I had the exact same problems.. and in fact I was originally reading the x axis as cycles (..and reading the Y incorrectly as well, if to a lesser extent).
What you said in your last post to me seems correct so I read it more carefully and got the very same "huh?" when looking at the x axis.
This paper gives a better idea of the process (page 8 discusses the time value):
http://www.klippel.de/fileadmin/klippel/Bilder/Our_Products/R-D_System/PDF/S1_LSI.pdf
So it looks like a lumped (frequency) or total distortion value, and the drivers were likely being tested either in free-air or its virtual equivalent with an open test baffle. So the plot is at best largely useless, and at worst - entirely misleading.
Apparently the x axis time value can also being used for music signals as well (page 5):
http://www.klippel.de/fileadmin/kli...rs/Objective_and_Subjective_Evaluation_04.pdf
Now this paper is more interesting:
http://www.klippel.de/uploads/media/Loudspeaker_Nonlinearities–Causes_Parameters_Symptoms_01.pdf
-and on page 59 you see a simulation that better represents the Suspension/Force/Inductance effects.
A problem here is that it references "all" non-linearities, but as the paper mentions there is a lot more to it than those three effects - most particularly effects generated by cone & surround (geometry and material).
The Klippel is probably useful for a driver designer - IF they now how to use it properly. But that's a big "if".. and just trying to find good information on the time-scale x axis was difficult and doesn't exactly inspire confidence.
As for it's usefulness for an end-user.. I doubt it.
Better-off just using current tools with HD sweeps and IMD tests (for non-linear analysis), but particularly over a range of spl (which seems to be rare) and most especially in a design context (or how the driver is used).
Last edited:
That just gave me an idea. Much of Tony's impressions on sound are close to what I experience.
I tend to look at it as: "how am I going to use this driver?" -and start from there.
In the case of the Scanspeak 18wu/4741t00 it's a traditional 4 ohm mid-bass driver. Its real benefit seems to be linear stroke, but even then the distortion performance is best above 90 Hz.. and if you are going to high-pass the driver (which would lower distortion by limiting lower freq. output) then what's the point? The only thing I can come up with is high-output/power midrange OR lower output dipole midrange. That's a pretty small range of operation, but it might be worth it in the right design.
Some might say higher-output/power "full-range" 2-way, but there are so many drivers available and it's quite likely that a better 3-way can be produced for less money.
Last edited:
I think it is necessary to see how close the mic is located during the test. Of course,maps with any kind of test, one needs to understand what is being done and the constrains of the results due to a test method. For example, if a laser was used to detect driver motion, then you eleminate the limitations of acoustic measurements, but then you are limited to a single point where cone modes may effect your data at the higher frequencies when you expect to look at just the effect of motor, and suspension.The Klippel is probably useful for a driver designer - IF they now how to use it properly. But that's a big "if".. and just trying to find good information on the time-scale x axis was difficult and doesn't exactly inspire confidence.
As for it's usefulness for an end-user.. I doubt it.
Better-off just using current tools with HD sweeps and IMD tests (for non-linear analysis), but particularly over a range of spl (which seems to be rare) and most especially in a design context (or how the driver is used).
Some of the data produced will aid proper selection of a driver.
EBP = 126. It ain't gonna produce much low bass... The same applies to many (most) PA woofers like B&C, etc. For a SUBwoofer, you want EBP < ~70, IMHO. The old JBL 2245H had EBP = 73, for instance, and it made for a great subwoofer for home hi-fi. The later 2241 and 2242 instead are more PA oriented: much higher EBP, and more robust if abused, but it's almost impossible to get them to go lower than 40Hz at best.
Marco
Nobody seems to be taking my idea for the Rythmik Audio GR Research motional feedback subs seriously, but I stand by it. Two 12" subs have almost as much Sd as a single 18" woofer (circa 1060sq.cm for dual 12" vs. 1200sq.cm for an 18"). Add to that the benefit of a good ampifier (dual Hypex modules) and you have a good sub system. I suggested two pairs (or more) because it will be better for serious low bass reproduction. If you think I'm crazy, what about eight 21" drivers per side (StigErik) or eight IB15s in a single sub as mentioned earlier. I still think that two pairs of RA GR Research CI 12" subs is the best option for quality and depth. Add to that the fact that the subs are paper cone units, I really think it is a winner. If I am wrong, please explain to me why (other than cost implications, but OTOH, look at the price of good 18" driver and add the price of a good amp...)
Deon
Deon
Last edited:
LAB12 or the 15" version would work for horns and direct radiator. I prefer it over a few other subwoofer drivers that I have heard. Goes lower than the typical PA driver. But I prefer a JBL 2202H above ~90Hz, although I am sure there are better options as the JBL cone is so heavy. I notice how 6.5" or 8" drivers sound snappier above 200Hz. But I don't like to cross the 12" JBL between just 100-200Hz. Even with 2nd order the sound is strangely muffled and 1st order reviels too much nastiness that the JBL has above 800Hz. Best way to use the JBL is to cross it at 400Hz and ignore my experience with snappy 8" drivers. It is a dilemma.
Above 400Hz the compression drivers kick in. With 2nd order slopes on the cone drivers below this means whatever cone I will use, I will still hear this cone driver until around 1500Hz. The JBL 12" JBL 2202H is full of garbage at 1500Hz.
Any 12" that plays down to 80Hz with confidence and has the snappiness and detail of a 6.5" driver?
Above 400Hz the compression drivers kick in. With 2nd order slopes on the cone drivers below this means whatever cone I will use, I will still hear this cone driver until around 1500Hz. The JBL 12" JBL 2202H is full of garbage at 1500Hz.
Any 12" that plays down to 80Hz with confidence and has the snappiness and detail of a 6.5" driver?
Last edited:
Actually, I think nobody is sure you gain much with a servo design taken into account the cost and effort involved when you compare them against well designed simple drivers.
Well, from all that I have read, good servo design is especially effective in the low bass. Servo feedback lowers the distortion in the lower bass quite a bit, increasing accuracy. I also have yet to read one negative comment on Rythmik Audio's implementation. I am not saying it is the absolute best out there, but at the price point, it is very hard to beat.
Deon
There are lots of parameters involved, you cannot just take any driver and expect it to work better than a well designed normal driver and a well designed amplifier. Additionally, if you have feedback where large cones start to flex, it gets a bit tricky.
And??? what were your impressions. I've personally never heard servo controlled subs.
Hi Pano
Please give me your impressions. I have not heard them. I have also read only good reviews so far, but I value your opinion highly so I would really appreciate your input. What did you think of them, sound and implementation wise? The concept seems sound, but does the implementation work?
Thanks,
Deon
I have been preaching this for years and years now. One has to be able to MOVE some air.
Above 500, a horn loaded compression driver, of course, and a ribbon for the extreme highs as well. Okay, now call me crazy. Happy listening to all
Nice! What driver do you use in the EV horns? Where do you crossover to the ribbon and what ribbon do you use?
OK, another suggestion, but this will make the LTO a sub-sat. How about a good 15" driver (Rythmik Audio DS1500 CI anyone?) in the Decware Imperial SO subwoofer horns. Then we have a horn for the upper-mids and low bass (not sure how low the Imperial SO goes though). Only problem with this is that it'll place more pressure on getting the mid-bass driver correct. Anybody got any experience with these that they can share?
Speaking of mid-bass, has anybody mentioned/considered the BD-Design BD15 15" bass drivers for the range between 700Hz and the sub? Pricing (per pair) is about in the same range as the GPA 515.
From the BD-Design site:
Some specific elements responsible for the performance of these fine sounding bass drivers are:
- Light weight and high strength and, most importantly, a good sounding cast aluminium basket.
- Light and very rigid custom made paper diaphragm.
- A diaphragm able to resist air pressure and therefore capable of actually delivering bass.
- Spiders and surrounds specified for both good sound and very long life.
- Flat wired, light weight, extremely short and underhung voice coil.
- Small but very strong "double" Neodynium magnet construction with a 14mm(!) thick pole plate.
- Nearly 100dB sensitivity with 16 ohm impedance (almost 103dB sensitivity at 8ohm when used in parallel).
- Linear and clean response without noticeable break up problems.
- Very light total weight saves on shipping costs.
Here are the specs (also from the BD-Design site):
Fs 35.696
BL 29.207
Qms 8.7300
Qes 0.2412
Qts 0.2347
Cms 0.2817
Mmd 56.400
Mms 70.557
VAS 287.79
Re 13.000
Rms 1.8127
dBSPL 99.369
SD 0.0855
P (RMS) 150.00
L 1kHz 1.3839
L 10kHz 0.7576
Z F1-F2 79.290
ZMax 483.61
X-Lin (p/p) 7,0000
Weight 4,9000
OK, I know I am a Johnny-come-lately, but with two drivers in parallel they give an 8-ohm and about 103dB/W SPL, and the Neo magnets should be a good match sonically for the Radian Neo. These were also made specifically to match up with a horn mid. Again, has anybody got any experience about these to share?
Deon
Speaking of mid-bass, has anybody mentioned/considered the BD-Design BD15 15" bass drivers for the range between 700Hz and the sub? Pricing (per pair) is about in the same range as the GPA 515.
From the BD-Design site:
Some specific elements responsible for the performance of these fine sounding bass drivers are:
- Light weight and high strength and, most importantly, a good sounding cast aluminium basket.
- Light and very rigid custom made paper diaphragm.
- A diaphragm able to resist air pressure and therefore capable of actually delivering bass.
- Spiders and surrounds specified for both good sound and very long life.
- Flat wired, light weight, extremely short and underhung voice coil.
- Small but very strong "double" Neodynium magnet construction with a 14mm(!) thick pole plate.
- Nearly 100dB sensitivity with 16 ohm impedance (almost 103dB sensitivity at 8ohm when used in parallel).
- Linear and clean response without noticeable break up problems.
- Very light total weight saves on shipping costs.
Here are the specs (also from the BD-Design site):
Fs 35.696
BL 29.207
Qms 8.7300
Qes 0.2412
Qts 0.2347
Cms 0.2817
Mmd 56.400
Mms 70.557
VAS 287.79
Re 13.000
Rms 1.8127
dBSPL 99.369
SD 0.0855
P (RMS) 150.00
L 1kHz 1.3839
L 10kHz 0.7576
Z F1-F2 79.290
ZMax 483.61
X-Lin (p/p) 7,0000
Weight 4,9000
OK, I know I am a Johnny-come-lately, but with two drivers in parallel they give an 8-ohm and about 103dB/W SPL, and the Neo magnets should be a good match sonically for the Radian Neo. These were also made specifically to match up with a horn mid. Again, has anybody got any experience about these to share?
Deon
Last edited:
Not myself, but have looked at them more than once - I know that my friend James, who worked with Bert Doppenberg to develop the Quasar Mk.1 OB speaker ( which use Supravox 285GMF) , is a big fan of these drivers. They are definitely worth investigating for a single bass driver OB project .
Thanks - good link . Reminds me I need to think about the 'issue' of the high 3-5k region on these sort of speakers . I'm not a fan of steep crossover slopes, but maybe a notch filter would be a better solution.
Thanks - good link . Reminds me I need to think about the 'issue' of the high 3-5k region on these sort of speakers . I'm not a fan of steep crossover slopes, but maybe a notch filter would be a better solution.
What have been your impressions? The only servo sub i have extended experience with is the Genesis 928 which has served me pretty well. On the other hand I have also very much enjoyed the H Frame Goldwood 18 inchers. Neither of these could be considered state of the art of course. Have head a bunch of other in show room and audio show conditions. I'd love to find something knocking on the door of the best which from reviews I have read seems to fit the Rythmik with a pair of GR Research 12's. The price seems at least reachable for someone not in the 1%. Would love to hear what your impressions were and what you consider would be better.
First it was Tonetubby AlNiCo guitar driver, then GPA 416 AlNiCo. I get the impression that there is a desire to generate a certain 'tone' here when reaching beyond, rather than neutrality of sound.
Interesting about AlNiCo magnet from GPA 416 spec sheet:
http://www.greatplainsaudio.com/downloads/416-8B%20Spec%20Sheet.pdf
It is clearly stated AlNiCo magnet is actually a signal compressor. Compression generates nonlinear distortion. Distortion = tone.
This may be the "vivid tone colors" Lynn expressed earlier describing GPA 416 driver sound.
I know I may be asking for nothing, but is there anyone measured the distortion of GPA or Tonetubby AlNiCo drivers ?
.
Interesting about AlNiCo magnet from GPA 416 spec sheet:
http://www.greatplainsaudio.com/downloads/416-8B%20Spec%20Sheet.pdf
It is clearly stated AlNiCo magnet is actually a signal compressor. Compression generates nonlinear distortion. Distortion = tone.
This may be the "vivid tone colors" Lynn expressed earlier describing GPA 416 driver sound.
I know I may be asking for nothing, but is there anyone measured the distortion of GPA or Tonetubby AlNiCo drivers ?
.