Percy

Hi guys!

It's been a while since I posted. Thought I'd run an idea by the gurus, e.e.'s here.

As you all know, Velodyne has stopped manufacturing their servo sub system - the worlds most accurate sub. What if the following were implemented in a suitable sub...

2 accelerometers, one to monitor cone motion and the other to monitor any motion from the sub basket. Accelerometers are fairly cheap. Most complex part is using a controller similar to the Vel control box. OR, just use the Velodyne box itself.

Feasable?

Percy

TTurboPimp

iTrader: (12)

WOW, I thought I knew my *****....you might as well be speaking Russian to me after reading that post.

jmecbr900

I would think that the servo's are not the difficult part, I think that the processing of the readings is going to be more difficult. If you can adapt the Velodyne unit to suit your needs, I don't see why it wouldn't work. Theoretically it should since all your doing is making your own servo and letting the control box figure it out and adjust accordingly. It probably wouldn't be a bad idea to get the tech specs on the servos in the Velodyne unit to make sure your servos match them.

engin_ear

iTrader: (1)

Hi Percy!
Yes, I agree with jmecbr900. Analog Devices makes some pretty sweet silicon MEMS-based accelerometers that can be used, but that is not the hard part. Once you get them glued down, it's what you do with their outputs (which become the inputs to your control circuit) in order to correct the signal to the driver. This is NOT trivial (even though it may seem so), and unless you have something to match up with from prior development work, you wind up potentially having to do some control system analysis to make sure that your servo is stable under all conditions, over frequency, and that its error correction time and amount are appropriate, etc, etc. If this is not done correctly, you can introduce more distortion than you're trying to eliminate, heh heh. If you're willing to experiment and you have the time generate some hardware and write some code, you might be able to use a PIC or a cheap DSP chip.
Here's another outrageous idea: How about taking the controller out of a Velodyne home subwoofer and adapting it for car sub use? Might be easier, but you still need to dig out the design details to see which accelerometer is appropriate....this may not be easy to copy, though, since the chips and code in the thing are probably proprietary.

Percy

Retro,

Surprisingly enough, nothing in it is even remotely DSP based! Just a few logic chips and comparitors as well as some buffers. Would be neat to find someone that can program a DSP to do the same thing - it shouldn't be too hard, plus the parts count will go way down.

Easiest way of adapting a Vel into a car that I can think of is to get a home Vel, take it out of the box and use it in the car. The AC can be taken from a very large power inverter, thus using the built in amp (200 to 300 wpc) that the Vel came with. Quick and dirty. Anyone ever try this?

Percy

engin_ear

iTrader: (1)

Percy - not Retro here, 'tis Jerry. :-)
Surprised to hear that the controller isn't too complicated. Sounds like it would be easy to clone, seems possible to trace the circuit out, given the discrete nature of the chips. You could also try capturing the signals on a digital scope and/or logic analyzer - that would tell you what exactly is being measured and corrected and the timing of it all.

jmecbr900

I've heard the Velodyne unit.

Do you think it's worth that kind of trouble to net those results? Couldn't accurate imaging and sound be produced in less difficult a manner?

Just playing devil's advocate here.

engin_ear

iTrader: (1)

I know that Percy is a very big fan of the Velodyne and its servo correction. For me personally, I think a good quality SQ sub can be great at low to moderate volumes without the need for any correction, and some even at high volume, but at huge volumes I think any sub would benefit from having the correction circuit, since you're mechanically whipping the cone around so fiercely, and the laws of inertia can't always be compensated for by the speaker manufacturers, along with the other characteristics that they're trying to achieve in a given woofer. While I haven't heard the Vel unit, I think it is, or was, geared to the low/moderate volume arena, not the huge W7/IDMAX/etc volumes. However, yours is a good question for debate.

As far as imaging, that does not come into play here, since woofer freqs are non-directional. All imaging comes from the mids and highs. Your ears use the timing/phase differences at higher frequencies as localization cues to determine location angle.

Percy

Jerry,

Give the Velodyne a listen at a nearby audio shop. Even at low levels, it's amazing the detail that can be picked out of bass!

Percy

jmecbr900

I have listened to them personally. They are accurate, but only a calibrated mic is going to REALLY tell any inherent difference at most sound levels. Yes it's better than most, but at what price? Most SQ competition vehicles are judged not with a mic only, but judges ears too. That's my question, is it really worth the kind of intense R&D work needed to reproduce the now discontinued model?

Again, just playing devil's advocate.

You are absolutely correc engin. I should have used a different term besides "imaging" for the sub.

Percy

jmecbr900 (aka dev's advocate) ,

With both the car version and the home version, 70dbA is more than loud enough to notice the difference...at least for me anyways. No "golden ears" here...just a knack for hearing and listening to what's going on. Home version is the F1200R while the car version is DF12SC. I've noticed that the newer Vel home subs weren't up to par for some reason. Changeover started happening around '98. Maybe trying to save parts cost on the circuitry?

R&D work is pretty much "on the board"...just add servos. That's what I was trying to go for...how well it can be adapted to other subs.

Percy

jmecbr900

Gotcha. I follow you.

But wouldn't you agree that you will still need to do the R&D to mate up the servos IF you are gonna build one from scratch copying the Vel unit? At 70db, couldn't you build a sub setup for substantially cheaper than a Vel unit that sounds as accurate? I know YOU would definetly have the skills to, based on your history here. So, isn't it fun to dream about it but actually more logical to do something a little less elaborate.

I just think that matching the servo units, which may not be expensive by themselves, to some kind of dampening/processing is going to require some pretty wicked engineering. Isn't that just the reason why Vel units cost so much money?

engin_ear

iTrader: (1)

I think the Vel costs more money (both car and home) not because the engineering is so difficult, but because they are one of the few companies to do the servo thing with a good driver (and amp/enclosure, in the home case). It winds up being black magic to the casual observer, and thus commands more money.
Yeah, Percy, they appear to be succumbing to the forces that other companies die from, that is, cheapening their stuff for mass marketing, and thus gradually falling out of favor with high-end audiophiles.

"At 70db, couldn't you build a sub setup for substantially cheaper than a Vel unit that sounds as accurate?"
Yeah, I think so. If you're not pushing the woofer too hard, many of today's overdesigned woofers will be QUITE accurate at THAT level, at least the more efficient ones.

Me, I'll still take a good W7 or IDMAX and skip the extra engineering - leave that for the home stuff. I'd still rather have cleaner radio signals, even with satellite radio.

RON430

Hey Percy, yet another one for the crank files. How about taking one of the ANR chips (pretty cheap now)and playing with the software a bit and let it drive another sub to damp out what you don't want from the primary driver? In addition to cleaning up a specific driver it might also compensate for different installations. Scary.

engin_ear

iTrader: (1)

Hi Ron,
Assuming I understand your original premise here, in that you're referring to TWO separate sub drivers, one primary and one secondary, you can drive the second one with the correction signal, but you have the problem of it being physically displaced from the first, so that correction will only happen where the second woofer's correction output is exactly in-phase (or out of phase, in this case) with the output from the first. At other locations, the correction will not be accurate, and will even make the original signal worse. You also can have the problem of two different, unmatched time responses if the two drivers are not the same brand/type.