Several members correctly pointed some common pitfalls in measuring TS parameters (especially with low current/high impedance technique), so here is my version on how to measure TS parameters:
1. Break-in drivers for 24 hours with pink noise (preferable) or sinusoidal signal. Signal must be strong enough to cause cone movement up to Xmax.
2. Precondition the driver for 2 hours with expected signal, to rise the voice coil temperature to the expected normal working level. Immediately after that measure the voice coil resistance Re (Rdc).
3. Measure the driver impedance vs frequency with "Constant Voltage Mode" (CLIO): use low-ohm resistor (0.22 ohm or so) as a sensing resistor in series with the driver. Use low output impedance solid state amplifier. Signal voltage should be at the expected working level (1W or so for home hi-fi and much more for sound reinforcement drivers).
4. Compute TS parameters from the measured driver impedance.
I have used this technique manually and with CLIO (I have no experience with Dayton DATS or Smith&Larson WT or such). Apart from internal port and box resonances, simulated and measured SPL vs frequency graphs are almost identical.
1. Break-in drivers for 24 hours with pink noise (preferable) or sinusoidal signal. Signal must be strong enough to cause cone movement up to Xmax.
2. Precondition the driver for 2 hours with expected signal, to rise the voice coil temperature to the expected normal working level. Immediately after that measure the voice coil resistance Re (Rdc).
3. Measure the driver impedance vs frequency with "Constant Voltage Mode" (CLIO): use low-ohm resistor (0.22 ohm or so) as a sensing resistor in series with the driver. Use low output impedance solid state amplifier. Signal voltage should be at the expected working level (1W or so for home hi-fi and much more for sound reinforcement drivers).
4. Compute TS parameters from the measured driver impedance.
I have used this technique manually and with CLIO (I have no experience with Dayton DATS or Smith&Larson WT or such). Apart from internal port and box resonances, simulated and measured SPL vs frequency graphs are almost identical.
Last edited:
I was not referering to one driver vs another, but DATS measured parameters vs the manufacturer's data because of too different measurement procedures, especially measurement level is too low with DATS/WT2 to give acurate values: Fs to high, Qts to high, and Vas to low.
I have to reflect on what I wrote here. Not only using a more compliant surround can be used to lower the Qes of a driver. The Bl product can be increased, too to decrease Qes. SB Acoustics may have chosen to go this way, and the specification on their site really indicate that the 8 Ohm version of their speakers have higher Bl than the 4 Ohm version.
Sorry for the inaccuracy of my former post.
I didnt read every post, but in my personal experience the added mass method sucks at getting it right. I now build a small sealed box every time I measure TS parameters. Yes its much more work but its much more accurate. And thats with me havibg used a lab calibrated scale to measure neo magnets weight, so that part was right.
Just IME.
Just IME.
Wish this thread cd be buried. The title kinda gives the impression it affects the whole SB range
Last edited:
They seem to be fine drivers. The SB Acoustics woofers I have sound good and appear to be well made. But the wrong Qts thing does seem to apply to a lot of them.
Some of the tests showing this were done by Vance Dickason who (literally) writes books on the topic. So either:
- everyone, including the experts, is measuring them wrong.
- they often have a high Qts.
Anecdotally, this is an email I got from a small distributor (background: in 2015, I bought a DATS and speakers from them. My new SB drivers were off spec, whereas some new Peerless drivers had measured perfectly to spec):
we set up a DATS test system here and tested some drivers ourselves, we found that our stock of SB drivers are out by a fair way but we don't have any vifa or peerless left here to test.
Published measurements that you can check yourself:
Zaph|Audio - SB12.3 3-Way Tower
Slightly higher than spec.
I do note that Zaph says "I always use high level delta compliance testing because it is much more accurate compared to delta mass testing"
www.audioexcite.com >> SB Acoustics SB17NAC35-4
Substantially high:
Spec is 0.32, measured is:
0.4 (Smith & Larson Audio WT2 woofer tester)
0.42 (Dayton Audio WT3 woofer tester)
Test Bench: The SB42FHCL75-6 15” Subwoofer from SB Acoustics
Slightly higher than spec.
VOICE COIL
A Satori 6.5” Midrange from SB Acoustics and a MBX Mid-Bass Woofer from B&C Speakers
By Vance Dickason
Qts variably over spec.
5 years ago, it appears they had some where Qts was on spec or low:
SB29NRX75-6
SB34SWNRX-75-6
(both in Voice Coil, 2012)
Last edited:
T/S parameters are curves not scalars. When you collapse them to a number its value depends on where you pick it off the curve.
All those measurements, factory & others can be accurate despite being different.
If they are all the same then you have horizontal T/S curves.
The title of this thread is unlikely the case. Use the same measuring system & context as SB you will get within unit-to-unit variation.
dave
OK. And maybe they chose to "pick it off the curve" in a different way to Zaph, Vance Dickason, the Smith & Larson Audio WT2 woofer tester and the Dayton Audio WT3 woofer tester.
It may be good for consumers to know that.
Agreed. I can accurately say that I can go faster than 1,000km/hour on a penny farthing. I'd have to be going off a cliff in a vacuum, but given that measuring system and context, it would be true 🙂
But that doesn't help much if you're trying to decide whether to buy a penny farthing.
Silliness aside, the thread title is a bit overwrought.
FWIW, you have to add enough mass to lower Fs by at least 1.56x [or raise it in a box], so a few quarters worth is only going to work consistently on a tweeter 😉.
GM
I always start with the factory numbers and have had few situations where i need to tweek things.
dave
I have now returned the drivers to SB, and they have tested them. What they found is that the measured TS parameters are different to those on the datasheet, but quite close to those on the datasheet for the SB17NAC (natural aluminium, not black) version. It seems that the NBAC datasheet is preliminary, and not correct.
My measurements were still quite different (higher fs and Qts) even from the NAC data, so there is still a discrepancy. They also did measurements at a much lower level (0.1V), and that gave values much closer to my results. This pointed the finger at the DATS using too low a signal level, but it is set to the default of +4dBu (1.23Vrms) so it should be OK.
As a sanity check, I also tested some Scanspeak Discovery woofers I had purchased previously (18W-8434G00) with the DATS, and they measured very close to the SS datasheet. Interestingly, Ulrik at SB said that whereas SB always use a constant voltage to measure their drivers, Scanspeak always use constant current.
My measurements were still quite different (higher fs and Qts) even from the NAC data, so there is still a discrepancy. They also did measurements at a much lower level (0.1V), and that gave values much closer to my results. This pointed the finger at the DATS using too low a signal level, but it is set to the default of +4dBu (1.23Vrms) so it should be OK.
As a sanity check, I also tested some Scanspeak Discovery woofers I had purchased previously (18W-8434G00) with the DATS, and they measured very close to the SS datasheet. Interestingly, Ulrik at SB said that whereas SB always use a constant voltage to measure their drivers, Scanspeak always use constant current.
What do we need to know Qts with 2 digits precision for? To nail a perfect QB3 or SBB4 BR alignment? 🙄
Most diyers can't stand "flaws" like a bumped response on their sims, but won't bother to check if the box they built matches or not their sims, nor probably hear any difference between a flat or slightly bumped or shelved response.
Maybe a non intentionaly bumped response will be prefered by most people to that ideal targeted flat response. Btw, most comercial designs are intentionally bumped, not maximally flat, to match mainsteam consumer taste.
Most diyers can't stand "flaws" like a bumped response on their sims, but won't bother to check if the box they built matches or not their sims, nor probably hear any difference between a flat or slightly bumped or shelved response.
Maybe a non intentionaly bumped response will be prefered by most people to that ideal targeted flat response. Btw, most comercial designs are intentionally bumped, not maximally flat, to match mainsteam consumer taste.
I have a DATS v2 system and there is no need to speculate on the size of the internal shunt resistor in the potted sound device.
I sent a 1kHz sine through it and measured 1.4V output into no load. Then I placed a 10Ω resistor across the alligator leads and measured 125.54mV. Here's how you figure out the internal resistor value (Rout):
Vnoload = 1.4V
Vload = 0.12554V
Rload = 10Ω
I sent a 1kHz sine through it and measured 1.4V output into no load. Then I placed a 10Ω resistor across the alligator leads and measured 125.54mV. Here's how you figure out the internal resistor value (Rout):
Vnoload = 1.4V
Vload = 0.12554V
Rload = 10Ω
That would give an internal series resistance of 101.5 ohms and an output current of only 12.554 milliamps. Now would that be sufficient to drive a large woofer and get accurate T/S parameter measurements ?
C.M
C.M
SB explains in their papers their measurements are made at 1V.
Btw I don't think you get more accuracy at 1V than at 100mV, simply you get different results.
There are no absolute true data beyond measurement conditions.
Btw I don't think you get more accuracy at 1V than at 100mV, simply you get different results.
There are no absolute true data beyond measurement conditions.
I had another message from SB to say that they had also tried the DATS V2 and were able to replicate my results, which were very close to the results they got from their in-house system at 0.1V.
So the SB drivers are quite sensitive to drive level when it comes to TS para measurements, but the Scanspeak Discovery units are not.
So the SB drivers are quite sensitive to drive level when it comes to TS para measurements, but the Scanspeak Discovery units are not.