Who makes the lowest distortion speaker drivers
- Thread starter mtoc
- Start date
What about Doppler distortion? Must be present and must be in relation to the amplitudes? Might be related to the passband of the driver or at least the audibility of it?
Doppler distortion can be reduced by using a true horn which lowers amplitudes for the same output loudness.
B.
Essentially, there is no way to produce clean sound by shaking heavy cardboard and copper coils at thin air. Just compare the distortion of electrostatic full-range and film tweeters to Rice-Kellogg cone drivers. And with power-madness gripping the audio world, moving masses are getting, well, massive, moving toward half-pound coils to handle (at least in marketing blurbs) big watts.
B.
B.
)since you mentioned "reduced" my suggestion is definitely valid as well. Horns with compression drivers do definitley have less FM distortion but they have increased 2nd order harmonic distortion due to the nonlinearity of the air in the compression chamber.
Regards
Charles
The Purifi drivers have amazing specs. I have yet to buy a pair. I tested an assortment of 7" drivers a few years ago and the Seas driver had the lowest distortion. The Dayton Reference drivers did well. My documentation isn't perfect. Some of the tests were done with a current drive amplifier, some with voltage drive. For smaller dirvers Aurasound 2" driver used by Linkwitz in the Pluto has very low distortion as well. I was targeting 0.1% distortion and was able to get close to that at usable listening levels with those drivers.
| Driver | Diameter | Two Tone (dB) Closest, Highest Harmonic | T Decay -30dB, Current Drive | T Decay -30 dB Voltage Drive |
| RadioShack | 6 | -38 | ||
| Tang Band | 6.5 | -45 * | ||
| ScanSpeak Illuminator | 7 | -42, -42 | 1.5 ms | 0.8 ms |
| Seas W18EX001 | 7 | -50, | 1.2 ms | |
| Dayton Ref | 7 | -48 | 0.8 ms | |
| Seas W22EX | 8 | -50, -42 | 1.5 ms | |
| Dayton Ref | 8 | -48 | 1.4 ms | |
| HiVi M8N | 8 | -38, -38 | 1.1 ms |
* Higher SPL
Last edited:
Hi Charles
That is not exactly correct. Air is nonlinear (weakly) and as such will create both harmonic as well as IM (intermodulation). They just use two different signals as "detectors". Both be very weak since the nonlinearity is weak.
If only you'd move beyond Brylcreem and a sundial on your wrist to tell time, there is a hell of a lot of advances in the last 50 years.
A couple of years ago, there was a lot of marketing wank about the Peerless STW 350F 15" subwoofer, with it's massive 7.4"coil, and an employee from Tymphany, Mr. George Bullimore was purveying fallacies about the product.
Nick from Stereo Integrity purchased one off the shelf and had it sent to Warkwyn for Klippel testing. It's suspension was the limiting factor for distortion, at a whole 5.95 mm one way. Nick sent one of his own drivers at the same time for Klippel testing, the HST 15, - it mopped the floor with that POS Peerless product.
Mr. Bullimore tried in vain to prop up his product, but he ran and was never heard from again.
Ad copy read - capable of 45.5 mm of cone excursion before any major loss in motor strength. ( Bl )
Nick from Stereo Integrity purchased one off the shelf and had it sent to Warkwyn for Klippel testing. It's suspension was the limiting factor for distortion, at a whole 5.95 mm one way. Nick sent one of his own drivers at the same time for Klippel testing, the HST 15, - it mopped the floor with that POS Peerless product.
Mr. Bullimore tried in vain to prop up his product, but he ran and was never heard from again.
Ad copy read - capable of 45.5 mm of cone excursion before any major loss in motor strength. ( Bl )
Attachments
Aside from your personal insult attack, can you actually name any way in which a 15-inch driver today has features which make it audibly better than a Stephens 150W (woofer) made in 1965 with a resonance of 20 Hz, a very light cone, substantial alnico magnet with hole in the middle, cast frame, edge-wound rectangular coil wire*, and accordion surround.
B.
* like the famous Western Electric horn drivers of 1935
Last edited:
Audibly better..... Well since flat native FR is not a necessity these days with the advent of DSP and FIR correction, modern 'stoves' of a couple of KW's worth of power, the only things left are suspension linearity and longevity, cone strength, and motor topology.
Split inside/ outside wind coils these days have much flatter Bl curves, drivers that are machine assembled have the same unit to unit consistency in terms of adhesive quantities. Shorting rings ! Le/x curves that are flat.
Neodymium and Ferrite do not have the same issues that Alnico does. A large mechanical shock from shipping could render your 1965 Stephens a paperweight. Also, a large burst of energy can cause partial demagnetization. It's likely down about 10% in gauss in the gap or more from age, unless you have had it recharged.
With a 20hz Fs and a light cone, ( how many grams? ) that is a very, very weak suspension, but don't worry, I am sure the doping on it has caused the suspension to become almost rock hard in the past 57 years since it was born.
Since you always love analogs, and deify Lotus, have you ever driven on a race track with a soft suspension vehicle and next to zero damping? Regardless if you put your beloved carbureted Stephens in a sealed box to give it some restorative force, it's still not going to track an input signal like a modern driver. Be sure to set the points first and sychronize those carbs. Maybe even pull out the bleach to give your bias plies a good burnout first.
Anytime you want to wax poetically about your Stephens, just pull out the Klippel results for it.
Split inside/ outside wind coils these days have much flatter Bl curves, drivers that are machine assembled have the same unit to unit consistency in terms of adhesive quantities. Shorting rings ! Le/x curves that are flat.
Neodymium and Ferrite do not have the same issues that Alnico does. A large mechanical shock from shipping could render your 1965 Stephens a paperweight. Also, a large burst of energy can cause partial demagnetization. It's likely down about 10% in gauss in the gap or more from age, unless you have had it recharged.
With a 20hz Fs and a light cone, ( how many grams? ) that is a very, very weak suspension, but don't worry, I am sure the doping on it has caused the suspension to become almost rock hard in the past 57 years since it was born.
Since you always love analogs, and deify Lotus, have you ever driven on a race track with a soft suspension vehicle and next to zero damping? Regardless if you put your beloved carbureted Stephens in a sealed box to give it some restorative force, it's still not going to track an input signal like a modern driver. Be sure to set the points first and sychronize those carbs. Maybe even pull out the bleach to give your bias plies a good burnout first.
Anytime you want to wax poetically about your Stephens, just pull out the Klippel results for it.
What you list is either audibly trivial and/or incremental or relates to manufacturing or durability*.
OK, I admit it: it was a trick question. I was wondering if the Stephens 150W is audibly and visually distinguishable from a couple of top of the line JBL drivers today (and for a real good reason, as some with a grasp of audio history may know)?
B.
* maybe not durability since folks are buying and selling old Stephens today, ahem, ahem.
This is an unproductive question to say the least. Who is supposed to know what a Stephens whatever-the-hell is supposed to sound like, let alone what its measurements are?
Objection, CONCLUSION.
Audibly trivial? How about you send off your Stephens to have it Klippel'd? Tell them to take it easy, since it won't really survive much power. Then once and for all, we can compare it to today's drivers instead of calling up conjecture, old nostalgia and belief.
People are still selling Ford Model T parts too. I suppose that makes the Model T durable?
https://www.modeltford.com/
https://www.modeltford.com/item/6200REB.aspx
You can buy a rebuilt Holley NH carb for a model T for $135.00 if you so desire, per the above link, $ 220.00 if you don't have a core.
I'd be interested in doing that; can you link us to nonlinear distortion measurements of these drivers under standardized measurement conditions? IMD would be great as well, but is rarer to find.
Am I the only one here reading this like; the vast majority of the world audio transducer industry are meekly trying to catch up to the epic ‘Stephens’ driver from a bygone era?