Good evening everyone
I just bought a Dayton Audio DATS V3. I know there are several previous posts regarding the use of this device. However, I just wanted to share my experience using DATS V3, and hope to get some feedback from people with more experience than me. At arrival, I quickly set up the DATS to run some T/S measurements on a handful of drivers I have laying around. These drivers are:
Scan speak 21W/8555-01 (4 pieces)
Scan speak 21W/8555-00 (2 pieces)
BMS 15S320 (measured only one)
Fostex FF85K (2 pieces)
Some observations:
Cheers
I just bought a Dayton Audio DATS V3. I know there are several previous posts regarding the use of this device. However, I just wanted to share my experience using DATS V3, and hope to get some feedback from people with more experience than me. At arrival, I quickly set up the DATS to run some T/S measurements on a handful of drivers I have laying around. These drivers are:
Scan speak 21W/8555-01 (4 pieces)
Scan speak 21W/8555-00 (2 pieces)
BMS 15S320 (measured only one)
Fostex FF85K (2 pieces)
Some observations:
- I have four 21W/8555-01, which all measures pretty much the same (very high degree of consistency). All four woofers new in boxes. I broke in two them by simply running a low a freq. signal in free air for three days before measuring them. The remaining two came right out of the box (no brake in). All four woofers pretty much measured the same (only some minor differences). For all of them, Fs is very close to factory spec (meas. Fs = 19-20 vs. fact. spec Fs = 19). However, Qts is way off (meas. Qts = 0.37-0.38 vs. fact. spec Qts = 0.26).
- For the two Scan speak 21W/8555-00, observations were similar to the observations for 21W/8555-00. Right out of the box, the two woofers measured the same (meas. Fs = 21 vs. fact. spec Fs = 20) and (meas. Qts = 0.44 vs. fact. spec Qts = 0.31).
- Measurements for the BMS 15S320 was bang on factory specs (meas. Fs = 42 vs. fact. spec Fs = 41) and (meas. Qts = 0.29 vs. fact. spec Qts = 0.28). Also, this woofer came right out of the box.
- Measured specs. for the Fostex drivers were quite different from factory spec. Higher Fs and Qts. I got these drivers from Planet10 some years ago, and they came with measurements performed by Planet10. By comparison, I got slightly lower Fs and Qts, but differences were marginal. That makes me feel somewhat confident that the DATS V3 works as intended.
Cheers
Last edited:
Hi,
Be aware that Thiele-Small Parameters (TSPs) vary with time, temperature, cone displacement and velocity. Often manufacturers publish parameters for drivers measured after breaking them in. DATv3 uses a high impedance output and a current source assumption to calculate impedance, so there are some differences from other measurement systems. The DATS HW itself is very good but it’s not intended for making lab grade measurements. That being said there are few things you can do to improve accuracy, here are a few tips for measuring small signal parameters:
Check that the DATS DCR value make sense, ideally with a 4-wire capable DMM. The DCR value/accuracy directly affects how a number of parameters are calculated. Make sure you run the measure in a quiet environment (noise moves the cone which causes measurement fluctuations). Make sure to measure at room temp (20°-25°C), as temp changes your DCR and suspension compliance.
I assume you’re collecting the TSPs to model an enclosure. Try modeling the published values and your measured values, you may be surprised how little difference it makes, but it’s still worthwhile to understand how to make better measurements.
Be aware that Thiele-Small Parameters (TSPs) vary with time, temperature, cone displacement and velocity. Often manufacturers publish parameters for drivers measured after breaking them in. DATv3 uses a high impedance output and a current source assumption to calculate impedance, so there are some differences from other measurement systems. The DATS HW itself is very good but it’s not intended for making lab grade measurements. That being said there are few things you can do to improve accuracy, here are a few tips for measuring small signal parameters:
Check that the DATS DCR value make sense, ideally with a 4-wire capable DMM. The DCR value/accuracy directly affects how a number of parameters are calculated. Make sure you run the measure in a quiet environment (noise moves the cone which causes measurement fluctuations). Make sure to measure at room temp (20°-25°C), as temp changes your DCR and suspension compliance.
I assume you’re collecting the TSPs to model an enclosure. Try modeling the published values and your measured values, you may be surprised how little difference it makes, but it’s still worthwhile to understand how to make better measurements.
Thanks for your reply Olen
For the scan Speak woofers, I found that Re obtained from DATS were approx. 6.2 Ohm. However, measuring these values using a multi-meter, I found the values to be around 5.7 Ohm (the same as the published specs.). I suspect that this difference may explain the differences in the Qts-values (but cannot be sure). I found a technical note on the Scan Speak web page outlining the technical details and the mathematical formulas for computing TSPs, I am new to this, but I will try to compute the Qts-values using their procedures. It appears to me that Scan Speak do more or less the same as the DATS unit. Thus, ideally there should not be much difference.
This is correct. I am currently making the enclosures. The volume of the enclosures is derived based on the published specs. I did model the enclosure volume using the results of the measurements. I then got a volume that was 2 liters more, so not much difference. However, there were some considerable differences in the modeled group delay, which became somewhat worse.
Cheers
For the scan Speak woofers, I found that Re obtained from DATS were approx. 6.2 Ohm. However, measuring these values using a multi-meter, I found the values to be around 5.7 Ohm (the same as the published specs.). I suspect that this difference may explain the differences in the Qts-values (but cannot be sure). I found a technical note on the Scan Speak web page outlining the technical details and the mathematical formulas for computing TSPs, I am new to this, but I will try to compute the Qts-values using their procedures. It appears to me that Scan Speak do more or less the same as the DATS unit. Thus, ideally there should not be much difference.
This is correct. I am currently making the enclosures. The volume of the enclosures is derived based on the published specs. I did model the enclosure volume using the results of the measurements. I then got a volume that was 2 liters more, so not much difference. However, there were some considerable differences in the modeled group delay, which became somewhat worse.
Cheers
What (SW) are you using to model your enclosure volume? I’m guessing it’s a ported box or TL with the lowish Qts. My thought was to try modeling the response with the published TSPs in the optimal enclosure and then using the same enclosure and model the response with the measured TSPs. For a ported box I wouldn’t obsess too much about the modeled group delay for a single point in space. I’m guessing the frequency response difference is just a fraction of a dB.
An old Radio Shack book has the same equations as the scan speak webpage but presented in a different way which may be a little clearer. It shows how to calculate Qts and how the DCR directly influences the result.
An old Radio Shack book has the same equations as the scan speak webpage but presented in a different way which may be a little clearer. It shows how to calculate Qts and how the DCR directly influences the result.
I am using WinISD to model a ported system. This is the first ported system that I am building. With a few exceptions, I always had a problem with the sound quality of ported systems. Somewhat difficult to describe, but I always had the impression that the sound at lower frequencies from such systems is “slow”, “messy” and “undefined”. To overcome this problem, while still retaining some low-end output, I decided to try a ported system with low Q woofers combined with a low tuning frequency (long port). By this, I am hoping to approach the sound quality of a sealed system.
To me it is of interest to better understand how to measure T/S parameters. Thus, I went back to do measurements of the Scan speak W21/8555-01 woofer. DATS gives an Re of 6.2 Ohm. Using a multi-meter, Re is in fact 5.7 Ohm, which is the same as the published spec. By using the formulas from the DATS documentation (similar to formulas given in your post), I managed to manually calculate the T/S parameters presented by DATS. Now, if I adjust Re to 5.7 Ohm (instead of using Re = 6.2 Ohm), Qtc reduces to around 0.33 (down from approx. 0.37. Note that the published spec is Qts = 0.26). If, in addition, I use the published spec for Fs = 19 Hz, Qts gets close to 0.30. Also, usage over time may lead to a Qts-value closer to the published spec.
cheers
Last edited: