My New PA Setup w/ ServODrive

[JMFahey]

Thanks a lot for the added details.
So the motor is simply driven straight by the audio signal?
Amazing.
Would love to know its DC resistance, although its impedance must be much higher, not only because it's inductive by itself, but because as a highly efficient transducer , the (considerable) mechanical impedance offered by cone, suspensions + mass of additional moving parts (including coupling belts and pulleys plus rotor inertia) *appears* as an equivalent electrical impedance at the motor terminals.
Yes, I agree, there's a lot of careful research and design behind it.
Personally I'm just intrigued by a novel idea and would love to make "something", even if it's flat only between 18 and 24Hz , simply as "proof of concept".
I am also amazed that he used rubber, because a narrow strip provides excellent extension force transmission, but very poor on compression ; I had imagined it was some kind of properly heat treated spring steel strip.
Probably the rubber strip is pre-tensioned, so it always works on traction, both ways.
Thanks again.

[FairfaxStu]

Originally Posted by JMFahey

So the motor is simply driven straight by the audio signal?
Amazing.
Thanks again.

Yes, the servo motor is driven directly by the audio signal. As for the rubber belts, they are quite rigid, surprisingly so. The mechanism is built in such a way that there is no noticable issue with them stretching. Keep in mind that the suspension on the drivers is extremely stiff. It appears to be so stiff that I could actually hit one of the driving rod ends with a hammer and still only move the cones about 1/4". I hit it pretty hard with my palm and the cones moved only about 1/8".

[JMFahey]

the suspension on the drivers is extremely stiff.

Mmmhhhh, interesting, thanks a lot for taking the time to check.
Have you ever measured the resonant frequency of those speakers?
I wouldn't be difficult, just an oscillator, a series 10 to 100 ohms resistor and the original amp.
We are not trying to measure full TS parameters here, not even close, just find a voltage peak .
Thanks again.

PS: and yes, the idea behind those speakers is fascinating: any decent "rotary motor" is typically around 90% efficient, while typical "linear motor" (VC+magnet) is in the area of 1%.

[FairfaxStu]

Well, I am a bit hamstrung by the fact that this beast is now tucked in against the wall and I cannot really move it easily because it weighs about 230 lbs. However, I was able to get the DC resistance from the amplifier side - 3.5 ohms. Given that I have a 50 foot 12 ga. cable and the internal wiring in the mix, I suspect it is slightly lower than that. I will try to report back on the resonant frequency soon. I will have to make up a cable to be able to do that so it may be a day or 3.

[JMFahey]

Hey, thanks a lot, don't worry
Anyway I won't clone one, just make an experiment.
Already have most of what's needed, including empty frames, cones, suspensions, etc.
Plus a well equipped production shop and access to many others.
I saw a home version which I found interesting, with free radiating (no horns involved) drivers, but most interesting was that they used regular fiber cones (cellulose/kevlar?, I have both) impregnated in epoxy .
Cool, because that was the original path I was going to follow.
Way Down Deep II ServoDrive Contrabass | Home Theater

[FairfaxStu]

Ok, now that I have had this setup for a couple of weeks, I have some initial impressions both bad and good.

First thing is the mechanical sound of the ServODrive. While many people have commented on this (well most everybody), I have a different take. As they commented it is most noticable at low volumes, but I say it is most noticable at low volumes on certain songs. Other songs sound fine at low volume. Obviously, it is frequency related but I have yet to quantify that. That being said, this is a sub that was designed for the outdoor PA environment, not as a sub for your home audio/theater setup.

Second note: I am sure that everybody here already knows this, but I want to spell it out. There is a reason that a setup like this is only used in a PA system. That reason is that the speakers are incredibly harsh at close range (near field) but sound extremely good at long range. My listening room is 25' by 15' and they are still harsh at times.

[JMFahey]

Thanks for posting your real world review.
I think that harshness you hear must be similar to what happens with music digitized with very few steps (as in 8 bit audio).
The mechanism would be that at low volume you run out of "steps", the carbons/brushes either touch a certain pair of rotor contacts ... or the next pair, ..... , there's nothing in between.
Some music program may mask this, some other not so well.
At low volume, sound becomes "grainy" .
But the concept is very good, and of course, at high power nothing can beat it.
Thanks for posting.

[FairfaxStu]

I should be more clear. The harshness exists not only with the ServODrive but also with my mains - the Peavey PR12's. Especially in the near field. That makes perfect sense because these speakers were not designed to play at low volumes nor to be listened to in the near field. Overall, I have no regrets (zippy, zero, zilch) in the spending the money to assemble the system because it sounds amazingly good at the intended volumes (loud).

That being said, I think I may need to augment the PR12's with another set of mains on tripods. Does anybody have a suggestion or should I stick with another set of PR12's?

[wg_ski]

The mains sound harsh beacuse they don't follow the general design guidelines that home speakers do regarding driver spacing and directivity match vs. frequency. That's what you get - position-dependent peaks and dips in the response.The further you get away the less it matters because the path lengths become equal. Higher end PA mains (and larger studio monitors) have big 2" compression drivers that can cross over at 500 hz and horn lenses with a more equal H vs. V pattern. That type of speaker will work nearfield.


Bass horns themselves have peaks and dips in their response. Look at the response of a single box in 2pi in Hornresp. It's bumpy - very bumpy. Those even out when boundary loaded (indoors) or in groups (outdoors). Of course you need to be far enough away to see those effects - and at bass frequencies the wavelength is longer so it takes more distance to do it. Even when you have a group of four of them that measures nice and smooth at 100 feet, put your head inside one of the horn mouths and have a listen. It sounds wonky, highly resonant and totally unlistenable.

[FairfaxStu]

Originally Posted by wg_ski

Higher end PA mains (and larger studio monitors) have big 2" compression drivers that can cross over at 500 hz and horn lenses with a more equal H vs. V pattern. That type of speaker will work nearfield.

You are absolutely correct. Can you recommend some higher end mains that will not break the bank? My mixer is quite powerful at 1600 watts so that will not be an issue.