"Fortunately, I was still able to use them."
That 's the important bit.
Lots of SB drivers seem to be a bit off spec. I had some 12" SB drivers where all four (not all from the same production run) had Qts and Fs higher than advertised.
SB Acoustics TS parameters dreadfully inaccurate
Real TS values are, by definition, small signal - "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)."
(from Wikipedia)
However, SB do not measure at small input levels. So what they publish are not real TS values.
The link is currently broken:
TECH. NOTES – Sbacoustics
...but their tech note used to state that they test at 1 volt, if I recall correctly.
A home woofer tester will give true small signal levels - because they use a tiny fraction of a volt.
At a tiny fraction of a volt, you are measuring the "essentially linear" part of the curve (see picture in post 3), whereas SB are measuring further to the right, where the curve has begun to dip.
TSP are scalar - queries
If you look at post 4 in that second link, you'll see how different the values are, when comparing a small signal measurement (real TS values) and a 1 volt measurement.
That difference - roughly 20% - is the amount of "fudge factor" you should expect to apply when converting SB spec sheets into real TS values.
That 's the important bit.
Lots of SB drivers seem to be a bit off spec. I had some 12" SB drivers where all four (not all from the same production run) had Qts and Fs higher than advertised.
SB Acoustics TS parameters dreadfully inaccurate
Real TS values are, by definition, small signal - "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)."
(from Wikipedia)
However, SB do not measure at small input levels. So what they publish are not real TS values.
The link is currently broken:
TECH. NOTES – Sbacoustics
...but their tech note used to state that they test at 1 volt, if I recall correctly.
A home woofer tester will give true small signal levels - because they use a tiny fraction of a volt.
At a tiny fraction of a volt, you are measuring the "essentially linear" part of the curve (see picture in post 3), whereas SB are measuring further to the right, where the curve has begun to dip.
TSP are scalar - queries
If you look at post 4 in that second link, you'll see how different the values are, when comparing a small signal measurement (real TS values) and a 1 volt measurement.
That difference - roughly 20% - is the amount of "fudge factor" you should expect to apply when converting SB spec sheets into real TS values.
If both Fs and Qts are higher by about the same percentage, and Vas is appropriately lower relative to the square of Fs, then the driver will perform the same as if based on the published specs. There will almost always be variations in actual measured T/S values in drivers even from the same production lot, but the drivers will perform the same in a given box.
Paul
Paul
"then the driver will perform the same"
Sure.
But my explanation is still needed, because people were berating the OP for not breaking the drivers in and/or not doing the tests properly, when, most of the difference in measurements has an explanation that is SB's fault, not the OP's fault.
If SB used small signals for their tests - the way everyone else does - this issue wouldn't come up.
Sure.
But my explanation is still needed, because people were berating the OP for not breaking the drivers in and/or not doing the tests properly, when, most of the difference in measurements has an explanation that is SB's fault, not the OP's fault.
If SB used small signals for their tests - the way everyone else does - this issue wouldn't come up.
Helpful. I'll get two for the center to test before I invest in 6.
Break in time does matter though. How long seems to vary with the materials.
I use published specs only to make a purchase choice. Next I use my trusty Woofer Tester II for the model, then prototype. Sometimes the model is correct, sometimes not. VAS is always an iffy measurement. For this project, I am using sealed boxes and have the ability to oversize considerably. Might even achieve critical.
Break in time does matter though. How long seems to vary with the materials.
I use published specs only to make a purchase choice. Next I use my trusty Woofer Tester II for the model, then prototype. Sometimes the model is correct, sometimes not. VAS is always an iffy measurement. For this project, I am using sealed boxes and have the ability to oversize considerably. Might even achieve critical.
Curious. Just put the published specs for all four cones, both 4 and 8 Ohm into WinISD.
Calculated parameters match, but SPL is 3 dB lower for all 8 drivers. My experience, drivers have all worked out in use to be quite close. FWIW, going to order a pair of the aluminum cone versions for a prototype center.
Calculated parameters match, but SPL is 3 dB lower for all 8 drivers. My experience, drivers have all worked out in use to be quite close. FWIW, going to order a pair of the aluminum cone versions for a prototype center.
Great posts Hollowboy. 🙂
Audioexcite has measured numerous SB Acoustics units, it can give some points in the cloud of measurements.
Audioexcite has measured numerous SB Acoustics units, it can give some points in the cloud of measurements.
Last edited:
Alas, darn near everything is sold out. Guess I'll research the SUB upgrade while we wait for a new container from China.
Actually it looks to me like maybe one is 8 ohm and one is 4 ohm-the screenshots are on different scales and too crude for detailed answers. And HOW the impedance is measured (constant current, constant voltage, in betwee, absolute drive level) unfortunately all give different yet correct results.
But 52 Hz versus 40 Hz, something is totally wrong there. My testing experience is that break-in is overblown; once ran a full power 280 hour test for Ford and while yes resonance dropped it was not crazy dramatic (though it WAS continual, it never settled down and stopped dropping). In fact any design that drastically drops with break-in I'd consider defective.
I think either the data sheet has been screwed up OR you were shipped the wrong driver (or if labeled correctly, maybe a production screwup).
What does SB say about it?
The thing is : should you look at the VAS when the driver motor is hot ot after 24 hours rest as some advice ? With SB it seems that gives 20% VAS difference with the 1V factory measurement versus the usual little voltage measurement technic.
Goran from Audioexcite measure them after 3 hours of rest after a breaking-in for the Fs and Vas comes again close enough of their factory specs.24 hours seems more accurate from what I readed here.
But this last way from Göran way seems legit : you want to load your driver with a Vb that reflects several hours of listening sessions but if you are enjoying your speaker just 30 minutes from time to time. It could make sense to break-in the drivers one or two hours in the box close to the one targetted with a minimal filter, then wait a little hour and proceed to the measurements close from a playback condition ? 😕 ... you don't mind the Vas when the music is not playing !
Does it make sense for the forth parts air cutters ? very serious guys tell a breaking in is fast and should not be at too much high volume for a long time at X-mech limits or a little below Xmech, or xorse with a hand massage that coul push the spider beyond its natural resilience (i.e. below X-max).?
For me I will measure my SB with Arta then will do an average box I will tune with ears : a little wool to climb up the Vb or plain things to reduce it ! 😱
Goran from Audioexcite measure them after 3 hours of rest after a breaking-in for the Fs and Vas comes again close enough of their factory specs.24 hours seems more accurate from what I readed here.
But this last way from Göran way seems legit : you want to load your driver with a Vb that reflects several hours of listening sessions but if you are enjoying your speaker just 30 minutes from time to time. It could make sense to break-in the drivers one or two hours in the box close to the one targetted with a minimal filter, then wait a little hour and proceed to the measurements close from a playback condition ? 😕 ... you don't mind the Vas when the music is not playing !
Does it make sense for the forth parts air cutters ? very serious guys tell a breaking in is fast and should not be at too much high volume for a long time at X-mech limits or a little below Xmech, or xorse with a hand massage that coul push the spider beyond its natural resilience (i.e. below X-max).?
For me I will measure my SB with Arta then will do an average box I will tune with ears : a little wool to climb up the Vb or plain things to reduce it ! 😱
Last edited: