Measuring driver parameters vs the data sheet values

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Just another Moderator
Joined 2003
Paid Member
I did try putting in different voice coil inductance, as that was one area where my measurements were way off the manufacturers ones, I wasn't finding it was making much of a differences in the resonance region, as expected it had a difference at higher frequencies.

Unibox does have extra le and re params but I have not tried playing with them.

Vituixcad does show the impedance, and lets you import a target impedance response. I loaded the actual and then tweaked to match it in the sim. it

edit: the biggest surprise was the qa value (basically zero absorbtion) even though I have poly bats on the walls of the enclosure. Perhaps that says more about the (in)effectiveness of the bats than the box modeling software though ;)

Tony.
 
Last edited:
I have some high inductance subwoofers that is almost impossible to simulate correctly due inductance losses, i use the old lspcad pro, basta!, unibox 3.00, and oldy loudspeaker lab for box simulation, but non does it in a trully convincing way. Unibox seems to be most accurate but the user interface is not very efficient, basta does many things right but lacks some loss features as lspcad does, so i can not get the simulation as close to real world measurements as i want to. Loudspeaker lab is more like a toy but has some nice post processing features that i keep coming back to.
Vituixcad looks very nice, but i could not get it to start, it said there where files missing, i think my computer might be to old.
I am glad you explored that boxes and losses does not always behaves as we predict, and knowing this i am sure our next loudspeakers will be better then the last :)
 
Hey everyone. Thank you all for sharing all this amazing information on measuring TS parameters. I have a similiar issue to ggidzinski. I have a Dats and a Clio Pocket. I have recently been asked to measure / verify TS parameters of a bunch of high powered car subwoofers. Ultimate goal is to recommend enclosure boxes for it. The manufacturer claims to have used an LMS to measure and I measured both with the Dats and the clio pocket. I am getting similiar VAS and FS with both tools but the QTS, QES and QMS are significantly different. Anyone have an opinion or idea as to which measurement device I should trust more? The Clio Pocket or Dats? Also I asked the manufacturer why they use hte LMS instead of the Clio (as they own a clio) and they said a clio is not accurate for TS parameters and is only accurate for Frequency response measurements. I never heard htat before. Can anyone Chime in and help me. Thanks!
 
interesting, that means you are getting different shape of the resonant peak but same resonant frequency, most common is that the shape and freq changes with drive levels and stimuli, do both use the same signal? What happens if you match drive levels? Also, you can export response curve from both programs and then do manually calc by looking at the numbers in the exported files. Lms uses stepped sine as stimul as far as i know and can be used with additional boxes for constant current/constant voltage method for higher accuracy, maybe that is why they getting different set of parameters
 
Also I asked the manufacturer why they use hte LMS instead of the Clio (as they own a clio) and they said a clio is not accurate for TS parameters and is only accurate for Frequency response measurements. I never heard htat before. Can anyone Chime in and help me. Thanks!


Now if we think a little more about the above underlined statement, it seems not to be an accurate one, especially without any evidence, and I see none so far. There is a Clio Pocket review by Joe D'Appolito where we can see the comparison of TSP for the Usher 8945A as measured by the manufacturer with LMS and the one by Joe reviewing Clio. I have added a paragraph from the same review.
 

Attachments

  • Clio Pocket vs. LMS.png
    Clio Pocket vs. LMS.png
    205.9 KB · Views: 128
I have recently been asked to measure / verify TS parameters of a bunch of high powered car subwoofers.

I've only skimmed this thread, but every 2nd comment seems to be talking T/S about things that technically aren't T/S parameters:

Thiele/Small parameters - Wikipedia

"These are the physical parameters of a loudspeaker driver, as measured at small signal levels"

"These values can be determined by measuring the input impedance of the driver, near the resonance frequency, at small input levels for which the mechanical behavior of the driver is effectively linear (i.e., proportional to its input)"

---

What happens if you match drive levels?

I'm not dismissing your answer (which seems to be very well informed), just mentioning that if the higher drive level gives a different result, then, by definition, the higher level is not giving T/S parameters.

it pulls the scalars off a different part of the T/S curves.

Ditto. If there is a "curve", then the measurement is not where the "behavior of the driver is effectively linear".

---

Me, I find (complex) mathematics non-intuitive, but pictures are easy. Looking at a graph I posted earlier (post 3):
TSP are scalar - queries

...it is hella obvious that Qts is "effectively linear" only for the left of the graph, with small signal levels.

---

Notice how, for several Morel drivers, they give 2 different numbers for Fs - one for small signal, and another for 1V?
https://www.morelhifi.com/product-category/raw-drivers/

If you look at the spec sheet of the 638Nd (for example), they give small signal T/S parameters, and their 1V equivalents for many values.

The 1V numbers are substantially more "impressive" than the small signal numbers.
 
I'm not dismissing your answer (which seems to be very well informed), just mentioning that if the higher drive level gives a different result, then, by definition, the higher level is not giving T/S parameters.

still within the small signal domain, some drivers, with high qms, reacts highly with each change in drive level. as well stimuli makes a difference too, stepped sine, sweept sine or chirp, or mls makes a huge difference on some drivers
 

Attachments

  • small_sig_vs_large_sig.jpg
    small_sig_vs_large_sig.jpg
    12.7 KB · Views: 77
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.