You could well have just been measuring the T/S parameters on a different part of the curve and the 55-2421 has a more horizontal set of curves.
When i want to stiffen cone & add mass to lower Fs (and efficiency) i use multiple coats of mod-podge, front & back.
dave
While operator error is possible, 0 degrees phase is what it is... I'm not doing it manually.
am not suggestting operator error. What you said assumes that T/S parameters are scalars. They are not. They are curves. It is possible to make 2 equally valid sets of measurements from different points on the curve and get 2 quite different sets of numbers.
That your numbers are different than MCM's does not automatically invalidate them.
dave
Well I'm not sure how many different methods there are to correctly measure but I do think that the equipment/software I use very precisely finds the point of 0 degrees phase at FS and then the points of maximum inductive and capacitive phase on either side. Then a second test may be performed with an added mass or a sealed test box to measure the shift in FS. VAS/compliance is then calculated based on the change in FS. I'm not so sure you can measure something else and get valid consistent TS results.
The problem with the MCM drivers was the motor force was very weak. The 18" driver had a Qts ~.7.
Not something a different test method will ameliorate methinks.
The problem with the MCM drivers was the motor force was very weak. The 18" driver had a Qts ~.7.
Not something a different test method will ameliorate methinks.Well I'm not sure how many different methods there are to correctly measure but I do think that the equipment/software I use very precisely finds the point of 0 degrees phase at FS and then the points of maximum inductive and capacitive phase on either side. Then a second test may be performed with an added mass or a sealed test box to measure the shift in FS. VAS/compliance is then calculated based on the change in FS. I'm not so sure you can measure something else and get valid consistent TS results.
The problem with the MCM drivers was the motor force was very weak. The 18" driver had a Qts ~.7.
I'm gonna have to chime in and second Dan here. While I have has some luck with MCM drivers in the past, and they are certainly not the only ones publishing "optimistic" specs, the last set of drivers I bought from MCM were nowhere close to advertised specs. If I recall, Qes was 1 full unit higher than specified, 1.3 instead of 0.3. The drivers were also extremely light weight - suggesting that there was not nearly enough steel and ferrite in the frame and motor.
Fortunately, they have a great return policy.
Dan,
You aren't getting the point. I am saying you cannot dismiss the factory numbers because they are different than yours.
I doubt there is anything wrong with your measures You are thou just measuring a single point on the T/S curves. Measure elsewhere and the numbers will be different. 2 ways that you can shift where you measure on the curves is change the voltage drive or the weather.
As an example, here are the results of 2 measures on the same driver (+ factory for ref).
Mine, which is likely using similar kit to yours, and one done with 1 of the 2 pieces of kit the factories use. Both measures are valid (and the factory also shown, when you consider that a batch of drivers within a case can be quite large. The fostex -- which are pretty tight parameter wize -- can easily vary +/- 15%, i'd be surprised if the MCM drivers didn't have a greater variation.
You aren't getting the point. I am saying you cannot dismiss the factory numbers because they are different than yours.
I doubt there is anything wrong with your measures You are thou just measuring a single point on the T/S curves. Measure elsewhere and the numbers will be different. 2 ways that you can shift where you measure on the curves is change the voltage drive or the weather.
As an example, here are the results of 2 measures on the same driver (+ factory for ref).
Mine, which is likely using similar kit to yours, and one done with 1 of the 2 pieces of kit the factories use. Both measures are valid (and the factory also shown, when you consider that a batch of drivers within a case can be quite large. The fostex -- which are pretty tight parameter wize -- can easily vary +/- 15%, i'd be surprised if the MCM drivers didn't have a greater variation.
You are missing a couple numbers there:
SD = 62.534
Cms = 1782.56 um/N
Bl = 4.097
The MCM drivers parameters are orders of magnitude different than the driver you are using as an example. I'd venture to guess that 20 seconds of heavy breathing near that driver of yours would cause a significant shift in parameters. Not so with the MCM, it just sucks...
I measure drivers just out of the box, after a short pre conditioning, again after it's run in for several hours, and yet again after it returns to room temp. That driver was so far off that I didn't save any data, otherwise I'd have already posted it with an explanation of what specifically was the source of the deviation.
SD = 62.534
Cms = 1782.56 um/N
Bl = 4.097
The MCM drivers parameters are orders of magnitude different than the driver you are using as an example. I'd venture to guess that 20 seconds of heavy breathing near that driver of yours would cause a significant shift in parameters. Not so with the MCM, it just sucks...
I measure drivers just out of the box, after a short pre conditioning, again after it's run in for several hours, and yet again after it returns to room temp. That driver was so far off that I didn't save any data, otherwise I'd have already posted it with an explanation of what specifically was the source of the deviation.
Last edited:
What he was trying to say by "curves" is that Thiele-Small parameters change with the drive level, the test signal (sine vs. MLS etc), and so on. The exact same setup will deliver different numbers on different days. A different setup could yield noticeably different numbers, for Q in particular. Both could be "right" and they will both be "wrong" in the sense that you cannot represent a driver too precisely with one set of these numbers. It's kinda like Heisenberg!
Now, things shouldn't be off by an order of magnitude, which I doubt you meant literally. Resonance and mass and compliance I'd still expect within 10% on different setups with small signals. Q can vary more since it's calculated from that-one setup might give you 0.4 and another 0.5. Unfortunately this makes huge differences in non-sealed box calculations.
MCM
Dan, what was the fs of the 55-2954 for use as a sub-woofer?
I tried using Titebond to seal and hold some 5.5" drivers in an array. It seemed to work pretty well if the hole was routed accurately but I did not use this much because the comb filtering was too bad. I learned a lot about arrays. My present design uses a horn tweeter crossed low with the midbass time aligned and focused.
Thanks for all that have contributed.
Ted
Dan, what was the fs of the 55-2954 for use as a sub-woofer?
I tried using Titebond to seal and hold some 5.5" drivers in an array. It seemed to work pretty well if the hole was routed accurately but I did not use this much because the comb filtering was too bad. I learned a lot about arrays. My present design uses a horn tweeter crossed low with the midbass time aligned and focused.
Thanks for all that have contributed.
Ted
I regularily see larger deviations across batches of drivers from the same case on the same day measured at the same time.
dave
Hi Ted,
I'm sorry but I really don't remember. I was testing several different drivers and I just don't recall. All I really remember was being
nothing in that batch was a keeper. IIRC there were quality issues with that one too, due to a lack of adhesive at the back of the cone surround joint, so it started making clicking noises under light to moderate excursion.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.