Examples of DIY motional feedback?

[Mr Evil]

I've been doing some casual experimentation with motional feedback (MFB) recently. Using a piezo transducer ripped out of a telephone I've been able to get it working, showing a measurable effect.

As I do more experiments so the system improves, but I get the feeling I must surely be duplicating work done by others. However I can find almost no evidence of other people doing DIY MFB. There are one or two examples and a bit of theoretical information on the web, plus some mention of such systems on this forum, but no details.

So I ask: How many people here have done this before, how did you go about it and what were the results? I'm particularly interested in any examples with piezo transducers, this being a simple and cheap method that can be applied to any driver.

[SY]

I used actual accelerometers- piezo looked like too much work. I've got some new ones, courtesy of AD, and they'll be on deck sometime after I get the two amp projects on my bench finished, out of my lab, and into my living room.

Some simple signal conditioning and days spent playing around with feedback percentage and delay got the whole thing working. If I had to do it again, I'd be doing a lot of it digitally. Behringer is a gift to the speaker design community.

[tvi]

ou might find the www.mfbfreaks.nl of some interest, mainly about philips mfb systems.

hope this helps

regards
james

[Mr Evil]

Originally posted by SY
I used actual accelerometers...

I wanted to use an accelerometer at first too, but the only ones I can find are expensive and have significant mass. Plain piezo transducers on the other hand can be had for free.

Originally posted by SY
...If I had to do it again, I'd be doing a lot of it digitally. Behringer is a gift to the speaker design community.

That would be ideal, but in my case I want to do it as cheaply and simply as possible, so analogue it is.

Originally posted by tvi
ou might find the www.mfbfreaks.nl of some interest, mainly about philips mfb systems.

hope this helps

regards
james

An interesting site. Unfortunately the language barrier keeps most of it from me, but there was one article in English about a system very similar to the one I have set up at the moment.

[SY]

Yeah, accelerometers aren't cheap. But for amateur one-off purposes, they can often be gotten as samples (the route I took). And compared to the cost of the woofers I mounted them in... The new SMD packages are pretty light, though I'd still restrict them to big, heavy-coned drivers. The piezo may well be a better solution for most drivers if you can do the signal conditioning close by.

[Coolin]

Originally posted by SY
Some simple signal conditioning and days spent playing around with feedback percentage and delay got the whole thing working. If I had to do it again, I'd be doing a lot of it digitally. Behringer is a gift to the speaker design community.

Hello Sy,

Dont tell me theres a one box solution for this?

What specific model are you talking about using and modding i geuss?

[supra]

There was an article/design in the English version of Elector about 10-15 years ago. It used piezo as sensor. I built it and used it for quite a few years- had awesome bass with a 10" woofer in a 7' high concrete pipe.
Haven't got the article now though- it will be in the library here , and a similiar search in your country might produce it.

MickM

[Nelson Pass]

I think the accelerometer glued to a cone is not the same
thing as a piezo tweeter used as a microphone, if I infer
the original post. Both work if you know what you're doing,
but one reads cone acceleration and the other air pressure.

What have we got to play with? Accelerometers, microphones,
current + voltage feedback, and lasers. I've tried everything
but the lasers (although I do have a nice one....)

Oh, and I have made some money out of dual voice coils.

Any other ideas?

[mandat]

Try to follow the link below.
Microphone Feedback to correct Woofer

[Mr Evil]

Originally posted by Nelson Pass
I think the accelerometer glued to a cone is not the same
thing as a piezo tweeter used as a microphone, if I infer
the original post. Both work if you know what you're doing,
but one reads cone acceleration and the other air pressure.

What have we got to play with? Accelerometers, microphones,
current + voltage feedback, and lasers. I've tried everything
but the lasers (although I do have a nice one....)

Oh, and I have made some money out of dual voice coils.

Any other ideas?

Yes, a pieo transducer glued to the cone will detect air pressure as well as acceleration, but I don't think it matters, since pressure is proportional to acceleration, so the output from the sensor will be much the same either way.

Did you have any success with current feedback? I tried it but it's problematic because the phase of the cone motion flips 180 degrees relative to current at resonance.