Hi everyone,
What level of accuracy have people had with Speaker Workshop when measuring TS parameters?
For example, I recent bought two Audax HM100Z0 drivers. They are specified at....
Fs = 64Hz
Vas = 4.72l
Qts = 0.49
But they measured at (run-in for over 24 hours)....
Fs = 85Hz, 83Hz
Vas = 2.34l, 2.49l
Qts= 0.57, 0.55
I've measured several other Audax & Peerless drivers and Fs is always higher than spec whereas Vas is consistently lower than spec. Qts, Qms and Qes are all off spec but are still within the tolerances driver manufactures accept.
I'm running the 1.06 version of SW (the latest) with a Wallin Jig 2 and everything is calibrated. Impedance plots of the 4.7R and 16R calibration resisters show very good accuracy and linearity with respect to frequency.
Any ideas?
Nice one,
David.
What level of accuracy have people had with Speaker Workshop when measuring TS parameters?
For example, I recent bought two Audax HM100Z0 drivers. They are specified at....
Fs = 64Hz
Vas = 4.72l
Qts = 0.49
But they measured at (run-in for over 24 hours)....
Fs = 85Hz, 83Hz
Vas = 2.34l, 2.49l
Qts= 0.57, 0.55
I've measured several other Audax & Peerless drivers and Fs is always higher than spec whereas Vas is consistently lower than spec. Qts, Qms and Qes are all off spec but are still within the tolerances driver manufactures accept.
I'm running the 1.06 version of SW (the latest) with a Wallin Jig 2 and everything is calibrated. Impedance plots of the 4.7R and 16R calibration resisters show very good accuracy and linearity with respect to frequency.
Any ideas?
Nice one,
David.
Thanks Francois,
That's reassuring but I've measured six different Audax drivers (3" HT080G0, 4" AP100Z0 & 5.25" HM130C0) as welll as one pair of Peerless drivers (831858) and they ALL show higher Fs and lower Vas than specified. Is that your experience?
I've always accepted 30% as being the tolerences on drivers but the Vas measurement on my AP100Z0 is 100% off!! Am I doing something wrong? I used the added mass method (using coins for weight) with the magnet side of the driver clamped down on the bench.
Nice one,
David.
That's reassuring but I've measured six different Audax drivers (3" HT080G0, 4" AP100Z0 & 5.25" HM130C0) as welll as one pair of Peerless drivers (831858) and they ALL show higher Fs and lower Vas than specified. Is that your experience?
I've always accepted 30% as being the tolerences on drivers but the Vas measurement on my AP100Z0 is 100% off!! Am I doing something wrong? I used the added mass method (using coins for weight) with the magnet side of the driver clamped down on the bench.
Nice one,
David.
Yes, I see that often.
To give you an example, here are the specs for a Peerless 850102.
manufacturer specs:
fs 74.2
vas 3.4
qts .43
measured specs:
fs 102
vas 2
qts .52
(after my usual 100 hours of break-in)
I know, 30% is big but as long as all your drivers are close to each other this is not a problem. Cheap drivers may have 30% off the manufacturer specs but worse they may have 30% off between each other.
To give you an example, here are the specs for a Peerless 850102.
manufacturer specs:
fs 74.2
vas 3.4
qts .43
measured specs:
fs 102
vas 2
qts .52
(after my usual 100 hours of break-in)
I know, 30% is big but as long as all your drivers are close to each other this is not a problem. Cheap drivers may have 30% off the manufacturer specs but worse they may have 30% off between each other.
David - How much difference does this make to your enclosure design? Personally, i usually use WinISD and put the manufacturers specs in, then fiddle with volume and tuning until I get something like what i want. But, once I've built the cabinet i use the calculated port length as a rough guide and just hack bits off until it sounds right (I always trust my ears over my computer). I'm just interested to see how much of an effect this difference in measured and manufacturer data has on your preferred enclosure size?
Higher Fs and lower Vas are two things that tend to cancel each other.
The following is a WinISD simulation of the two sets of numbers, stated and measured, for the Audax HM100Z0. this is for a sealed box of 4 liters.
The stated numbers are in red.
The measured numbers are in blue.
I would call this pretty good agreement with the stated specs once this thing is put into a box. Calling the F3 of the measured spec 103 Hz, I would say that the stated spec gives an F3 less than .2 of an octave lower than the measured spec.
Put another way, where the stated specs are 3 dB down, (90 Hz), the measured specs are 4.5 dB down.
For good makes like Audax and Peerless, the measured and stated specs do tend to work out this way. This is good for ported and sealed boxes. For transmission lines, maybe not so good.
The following is a WinISD simulation of the two sets of numbers, stated and measured, for the Audax HM100Z0. this is for a sealed box of 4 liters.
The stated numbers are in red.
The measured numbers are in blue.
I would call this pretty good agreement with the stated specs once this thing is put into a box. Calling the F3 of the measured spec 103 Hz, I would say that the stated spec gives an F3 less than .2 of an octave lower than the measured spec.
Put another way, where the stated specs are 3 dB down, (90 Hz), the measured specs are 4.5 dB down.
For good makes like Audax and Peerless, the measured and stated specs do tend to work out this way. This is good for ported and sealed boxes. For transmission lines, maybe not so good.
Attachments
And just to further illustrate, here is Francois' Peerless 850102 in simulations of both published and measured specs. This is for a sealed box of 2.5 liters
The published specs are in red.
The measured specs are in blue.
At 121 Hz, where the published specs it should be 3 dB down, it is actually 4.5 dB down. This is not even taking into account a possible SPL shift with the difference in numbers.
This squares with my experience with Peerless too. It also agrees with the experience of Madisound, who measures the drivers when they come in. The Peerless model I ordered from them had different specs, but similar performance to published specs when placed in a box. In that case, it was the opposite direction-the speaker had lower fs, but higher Vas. But it all worked out when put into the box. (Actually, optimum vent tuning would shifted slightly with the new numbers-but even then, the old vent tuning would have worked OK.)
In a better world, the published specs would be a lot closer to the actual specs. But in this world, I guess we can live with the thing just working out reasonably all right in the end.
The published specs are in red.
The measured specs are in blue.
At 121 Hz, where the published specs it should be 3 dB down, it is actually 4.5 dB down. This is not even taking into account a possible SPL shift with the difference in numbers.
This squares with my experience with Peerless too. It also agrees with the experience of Madisound, who measures the drivers when they come in. The Peerless model I ordered from them had different specs, but similar performance to published specs when placed in a box. In that case, it was the opposite direction-the speaker had lower fs, but higher Vas. But it all worked out when put into the box. (Actually, optimum vent tuning would shifted slightly with the new numbers-but even then, the old vent tuning would have worked OK.)
In a better world, the published specs would be a lot closer to the actual specs. But in this world, I guess we can live with the thing just working out reasonably all right in the end.
Attachments
I'm, indeed, building a TLine, and I'm without means to measure my drivers; that's why I bought Vifa.
Care to comment, O Kelticwizard? Or anyone?
Vifa P17WG00-06
Fs=35Hz
Qms=1.34
Qes=.58
(Qts=.4)
Vas=49.4
Xmax=3.mm
I got these numbers off the Vifa Catalogue .pdf file that Madisound has on their site. Is there a place to find a "real" measurement?
Dave
Care to comment, O Kelticwizard? Or anyone?
Vifa P17WG00-06
Fs=35Hz
Qms=1.34
Qes=.58
(Qts=.4)
Vas=49.4
Xmax=3.mm
I got these numbers off the Vifa Catalogue .pdf file that Madisound has on their site. Is there a place to find a "real" measurement?
Dave
I wouldn't care to predict on the relationship between a TL line and Theile-Small parameters, because I haven't built a Transmission Line. This forum is lucky to have two members, Planet 10 and Martin J King, (MJK), who have their own websites devoted to Transmission Lines. Those are the guys to talk to.
I just excluded the Transmission Line from my posts above becase in the reading I have done, I think it is possible that the relationship might be between the speaker's Fs and line length-that the balance between Fs and Vas might not work out quite so nicely as it does in the ported or sealed box. But again, I do not know for sure.
Are you buying your speakers from Madisound? Some years ago, when I bought my Peerless there, they had started the process of measuring the Thiele Small parameters themselves for the speakers they received. Not individually for each speaker, of course, but averaging the group. If you buy from Madisound, give them a toll-free call. They still might be doing that.
I just excluded the Transmission Line from my posts above becase in the reading I have done, I think it is possible that the relationship might be between the speaker's Fs and line length-that the balance between Fs and Vas might not work out quite so nicely as it does in the ported or sealed box. But again, I do not know for sure.
Are you buying your speakers from Madisound? Some years ago, when I bought my Peerless there, they had started the process of measuring the Thiele Small parameters themselves for the speakers they received. Not individually for each speaker, of course, but averaging the group. If you buy from Madisound, give them a toll-free call. They still might be doing that.
I've been in touch with MJK (that guy knows his T-Line). Thanks for the Madisound advice.
I'll let the forum know how my T-Line project turns out--
even if it's a HORRIBLE failure.
Thanks,
Dave
I'll let the forum know how my T-Line project turns out--
even if it's a HORRIBLE failure.
Thanks,
Dave
Redeye,
Here's a comparison of box sizes & -3dB freqs for a spec'ed Audax AP100Z0 and the two drivers I measured. All boxes are sealed with Qtc = 0.71....
Spec'ed driver: Vb = 4.4 liters, F3 = 90Hz
Measured #1: Vb = 4.3 liters, F3 = 108Hz
Measured #2: Vb = 3.8 liters, F3 = 108Hz
I'm actually using 3 litre sealed boxes which increases the Qtc to around 0.74 and 0.76 respectivly. Ported boxes would be more unforgiving of mismatched parameters!
The box size and driver parameter issues have never really concerned me - I just wanted some feedback on the accuracy of Speaker Workshop's parameter estimation routines.
Nice one,
David.
Here's a comparison of box sizes & -3dB freqs for a spec'ed Audax AP100Z0 and the two drivers I measured. All boxes are sealed with Qtc = 0.71....
Spec'ed driver: Vb = 4.4 liters, F3 = 90Hz
Measured #1: Vb = 4.3 liters, F3 = 108Hz
Measured #2: Vb = 3.8 liters, F3 = 108Hz
I'm actually using 3 litre sealed boxes which increases the Qtc to around 0.74 and 0.76 respectivly. Ported boxes would be more unforgiving of mismatched parameters!
The box size and driver parameter issues have never really concerned me - I just wanted some feedback on the accuracy of Speaker Workshop's parameter estimation routines.
Nice one,
David.
- Status
- Not open for further replies.