For VAS, we only need measure downshift in Fs. Nothing
else "bad" that coins or vertical orientation may otherwise
goof the rest of VAS sweep data matters or is retained.
Only concern is cone acceleration faster than gravity.
Even if that should happen, wouldn't matter! So long as
does not happen at frequencies below the unshifted Fs.
WT3 test power isn't enough for this to be a problem.
You worry bout affecting accuracy in ways that have
no significance to measurement of VAS. And that all
that this step does with added mass, nothing more...
Last edited:
Hi there: The 1st try at Vas measurement provided a different value (66.6936) from the manufacturers (95.075) and a calculated value (78.604), however all of the other measured values on this run were also different from the origional tests of driver #2 and #1 (Fs#2=49.12, Fs#1=48.78, Fs=52.49, etc). So how are we to expect these methods to translate into acurate data for simulations? ...regards, Michael
Well my experience with the WT3 has shown that an added mass equal to or more than the Mms gives very repeatable results. Whether the results are accurate I can't say. But they are consistent. Smaller masses give me a different number every run.
I tend to run the measurements with the driver vertical, clamped in a vise. Works for me.
I tend to run the measurements with the driver vertical, clamped in a vise. Works for me.
I don't have good temperature control in that room - after running the window AC unit all night the first fs I get is ~60Hz. Yesterday's fs came pretty fresh off a Fedex truck and that room and drivers were quite a bit warmer than a few minutes ago. Seems to me TS need to be taken at a specific room and driver temperature to get a reference. Once the cone moves enough to drop BL, things are into chaos anyhow. Is there a temperature standard for gathering T-S? with AN10 I felt more comfortable getting Vas with a sealed box but out of maybe 60 speaker enclosures I have, none are simple boxes to hold an 8. (i precariously balance drivers with their magnet on the rod of an old Atlas mic stand to reduce effects of reflections)
Last edited:
BL hangout gots nuthin' to do wit Fs of sprung weight.
Measure of allthing harizonteel, cept-for added mass.
Obfuscate verti VAS result appropo to pig in a poke.
Remember that WT3 will throw away all data from the
added mass pass, except FS downshift. Why care the
vertical orientation's misrepresentation of BL??? BL is
computed from Fs shift, and the previous pass.
Temperature (and humidity?) probably does matter.
Cold wet coins, go figure... Use a milk crate...
Measure of allthing harizonteel, cept-for added mass.
Obfuscate verti VAS result appropo to pig in a poke.
Remember that WT3 will throw away all data from the
added mass pass, except FS downshift. Why care the
vertical orientation's misrepresentation of BL??? BL is
computed from Fs shift, and the previous pass.
Temperature (and humidity?) probably does matter.
Cold wet coins, go figure... Use a milk crate...
Last edited:
re:BL - I meant it doesn't make much sense to have faith say in a reflex and single driver like AN, etc. and TS alignments when the cone is traveling much on a dynamic passage. Earthworks has a demo cd with drums - but its one drum tapped at a time which is easier to get an illusion than a complex passage.
BL is whack if the gap is void, or the field shifts.
But consider Nirvana low Le! Surely that copper
nosecone must short the full length of motion.
Shorted BL may not be as chaotic as you think...
But consider Nirvana low Le! Surely that copper
nosecone must short the full length of motion.
Shorted BL may not be as chaotic as you think...
The AES paper I have lists 25 deg C (77 F) +/- 5deg C (68-86 F) on a IEC baffle that varies based on driver frame size.
For me, sealed is the only way to measure Vas since I'm really not interested in small signal measurements.
GM
Traditionally, raising Fs 1.56x in a sealed cab was considered the ideal, so it seems reasonable to me that to measure Vas using mass loading it needs to be enough to lower it 0.641x.
GM
Ideal for what exactly? I suspect those constants merely
made someone's math easier elsewhere in the equation.
made someone's math easier elsewhere in the equation.
The ideal box acoustic compliance (net Vb) for a given set of driver acoustical-mechanical properties. Not sure how it was arrived at though. AFAIK I've never seen all the data used to come to this conclusion, but extensive acoustic measurements done in Bell Labs anechoic chamber were most likely a part of it. For sure, it's the mean between two extremes, but if I've ever seen them I didn't realize it as such.
Even if you're right, it doesn't negate the fact that up to a point of diminishing returns, adding more weight should increase Vas measurement accuracy and that since the mass increase ratio is based on radiator size (Sd) and moving mass (Mms), a few coins may be fine for a small, high Fs driver, but I don't see how it can be for larger and/or lower Fs units if repeatable accuracy is desired.
GM
Even if you're right, it doesn't negate the fact that up to a point of diminishing returns, adding more weight should increase Vas measurement accuracy and that since the mass increase ratio is based on radiator size (Sd) and moving mass (Mms), a few coins may be fine for a small, high Fs driver, but I don't see how it can be for larger and/or lower Fs units if repeatable accuracy is desired.
GM
reading though a few posts it looks like adding enough mass to shift Fs down a half-octave is considered good - then do the Vas delta-mass test. I'll have to buy a thermometer. Sealed box is a good way to go. GM - do you do large signal Vas tests? - (not sure what that means in the world of single drivers) - -Novak had his way of porting boxes.
J D'Appolito recommends Fs be increased at least 1.5x in a sealed cab, but only decreased >25% with the added mass method, so I assume the latter was derived empirically.
1 W originally simply because this was Altec's standard measurement power, though later I switched to 2.83 V for ease of comparison between the older 16 ohm and newer 8 ohm drivers, so a relatively large signal. Don't see why it would be too high for 'FR' drivers, i mean it seemed to work fine for me, but then I never tried to see how small the cab could be made to meet a given app since it limits tuning options.
Yes, it's seems to be basically a simple way to implement Thuras's theoretically ideal mathematical model. For sure, it was a Godsend for folks like me that couldn't do the complex math. The only downside is it being based on a Fb = Fs tuning, so while you can get a usable net Vb for high Qts drivers, tuning is strictly empirical.
GM
1 W originally simply because this was Altec's standard measurement power, though later I switched to 2.83 V for ease of comparison between the older 16 ohm and newer 8 ohm drivers, so a relatively large signal. Don't see why it would be too high for 'FR' drivers, i mean it seemed to work fine for me, but then I never tried to see how small the cab could be made to meet a given app since it limits tuning options.
Yes, it's seems to be basically a simple way to implement Thuras's theoretically ideal mathematical model. For sure, it was a Godsend for folks like me that couldn't do the complex math. The only downside is it being based on a Fb = Fs tuning, so while you can get a usable net Vb for high Qts drivers, tuning is strictly empirical.
GM
- Status
- Not open for further replies.