore find the true impedance of the driver at any point. From those two graphs you should be able to find Fl and Fh etc. Maybe I'm just not seeing the downside to this method.
Never done it but REW will calculate T/S parameters from an impedance measurement, well two of them. The impedance measurements themselves are pretty easy to do.
Impedance Measurement
Thiele Small Parameters
Impedance Measurement
Thiele Small Parameters
Last edited:
No, the calibration is still correct because you have calibrated two things out - the meter and the amp. For example the meter reads 6mV low at 30Hz, this will always be the case whether it is displaying 600mV or 100mV as it is dependent on frequency. With regards to the amp, yes you have a good point there, and the calibration will be most accurate around the 10 ohms area. But, when the impedance rises to say 60 ohms, whereas your calibration data may show for example you are down 8mV at 25Hz this now becomes a small number in relation to the reading of 600mV, so the result is valid. What you are doing in this situation is helping the reading get more accurate rather than being totally accurate - every little helps as it were.
When using the current source module I designed, it's frequency response is flat down to very low frequencies, so the main source of error is the meter. And the calibration will successfully null that out.
Thanks for the comments and good luck with your testing.
Last edited:
There are some issues with using current source measurement of loudspeaker impedance, here is a short extract from a Scan-Speak paper presented at BNAM2012 concerning measurement of loudspeaker parameters:
Loudspeakers in practice are generally driven from a low impedance source (amplifiers with negative feedback), but loudspeaker measurements are often executed using a constant current source, and this is not the most appropriate.
The magnetic damping, which at least for small signals is very linear, is inactivated. This means that the impedance peak is determined by QMS and not QTS, the mechanical instead of the much more regular total Q-factor. This results in large peak amplitudes and extra high velocities around the resonance - in particular for modern loudspeakers with low loss surrounds and non-conducting voice coil formers (which not like an aluminum voice-coil former contributes with damping, due to eddy current losses induced by its moving up and down in the dc-field in the air gap).
The result is that the measured impedance curves around resonance are deviating from the shape conforming to the model. This is caused by maximum in both amplitudes and velocities; amplitudes combined with the lacking tie on the nonlinearities in the compliance and damping, velocities softening the suspension caused by “thixotropy”.
These factors result in compliance softer around resonance and misshapen impedance curves. This results in less satisfying fits and an imprecise derivation of parameters.
JohnPM,
very good point you brought up there. I've quit using constant current methods long ago, now using the source impedance that will actually seen by the driver in the application (typically lowish), and most importantly, use sufficient/realistic levels. With 20mA constant current (as used by Clio, for example) results are pretty useless (off by often more than 20%), notably with bass drivers. Plus, some drivers can have severe so-called "jump resonance" (a chaotic behaviour) without any electrical damping and higher levels of constant current applied.
very good point you brought up there. I've quit using constant current methods long ago, now using the source impedance that will actually seen by the driver in the application (typically lowish), and most importantly, use sufficient/realistic levels. With 20mA constant current (as used by Clio, for example) results are pretty useless (off by often more than 20%), notably with bass drivers. Plus, some drivers can have severe so-called "jump resonance" (a chaotic behaviour) without any electrical damping and higher levels of constant current applied.
Old threat but oke, I have try to measure thiele small parameters but get not what I want, like the VAS is 51 liters in stead of the 220 liters from visaton.
Also the qms is to high, fs etc, maybe this is because of the use of the pc headphone output or wrong resistor in measurement jigg.
thanks
Also the qms is to high, fs etc, maybe this is because of the use of the pc headphone output or wrong resistor in measurement jigg.
thanks
I have used REW with a sound card and 150ohm resistor with good success. But some back EMF from a quad woofer array generated enough voltage to fry the input FETs on my soundcard.
Now I use Dayton DATS v2 - works very well and is a great tool. Has LCR meter too for getting precise values of inductors and caps.
I use added mass method for Vas. Use silly putty (Dow dilatation compound) for added mass. It is sticky and sticks to cone but doesn't leave residue on delicate paper or aluminum cones. You need a digital gr scale to weigh the putty donuts that go on cone at interface of dustcap and cone.
Now I use Dayton DATS v2 - works very well and is a great tool. Has LCR meter too for getting precise values of inductors and caps.
I use added mass method for Vas. Use silly putty (Dow dilatation compound) for added mass. It is sticky and sticks to cone but doesn't leave residue on delicate paper or aluminum cones. You need a digital gr scale to weigh the putty donuts that go on cone at interface of dustcap and cone.
Last edited:
DATS V2. It's been dead on accurate for me. $99 at Parts Express
It's push button, has all the hardware I need and has fantastic L,C and R measurements. It's a $50 if you buy it with OmniMic (Usually $299).
I know many here like to DIY their test tools. I don't. I like to get to done fast. They do have limitations and occasionally I lust after something like Pocket CLIO.
But based on my measurements with ScanSpeak and Focal drivers, and a bunch of coils and caps it's just dead on accurate for my work. I also know of at least one California Bay Area megabuck speaker maker that has a couple they use in their QA lab, in addition to their really expensive tools.
Best,
Erik
It's push button, has all the hardware I need and has fantastic L,C and R measurements. It's a $50 if you buy it with OmniMic (Usually $299).
I know many here like to DIY their test tools. I don't. I like to get to done fast. They do have limitations and occasionally I lust after something like Pocket CLIO.
But based on my measurements with ScanSpeak and Focal drivers, and a bunch of coils and caps it's just dead on accurate for my work. I also know of at least one California Bay Area megabuck speaker maker that has a couple they use in their QA lab, in addition to their really expensive tools.
Best,
Erik
Hum??? DATS can measure with a cabinet or added mass. Added mass worked really well for me, but it takes a very sensitive little scale.
Best,
Erik
I have a little scale, precise to 0.01 grams, still vas what has to be 220 liters give 130 liters. I have read about that and it is very difficult to measure.
piece of tekst from here Measuring Loudspeaker Driver Parameters
Measuring Vas (equivalent air compliance), Method 2
The second method is to use an added mass, M1. Typically modelling clay or Blu-Tak is simply stuck to the cone close to the voicecoil, and the change of resonant frequency allows you to determine the moving mass of the cone. Armed with this, you can then calculate the Vas.
For speakers less than 200mm (8"), use 5 grams, for 200mm use 10g, and for 250mm (10") or larger, use 20g. You may need to add more if the mass chosen does not reduce resonance by at least 10%. The mass must be measured accurately! Even a small error can cause a large variation in the calculated Vas, so a precision scale (accurate to at least 0.1g) is essential.
You also need to measure the effective cone diameter. This is generally taken as a measurement that includes half the surround. Again, an inaccurate reading will make a big difference. Because of this, the test box method is probably more accurate. You don't need to worry about extremely accurate measurements that have a profound effect on the measurement result. Still, the added mass method is quick and convenient, and many people (including me) find it quicker and easier than using a known volume.
regards
Kee,
I am not sure what you are saying, whether you agree or disagree or are just adding details.
Personally I've measured 3 different pairs of 6.5" speakers with the added mass method and found everything spot in. Two were Focal, one was ScanSpeak. The SS was OFF from listed, but that's a known issue. The measured SS figures were accurate. The Focal measurements coincided with the published specs for the 2 different mid-woofers I tested.
I know the SS published figures are wrong because their recommended cabinet volumes don't jive with the published T/s parameters they give, but were very close to what you'd get if you used what DATS told me.
Best,
Erik
I am not sure what you are saying, whether you agree or disagree or are just adding details.
Personally I've measured 3 different pairs of 6.5" speakers with the added mass method and found everything spot in. Two were Focal, one was ScanSpeak. The SS was OFF from listed, but that's a known issue. The measured SS figures were accurate. The Focal measurements coincided with the published specs for the 2 different mid-woofers I tested.
I know the SS published figures are wrong because their recommended cabinet volumes don't jive with the published T/s parameters they give, but were very close to what you'd get if you used what DATS told me.
Best,
Erik
It is possible I do something wrong, or the pc output amp with the arta jigg is to weak, therefore i did make a electronic jigg who go outside pc soundcard.
First I go sheck my scale, it can be corrected I need to go for store to do that.
I did measure a visaton wsp26s 8 inch and VAS was uge out of spec, normally be 220 liters, and giving 130 liters.
Today because it raining again, I go make a standard for the speakers to put him in free air, this can also be a problem.
I did read that the voltage on the speaker may never be go above 1 volts for accurate measurement.
also DAT do just calculate it like arta and rew do, so this must also work. Only possible the DAT jigg has electronics in them to
measure more accurately, like the jigg I made.
kees
First I go sheck my scale, it can be corrected I need to go for store to do that.
I did measure a visaton wsp26s 8 inch and VAS was uge out of spec, normally be 220 liters, and giving 130 liters.
Today because it raining again, I go make a standard for the speakers to put him in free air, this can also be a problem.
I did read that the voltage on the speaker may never be go above 1 volts for accurate measurement.
also DAT do just calculate it like arta and rew do, so this must also work. Only possible the DAT jigg has electronics in them to
measure more accurately, like the jigg I made.
kees
Attachments
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.