[rythmik]

Brian,

In your originall A370PEQ pate amp post you showed the plate amp with XO freq of max 120Hz. You also subseq recommended an XO point not higher than -6dB @ 150 cycles (fair enough).

So, the question is: Is this a another plate amp, or a new version, or  I am simply mixing things up?

An W OB bass unit with Direct Servo, EQable and good up to 200Hz  (i.e. LR 12 or 24db XO @ _max_ 200Hz)... Now _that_ I would like to hear. And eventually use.

The 120hz is for 4ohm version and practically that model can be xover at 150hz because the baffle step function needs a stepup at around 100hz and that means one can even xover at 150hz with xover point at less than -6db. You can imagine in our published response curve,  move up the subwoofer level by 3db to get baffle step function, which literally move up the xover frequency.

Now for the 16-ohms OB version, it is about 150hz without considering the OB cancellation (which is +6db/oct). If I add another slight sheving function between 100 and 300hz (again +6db/oct),  that becomes a +12db/oct positive slope and that will compesate the 2nd roll-off originally occured at 150hz or so. To accommodate this, quite a few modification needs to be done for the amp. First, a shelving circuit needs to be added. The 12/24db selector needs  to move the frequency from 80hz/50hz to 200hz/120hz, and xover range needs to change from 25/120 to 50/240hz. All of these can be modified directly to the amp.  It is unfortunate that I cannot change the decal  on the plate amp unless this configuartion gets very popular.  So in short, it is the same amp with component value changes done in here  TX.  When one place order, one needs to specify if he wants regular xover frequency 120hz or higher 200hz. 

-Brian   

[TRADERXFAN]

Danny,
What setttings did you use on the amp to get it to play flat to 20hz in your room with the ob7's?  How did you place the sub and speakers?   

I know each room is different, but thought it may cut down on some of the experimentation...

Linkwitz suggested placing the dipole woofer along the wall even distance away from the listener as the listener is to the mains speaker. "A dipole woofer should be placed near the side walls, provided the distance from woofer to listener is nearly the same as that from the midrange. Again, most likely practical limitations will have to guide best dipole speaker placement for room modes as well as for first reflections off room boundaries." He also mentioned angling this toward the listener somewhere...
http://www.linkwitzlab.com/faq.htm#Q31

Is that how you set it up?

-Tony

[Danny Richie]

In my room I had to attenuate a peak around 25 to 30Hz on the speaker set on the left side of my room and had to do no EQ to the speaker set on the right side of the room.

Damping settings at 20Hz seemed to work well for me with medium to high damping settings.

I had them about 3 to 4 feet from the side walls and about 5 to 6 feet from the rear walls.

My room is 17' wide by 23' deep and has 9' ceilings.

[TRADERXFAN]

Oh, I didn't realize you had 2 subs in your room...

Thanks.

[Danny Richie]

Actually they are part of a full range pair of speakers (all open baffle).

[TRADERXFAN]

Do you remember where you crossed them to the ob7s when you had them with the ob sub?

[Danny Richie]

When Brian was up here, and we were playing the OB-7's that he brought up, we were using a single SW-12-04 in a sealed box.

In the design that I was using the W frame SW-12-16FR's in I crossed them at 200Hz.

[TRADERXFAN]

In my room I had to attenuate a peak around 25 to 30Hz on the speaker set on the left side of my room and had to do no EQ to the speaker set on the right side of the room.

Damping settings at 20Hz seemed to work well for me with medium to high damping settings.

I had them about 3 to 4 feet from the side walls and about 5 to 6 feet from the rear walls.

My room is 17' wide by 23' deep and has 9' ceilings.

Did you have the shelving function put on the amp for these? Did you use an external x-over for the 200hz?

In the design that I was using the W frame SW-12-16FR's in I crossed them at 200Hz.

[rythmik]

In my room I had to attenuate a peak around 25 to 30Hz on the speaker set on the left side of my room and had to do no EQ to the speaker set on the right side of the room.

Damping settings at 20Hz seemed to work well for me with medium to high damping settings.

I had them about 3 to 4 feet from the side walls and about 5 to 6 feet from the rear walls.

My room is 17' wide by 23' deep and has 9' ceilings.

Did you have the shelving function put on the amp for these? Did you use an external x-over for the 200hz?

In the design that I was using the W frame SW-12-16FR's in I crossed them at 200Hz.

I designed and setup the amplifiers, so I can answer better here.  There are two subs in there. A single SW-4 sealed sub and a dual SW-16 on OB.  On the SW-4 I used xover of 60hz, and add about 1 o'clock of phase shift to xover to OB7.  The LP switch is at 80hz/24db which I think at this point is more appropriate than 50hz/24db.

The same sub also was used to work with Neo2X and I similarly xover to 40hz, but add about 11 o'clock of phase shift.

There is this shelving (20hz to 80hz) built in all plates for OB.  The only difference whether we should add another boost shelving between 100hz and 300hz for those wants to xover at 200hz. The most important thing to remember is he xover on our plate amp can open up to 250hz if the driver supposed. The decal list 120hz because the highest extension model only goes to 120hz. So it is driver inductance bound. In OB, there is a natural boost already +6db/oct. This boost will improve the xover point by another 50hz to 200hz and that is what Danny uses and he keeps the LP to 12db/ext. I do want to add a shelving of 100hz and 300hz and use a 24db roll-off instead of 12db/ext for all of those xover at 200hz.  But that is just my thought. 

[Telstar]

Hi Danny,

Could you post the Le and voice coil diameter specs?
Also, a translation of all specs to the metric system would be helpful for we european folk

[Hank]

Telstar (and everyone else), here is a shareware program that I have on my computer desktops at work and at home.  VERY convenient conversion program: 
http://joshmadison.com/article/convert-for-windows/

[Kevin R]

Danny,

Where are you mounting the plate amp?

How easy do you think the box could be veneered? I'm running a OB system with an X-Voce, Statiks, and Omni's and this sub sounds like it could be the ticket. Think I would build two of them for each side.

[Danny Richie]

Where are you mounting the plate amp?

It gets put in its own little box that sits behind the sub.

[Kevin R]

Where are you mounting the plate amp?

It gets put in its own little box that sits behind the sub.

Would it hurt to mount the 'little box' on the side of the Sub? I've got kids that would be in the room so I'm not so sure it would be wise for me to have it just laying there. I'd turn it on and they would have it turned all the way or something.

[Danny Richie]

No problem. You can mount it to the sub, behind it, on top of it, or wherever you like.

[rick57]

Hi

The 45 degree W baffle looks simple to build.

Though if we're prepared to do a bit more more woodwork, this older Linkwitz design http://www.linkwitzlab.com/woofer.htm
- saves a little space, and
- sonically - "mechanical forces cancel and only a minimum of vibration is transferred to the cabinet's outside surfaces for possible re-radiation".
"Force cancellation is important because of the large cone excursions necessary for dipole operation and the associated large reaction forces". This isn't essential, but other things being equal, is desirable.

Those particular dimensions produce a resonance at 270 Hz 
http://www.linkwitzlab.com/sys_test.htm - half way down under Dipole woofer.
As SL says, this can be attenuated with a notch filter . .
I suspect the servo woofers are deeper than those used in SL's design. A slightly larger enclosure would lower the resonance.

Linkwitz uses this design up to 100 Hz. (I was going to cross at about 120 Hz).

Any thoughts on this?
How would a notch filter be incorporated with the plate amp?s XO, or would it better be a passive notch in the operating range of the "next driver up" ie the mid-woofer?

Thanks

[rick57]

~ Bump ~ 

Brian??

Any comments on Force cancellation with this layout?

And how a notch filter should be incorporated with the plate amp's XO

Thank you

[TRADERXFAN]

http://www.audiocircle.com/index.php?topic=54875.msg490375#msg490375

I was able to get a pretty flat response from 20Hz to 300Hz with no resonance at all.

Traderxfan,

Try applying that information to a box with a wider opening (22" tall by 13.5" wide) and only 13.5" deep and see what you get. 

Rick 57,
A couple of your questions are very similar to the initial one's I had and this was Danny's response...

[rick57]

Hi Tony

The discussion in May was about H frames . . Yes smaller configurations lead to higher resonances, in dipole frames whether they're H, Linkwitz compact, W or. .

However Linkwitz acknowledges that his compact design (only 2! inches taller than the driver) cause a resonance that's large: the link I posted shows the peak is about + 11 dB. That compact style design pretty much couldn't be any more compact. As we know, he co-developed LR crossovers, so a notch filter is 2nd nature to him.

So unless I misunderstand it, a notch filter would be needed. Maybe Brian can say how would it best be incorporated his plate amp's XO?

Regards

Rick

[rythmik]

"Force cancellation is important because of the large cone excursions necessary for dipole operation and the associated large reaction forces". This isn't essential, but other things being equal, is desirable.

Danny's SW12-16FR driver has an Mms of 78g (I will ask Danny to change the spec on this web site, the old parameters were provided by manufacturer). So it is already very light weight. That is a variable Siegfried has not discussed. The reactive force is proportional to the mechanical force it needs to push the cone. The means the reactive force of SW12-16FR is already 20% smaller than a woofer with 100g Mms and only half of tose with 155g. These drivers are purposedly build for OB/IB application.   
Mms: 77.8g
Mmd: 71.9g
BL:  12.81
Re: 13.46
Qes:1.07
Qts: 0.699
Vas: 148L
Fs: 27.6hz

One thing to note about force cancellation is that the frame needs to structured such that the force tranfers at the shortest distance between them for the cancellation. If the force needs to propagate via the long panel dimension, it will cause vibration, and propagate and lose the effectiveness. 

Another thing to consider is the "structure" to cause resonance is still there. So one should not place any driver (other than subwoofer) close to it. The filtering only filters out the incoming signal. Once those are generated as energy in the form of acoustic energy or harmonic distortion, that filter has no effect at all. I personally don't like odd shape filtering too much for exactly the same reason.

Those particular dimensions produce a resonance at 270 Hz 
http://www.linkwitzlab.com/sys_test.htm - half way down under Dipole woofer.
As SL says, this can be attenuated with a notch filter . .
I suspect the servo woofers are deeper than those used in SL's design. A slightly larger enclosure would lower the resonance.

The amp is very flexible with the connector in place. It is almost like a tweaker's dream. You can add all sorts of build-in functions to it. However, there will be a nominal charge for it and I hope any circuit board can get wider use before we commit to an add on function.  Even the PCB sampling service requires 10pcs minimal and $10 each.

Linkwitz uses this design up to 100 Hz. (I was going to cross at about 120 Hz). 

Any thoughts on this?
How would a notch filter be incorporated with the plate amp?s XO, or would it better be a passive notch in the operating range of the "next driver up" ie the mid-woofer?

Thanks

As I say, try to use other method such as distance to minimize the effect. Using filtering can only fix the problem on the surface. A similar analogy is a ringing metal midrange cone. Does a filter fix the problem?  All "experts" would say it doesn't while most junior engineers would say "yes".