Madisound had a special on the Faital Pro 12FH500 12" Woofers, so I picked up a couple. The measurements in DATS are quite different from the mfg published specifications. For example I measured Fs at 67 hz. and the specification is 45 hz. See my DATS measurements below.
But, both drivers measure very close to each other and I have not broken in the drivers. For my purposes I am not too troubled by the discrepancy in measurements, I was interested in experimenting with a neodymium magnet driver with high efficiency and low Q and I always planned to use measured values. But, this might suggest this driver might not provide as much LF extension as some might be expecting from the published specs.
* This data was exported from the Dayton Audio Test System: DATS
*
But, both drivers measure very close to each other and I have not broken in the drivers. For my purposes I am not too troubled by the discrepancy in measurements, I was interested in experimenting with a neodymium magnet driver with high efficiency and low Q and I always planned to use measured values. But, this might suggest this driver might not provide as much LF extension as some might be expecting from the published specs.
* This data was exported from the Dayton Audio Test System: DATS
*
- Piston Diameter = 0 mm
- f(s)= 66.82 Hz
- R(e)= 5.34 Ohms
- Z(max)= 64.63 Ohms
- Q(ms)= 4.897
- Q(es)= 0.441
- Q(ts)= 0.4046
- V(as)= 0 liters (0 cubic feet)
- L(e)= 0.9781 mH
- n(0)= 0 %
- SPL= 0 1W/1m
- M(ms)= 0 grams
- C(ms)= 0 mm/N
- BL= 0
- K(r)= 0.04387
- X(r)= 0.5812
- K(i)= 0.07071
- X(i)= 0.5486
Fairly common with prosound drivers. Fs will drop with use. I think DATS can test at different voltages as well. That might give you more accurate numbers.
I had the same issue with some SB Audience 12's. Took quite a few hours of high power use before they made the bass I expected.
I had the same issue with some SB Audience 12's. Took quite a few hours of high power use before they made the bass I expected.
It's likely that the difference is due to the fact that Faital uses Klippel which drives the speaker to more realistic voltages while DATS uses the old low-level method of measurement
Fully expected. According the the developer of the DATS it guesses teh T/S from the impedance curve, decreasing accuracy, the QC on these seems poor, and they collapse the T/S curves in a different place.
I have not had any experience with Faital Pro woofers, but in general i will use the factory numbers first for modeling.
DATS is useful for matching drivers, if you use the numbers generated you will likely not design as good a box.
dave
You are gonna do two things: break it in, and put it in a box. Break-in reduces the stiffness. Box increases stiffness. The mass is not going to change, and the BL can hardly change.
Changing the stiffness changes Fs but not mass or BL, so is not real important, except that we like to work with it. The Q (all of them) will change with Fs in predictable ways.
As a practical mechanic, I like to know Fs/Qts, a high-order figure of "merit". If I put it in a box so small that in-box Fs doubles, Qts also doubles (nearly).
DATS
f(s)= 66.82 Hz
Q(ts)= 0.4046
------------- Fs/Qts= 165
FaitalPRO 12FH520
Resonant Frequency (Fs) 50Hz
Total Q (Qts) 0.29
------------- Fs/Qts= 172
The test and factory numbers are within 5%. The worst discrepancy in real life is half a dB over about an octave. That's close enough for anybody. (If you work to sub-dB precision you want detailed EQ and a magic room).
DATS does not tell Mass? It will if you get Fs, add an ounce or two of modeling clay to the cone, and get Fs again, do math. I'd be shocked if DATS does not tell you how to do this.
Knowing actual Mass against spec Mms we can know if they gave you a heavy cone or a weak magnet. It makes little difference in the midrange but the bass may tune slightly different.
If you "need" response to 50Hz, the 67Hz driver need a lot of break-in. But 600 Watts (even 100!) at 50Hz from a hot Twelve is silly.
Changing the stiffness changes Fs but not mass or BL, so is not real important, except that we like to work with it. The Q (all of them) will change with Fs in predictable ways.
As a practical mechanic, I like to know Fs/Qts, a high-order figure of "merit". If I put it in a box so small that in-box Fs doubles, Qts also doubles (nearly).
DATS
f(s)= 66.82 Hz
Q(ts)= 0.4046
------------- Fs/Qts= 165
FaitalPRO 12FH520
Resonant Frequency (Fs) 50Hz
Total Q (Qts) 0.29
------------- Fs/Qts= 172
The test and factory numbers are within 5%. The worst discrepancy in real life is half a dB over about an octave. That's close enough for anybody. (If you work to sub-dB precision you want detailed EQ and a magic room).
DATS does not tell Mass? It will if you get Fs, add an ounce or two of modeling clay to the cone, and get Fs again, do math. I'd be shocked if DATS does not tell you how to do this.
Knowing actual Mass against spec Mms we can know if they gave you a heavy cone or a weak magnet. It makes little difference in the midrange but the bass may tune slightly different.
If you "need" response to 50Hz, the 67Hz driver need a lot of break-in. But 600 Watts (even 100!) at 50Hz from a hot Twelve is silly.
@PRR Thank you, that is helpful.
DATS has a procedure for computing Vas from the impedance sweep when a known mass is attached to the cone and a second sweep is done. So, I think it would compute mass as well. Of course I'd have to fire up the WindDoz to confirm.
Vance Dickason in "Loudspeaker Design Cookbook" describes an Efficiency Bandwidth Product which is fs/Qes and how that indicates driver compatibility with sealed enclosures by including the effect of strong motors, Qes in this ratio comparison. It is only a very brief discussion and made within the context of sealed box design and related to some guidelines about the transition from acoustic suspension to IB.
Your measure of "merit" appears a bit similar and something I may be chasing without really knowing it, yet.
I must confess to an intuitive exploration of these drivers as the subtleties of the Qc, Qts and Fs interactions are escaping me. I am interested in the Altec 515 style of drivers with low Q (.3) and massive magnet motors. These drivers have good extension into the midrange, high sensitivity and often lauded for tonal character. But, they are expensive.
If I'm reading the trend correctly, it appears that some of these pro drivers have interesting qualities for flea amp audiophile applications. High sensitivity with inexpensive Nd magnets, low Q which allows for higher Q cabinets and amps with low damping and I don't think it's unusual for the vintage drivers to be in cabinets without very low extension. Faital comes up often in DIY discussion and the sensitivity and motor appeared interesting to me for the very low price.
Considering the possibility that DATS is not measuring correctly, sets me back with respect to attempting to follow a cookbook recipe, but I still think these drivers have something to teach me with respect to my current interests.
Cheers,
Jamie
DATS has a procedure for computing Vas from the impedance sweep when a known mass is attached to the cone and a second sweep is done. So, I think it would compute mass as well. Of course I'd have to fire up the WindDoz to confirm.
Vance Dickason in "Loudspeaker Design Cookbook" describes an Efficiency Bandwidth Product which is fs/Qes and how that indicates driver compatibility with sealed enclosures by including the effect of strong motors, Qes in this ratio comparison. It is only a very brief discussion and made within the context of sealed box design and related to some guidelines about the transition from acoustic suspension to IB.
Your measure of "merit" appears a bit similar and something I may be chasing without really knowing it, yet.
I must confess to an intuitive exploration of these drivers as the subtleties of the Qc, Qts and Fs interactions are escaping me. I am interested in the Altec 515 style of drivers with low Q (.3) and massive magnet motors. These drivers have good extension into the midrange, high sensitivity and often lauded for tonal character. But, they are expensive.
If I'm reading the trend correctly, it appears that some of these pro drivers have interesting qualities for flea amp audiophile applications. High sensitivity with inexpensive Nd magnets, low Q which allows for higher Q cabinets and amps with low damping and I don't think it's unusual for the vintage drivers to be in cabinets without very low extension. Faital comes up often in DIY discussion and the sensitivity and motor appeared interesting to me for the very low price.
Considering the possibility that DATS is not measuring correctly, sets me back with respect to attempting to follow a cookbook recipe, but I still think these drivers have something to teach me with respect to my current interests.
Cheers,
Jamie
@planet10 Dave, I'm pleasantly surprised to have you contribute to this. I had hoped these numbers might spur a more interesting discussion and while I am disappointed to hear that DATS may not be as reliable as I had hoped, I am so pleased to have attracted this feedback. I wasn't even really seeking help, and now I have already received feedback to improve my thinking in general.
But, this does set me back a bit, and I am still a bit confused and curious. For example, I would assume the the Fs value would be rather robust given it only requires finding the impedance peak and those have coincided with my microphone measurements. Of course having just one T/S parameter doesn't a simulation make.
In any case, I am happy to borrow your knowledge and experience and start my models with the mfg data. It is a freeing idea to trust those numbers. I wanted to see what DATS would do and I had some old drivers I was curious about, and I was also curious to see how much break in might change the numbers since I've rarely seen anyone post a measurement in any debate.
Cheers,
Jamie
But, this does set me back a bit, and I am still a bit confused and curious. For example, I would assume the the Fs value would be rather robust given it only requires finding the impedance peak and those have coincided with my microphone measurements. Of course having just one T/S parameter doesn't a simulation make.
In any case, I am happy to borrow your knowledge and experience and start my models with the mfg data. It is a freeing idea to trust those numbers. I wanted to see what DATS would do and I had some old drivers I was curious about, and I was also curious to see how much break in might change the numbers since I've rarely seen anyone post a measurement in any debate.
Cheers,
Jamie
I would say that you just found out why these drivers were being sold on the cheap by Madisound... the Fs and Qts variation is a bit towards the extreme but definitely not unexpected. IIRC something like up to 20% variation is within the realm of possibilities, or perhaps there was a manufacturing flaw or mistake. For example, I once bought a "high excursion 10" subwoofer" and when I tested it I could bottom it out with very little cone movement. I phoned up the distributor and took one apart to find that the cone had been glued to the voice coil former in the wrong place or something like that. The entire batch was manufactured incorrectly, but that would only have showed up under certain testing, which was not done by the build house.
I would just reconsider how you use them. Avoid the area around resonance, e.g. use from 100Hz up to almost 1kHz, and let something else handle the low bass.
I would just reconsider how you use them. Avoid the area around resonance, e.g. use from 100Hz up to almost 1kHz, and let something else handle the low bass.
I have yet to see measuring kit for diyers that generates the same numbers as the factory data. S+L WT2 is probably best, but the enclosures i have done based on numbers generated by it i would like a second try, something that has been very rare when using factory numbers and the one instance i recall was not the alignment, but a too narrow a box.
I use factory numbers to design, the numbers i measure for QC.
dave
Depends on the driver, but the breakin is more important for gently easong the virgin spider into a condition suitable for the best performance of the driver.
dave
I apparently derived a similar idea (without publication or extended thought) before Dickason. (Many insights are "in the air" a while before someone nails them down.) I used Qts for no very good reason, Qes may be more what we want; but I was working with very high BL drivers so Qes ~~= Qts and dominated all other Q factors.
Fs looks easy to measure but--
- it varies with drive level (came up in this year's Burning Amp Festival),
- for most boxen, stiffness is a very minor parameter so the maker does not control it like pounds of pulp or Neo.
Once you put a speaker in a box to stiffen it the response depends on air more than doped canvas or pulp-rim. In a horn the acoustic resistance 'should' dominate driver stiffness in the range where it won't rip itself apart. Vented boxes can free the cone over a narrow band.
@PRR. Thank you again, these observations are really helping me at the moment about the topics I'm studying. I want to use a mid horn for a two way, and I'm trying to study high efficiency drivers that extend to the mids and have low Q values for tonality. All of the compromises for a two way intending mid horn and passive xover can be overwhelming, so I'm trying to start my experiments with drivers that are a good potential match to a horn from 800 hz. to 1200 hz. The pro drivers crossing over to Diy like the Faital's and the new build GPA Altecs tend to have lower Q, I think. So, that's the animal I need to learn to tame.