[richidoo]

Impulse is not the right tool to look at sound characteristic. It has a limit in analytical resolution. With our current state of the art, we already passed that limit.  You can look at it this way, impulse can be FFT (fast fourier transform) into frequency response. So impulse is related to frequency response (I am short of saying impulse response is frequency response because in my test tool, after you get a impulse response, you can then plot the full frequency response with impulse response).  Of course this is when the system is linear as described in textbook. In real world with nonlinear behaviour, impulse response of different signal strength will result in slight variation in frequency response. But interpretation of these results are not as simple.  What we need is something that goes beyond anything related to just frequency response, such as coherence (or consistency) and recovery from external disturbance...  and in short, complicated and unconventional method.   

The clarity and definition of the sound is from current portrait of small signals and large signals at the same time.

Brian, thanks for your response. I was thinking impulse would allow viewing the settling time of impulse signal, which would be obviously better than typical .7Q 12" driver and that's good enough for me. But your explanation is more valuable.

I hesitate to try the system because I once heard a second hand DIY OB sub wth GR Rhythmik paper OB servo drivers and it sounded just like any other high Q drivers in OB. My ported woofers with tube amps sounded more detailed!    Reading you and Danny explaining the potential of the technology and the degree of feedback employed I think there must have been something wrong with that setup which caused the servo feature to not work, leaving the drivers bouncing around as a normal speaker. Is this possible? Is the system vulnerable to rough handling or overheating the sensor coil, or wrong connections of servo coil to amp, etc? I guess hearing is believing, I just need to get my ears on a known good Rhythmik servo speaker...
Thanks
Rich

[ebag4]

Rich, if I remember correctly the OB Servos you heard were sold by a gentleman in your area that ultimately were shipped to George (Zybar), that pair was found to be wired incorrectly, see this thread and specifically post #45 by George:

http://www.audiocircle.com/index.php?topic=73327.40

Best,
Ed

[rythmik]

I think the Genesis solution is more up to date (compared to Velodyne), and the deltas regarding servo response time would be negligible in the real world both sonically and measurement wise. 

I am not familiar with Genesis and cannot comment on it.

I also question the point made about excursion in the three dimensional plane.  It would seem both the accelerometer and direct servo approach would react in a very similar manner.

Not exactly.  The magnetic flux of the speaker is designed such that there should be only z direction movement.  Any movement perpendicular to z direction is ignored by sensing voice coil. That is called self-alignment because the sensing coil and the driver coil are on exactly the same location and cone breakup or bending is completely irrelavent. On the other hand, accelerometer is not self aligned. How they are put together and structure break-up and cone bending changes their relative coordinates is beyond what instrument can measure. BTW, the non-z axis movement is not caused by the magnetic flux, instead it is caused by the assembly of spider/surround onto the speaker.

To demonstrate how stable our system is, first proof is we can offer this as a kit. Kit has the highest stability/fool proof requirement of all products. Second, we don't have relay in our system. There is no particular power up sequence it has to follows, which means it is un-conditionally stable.  Try that with accelerometer-based servo system.   

[JohnR]

Brian - thanks for the detailed info. What prevents the servo coil from picking up EM radiation directly from the voice coil? (If it is "at exactly the same location")

[rythmik]

Brian - thanks for the detailed info. What prevents the servo coil from picking up EM radiation directly from the voice coil? (If it is "at exactly the same location")

I guess you meant mutal inductance. Earlier approach from other people had tried to compensate for such effect. However, we just leave it in the system (by first minimizing the amount of mutual inductance using shorting rings etc which is good for sound quality anyway) and use that as the natural roll-off at the upper end extension of F/R. So far that that upper extension has been around 100hz for 12" sealed/vented and about 200hz for OB because OB has 6db/oct increase. It is an elegant design. That is why our system is very stable. Those tried to solve it with compensation would only look good on paper and then later becomes a source of long term stabibility problem.  To increase upper end extension, we use smaller diamter drivers such as the upcoming 8" drivers.

[Jonathon Janusz]

To increase upper end extension, we use smaller diamter drivers such as the upcoming 8" drivers.

Hi, Brian!

This is a very interesting piece of information that I don't think I've seen directly mentioned about the 8" servo drivers yet.     Considering that you and/or Danny have already modeled and tested this, how high up can the new drivers comfortably play sealed/OB?  I'm guessing not high enough (yet? ) to cross them directly to a carefully selected (or wave guided?) tweeter with a very low crossover point or even maybe just natural roll-off?

Slight aside, but regarding the 8" drivers at the lowest frequencies, would something like a pair of 8" sealed play comparably to a single 12", and similarly say a triple stack of 8" play like pair of 12" in OB, using the currently available amplifiers, or are the 8" drivers really waiting for the new amplifiers to optimize their use/performance?

[Danny Richie]

This is a very interesting piece of information that I don't think I've seen directly mentioned about the 8" servo drivers yet.     Considering that you and/or Danny have already modeled and tested this, how high up can the new drivers comfortably play sealed/OB?  I'm guessing not high enough (yet? ) to cross them directly to a carefully selected (or wave guided?) tweeter with a very low crossover point or even maybe just natural roll-off?

One time Brian brought over some small monitors with 5" woofers in an MTM configuration. And the little woofers were designed for servo control.

We had a passive crossover on the tweeter only and the servo amps on the 5's. We also had a passive network for the 5's that matched the crossover point used by the servo amp. And we could flip back and forth between the servo amp powering the 5's or my tube amps powering them with the passive network.

It was an interesting comparison. 

With the servo amps on those things, they had the tightest, cleanest bass you ever heard from a small speaker. They were super fast and pounded out a drum solo like no other small monitor speakers ever. However they really gave up a lot in the mid range compared to my tube amps.

With my tube amps they were musical and beautiful to listen to, but did not have the bass control.

It was easy to see that the servo system kills everything in the bottom end where there is driver exertion to control. But not so much about that. Above about 200Hz a top level amp should be used. Also, crossing the servo woofers below 200Hz keeps the crossover point out of the critical mid-range. Allowing that crossover point to start getting higher up in range starts asking for other issues to have to contend with.

[richidoo]

Rich, if I remember correctly the OB Servos you heard were sold by a gentleman in your area that ultimately were shipped to George (Zybar), that pair was found to be wired incorrectly, see this thread and specifically post #45 by George:

http://www.audiocircle.com/index.php?topic=73327.40

Best,
Ed

Thanks Ed.

[Jonathon Janusz]

Danny, maybe a silly question, but could one theoretically design the servo control circuit into a "top level amp" (read: something like the tubes that made the 5" drivers sing), delivering both the magic and the top level bass?  I know we would be all but talking about a tube-based active speaker at that point, but. . .?

Neat experiment!

. . . this just got me to thinking again that when AJ adds the servo bass to his coaxial based bookshelves. . . 

. . . and another thought comes back to mind regarding the servo amps.  Any thoughts on reconfiguring the case work for the amps from a plate to something in a component-style case?  Signal/power connections out the back, controls in the front - could make a nice bit of gear both easy to use and clean up the look in-room in some of the OB setups without the drivers being mounted in a box (that the amp is then also mounted to)?  Don't know if there would be enough of a market to justify it for the DIY crowd, but maybe something nice for OEM/commercial clients selling finished OB systems?

[Early B.]

One time Brian brought over some small monitors with 5" woofers in an MTM configuration. And the little woofers were designed for servo control.

We had a passive crossover on the tweeter only and the servo amps on the 5's. We also had a passive network for the 5's that matched the crossover point used by the servo amp. And we could flip back and forth between the servo amp powering the 5's or my tube amps powering them with the passive network.

It was an interesting comparison. 

With the servo amps on those things, they had the tightest, cleanest bass you ever heard from a small speaker. They were super fast and pounded out a drum solo like no other small monitor speakers ever. However they really gave up a lot in the mid range compared to my tube amps.

With my tube amps they were musical and beautiful to listen to, but did not have the bass control.

It was easy to see that the servo system kills everything in the bottom end where there is driver exertion to control. But not so much about that. Above about 200Hz a top level amp should be used. Also, crossing the servo woofers below 200Hz keeps the crossover point out of the critical mid-range. Allowing that crossover point to start getting higher up in range starts asking for other issues to have to contend with.

Hmmm... sounds like you guys could design a large, open baffle monitor with one 8" woofer, a midrange, and a tweeter, and power the woofer with a servo amp. Any plans for Brian to place a two-channel servo amp in a component chassis?

[Danny Richie]

Danny, maybe a silly question, but could one theoretically design the servo control circuit into a "top level amp" (read: something like the tubes that made the 5" drivers sing), delivering both the magic and the top level bass?  I know we would be all but talking about a tube-based active speaker at that point, but. . .?

Yeah, but that would be very expensive to implement and would only benefit an application of a two way speaker. If the servo control system is on drivers only having to cover from 200Hz and down then the plate amps are great.

Hmmm... sounds like you guys could design a large, open baffle monitor with one 8" woofer, a midrange, and a tweeter, and power the woofer with a servo amp. Any plans for Brian to place a two-channel servo amp in a component chassis?

There are still some issues with the amp getting too far away from the woofer. So having amps built into the speakers works out better.

[THROWBACK]

I could really use a step-by-step process for setting phase. If this has been covered before, I would appreciate the reference.

I have 3 servo-12's per side, crossed over at 80HZ through an electronic crossover (24dB/Octave, high pass and low pass). These are mated to my Sound Lab M-1, directly to the side of the M-1s, and out from the wall about 5 feet.

According to my Radio Shack Sound Level Meter, the best response is at zero phase difference; ears (better than mine) like them better at 90 degrees.

What is the best way to proceed? What should I be measuring? What should I be listening for?

The manual that comes with the speakers/amps is OK, but I am still floundering a bit.

Thanks much. 

[rythmik]

Slight aside, but regarding the 8" drivers at the lowest frequencies, would something like a pair of 8" sealed play comparably to a single 12", and similarly say a triple stack of 8" play like pair of 12" in OB, using the currently available amplifiers, or are the 8" drivers really waiting for the new amplifiers to optimize their use/performance?

We need to use a different amp for several reasons: 1) the profile of amp needs to be narrower. 2) the delay time and crossover adjustment needs to be in the appropriate range. 

I would say three of the 8" driver is slightly more than a 12" driver (but two 8" is slightly less than one 12" driver).  Those 8" drivers are really for design that requires higher crossover. The usage is to place the subs right beneath the front speaker so that it does not have phase alignment issue. That is why we need to drop the full phase adjustment range to 8ms instead of 16ms. 
 

[rythmik]

I could really use a step-by-step process for setting phase. If this has been covered before, I would appreciate the reference.

I have 3 servo-12's per side, crossed over at 80HZ through an electronic crossover (24dB/Octave, high pass and low pass). These are mated to my Sound Lab M-1, directly to the side of the M-1s, and out from the wall about 5 feet.

According to my Radio Shack Sound Level Meter, the best response is at zero phase difference; ears (better than mine) like them better at 90 degrees.

What is the best way to proceed? What should I be measuring? What should I be listening for?

The manual that comes with the speakers/amps is OK, but I am still floundering a bit.

Thanks much.

Right off the bat, your 24db/oct on both HPF and LPF requires the phase adjustment to be in the negative range.  You really need 12db/oct on the sealed front speakers, and no HPF on the bookshelf ported front speakers because the latter has more phase delay at bottom end.

If you have an AVR or prepr, these adjustment would be easy. But if you don't have one, you need to move the subwoofer to be closer.

[jtwrace]

I could really use a step-by-step process for setting phase. If this has been covered before, I would appreciate the reference.

The quick and dirty way is to wire your mains out of phase of the subwoofer.  Then you adjust the subwoofer phase to get the most bass from the listening position. 

This is NOT the best way but it can get you closer then doing nothing.  Of course, a computer, mic and soundcard will yield the best result.  However, you must do it properly too.   

[Jonathon Janusz]

We need to use a different amp for several reasons. . .

Thank you, Brian!  This and Danny's replies earlier are very helpful.  Cheers!

JT, using your quick and dirty method (reverse polarity, adjust phase), don't you listen for the LEAST sound rather than the MOST?  I might be remembering incorrectly (been a while since I've had to adjust subs), but isn't the point behind reversing polarity so that you can more easily adjust to the subjectively quietest/lowest sound instead of the loudest (easier to get to the objective "correct" value), then have it right when everything gets wired back up correctly?  Please correct/ignore if I'm remembering this wrong. . .