[OB_Newbie]

The number of exciters seems to make a difference in the HF response. With one exciter my 2'x4' gatorfoam panel goes up to 15Khz with 2 exciters the HF rolls off at 12Khz and with 3 or more it is lower still around 11 Khz. I presume the multiple exciters must be canceling out the high frequencies. I wonder what role spacing between exciters plays.

(I tried different spacing and it did not seem to make any difference)

After reading some of the earlier discussion about dipole vs. bipole, I thought I'd try something very different - exciters on both faces of the panel. Complete Fail! The sound was just awful. The exciters must be canceling out.

Well it looks like people decided to breeze over your testing and findings.  I am happy to read some logical testing, and your results are exactly what ELAC states; which is the dB coupling when using multiple exciters is less in the HF range than the LF.  Basically each additional exciter is adding more LF than HF, which ultimately equates to increasingly lower HF response.

Hey Odal, this might explain your poor HF response with 4 exciters on your plywood panels!  So much to read in this thread... so much to forget...

[hblester3]

 Just read about Rob (Ziggy) I have kept up with this for about five years, Ziggy inspired me to try some things I would have never could have thought of. We built Ziggy's Styrofoam speaker in my classroom. I have tried many of his and others suggestions over the years and as everyone has found out this can be very addictive. I am writing this as I listen to the latest just finished version with the latest suggestions from the group. Just when it looked like this thread might go away, most the time Ziggy had made a new breakthrough. He will surely be missed by his family and friend and by anyone who took the time to read this long thread. I will not be able to listen to this speakers driven by the gain-clone that I built without thinking of Ziggy. In his honor we should keep working on this I know I will.
THANK YOU ROB!!!

Hubert 

[Odal3]

Back again - been busy fixing up my basement to be a better listening room. Yes - let's definitely keep this going.

OB_Newbie - Thanks for posting the info. It's funny - this thread has tried most things, it's just a matter of finding it.

In addition to the slightly lower HF, I also see some sharp dips in the upper mids on my 4 exciter panels. Note however that the panels I posted the measurements for only had one exciter each. The close mic showed more level line in the HF vs. the sharp drop when the mic is moving away.

So I applied some pva/water mix to the front side of and sides of the 4 x 2 birch panel (didn't have time to remove the exciters and apply the glue mix on the back as well this time). Result - HF improved - especially the upper HF. Still a bit lower, but didn't show the same drop. But, the bad thing was that it created another hump just as my XPS panels that were treated with glue. This time more on the lower side (around 1000k) and the the hump was not as wide. 

So looking at my own measurements as well as others have posted, most looks like decent HF in close mic position, which then drops at listening distance (10-12ft). Is the theory for the cancellation of HF the same as the interference using multiple tweeters (or smaller full range drivers) in box and OB designs?

I read elsewhere that the rule of thumb is that drivers have to be within 1/4 wavelength of the highest freq they are producing in order to not cancel each other. So 10k Hz results in a wavelength of 1.3 inches, which would then require the limit for the spacing between drivers to be about 1/3rd of an inch.


Assuming the same applies to the panels in a similar way, but since the panels works in a different manner - where on the panels is the HF cancellation occurring to explain the drop of HF at distance? Is it the the distance to the edges on the sides, distance to corners? rounding and shape of panel? the diameter of the ring attachment between exciter and panel? OR??

One Z panel is good - two Z panels playing in stereo both measures and SOUNDS better in a normal room!

[OB_Newbie]

Hey Oda,
Good to hear from you!  In my phone so short messages today.

I always assumed that the tall panels act like a line array.  I don't know why it wouldn't as line as the panel is creating HF along the vertical axis.  But question that HF are created at the same amplitude.  So possibly more like a tapered line.  We need to research the different types of line arrays to see where these panels fall.  To me they sound and behave like a line array in regards to HF.

LF have me stumped.  I would have thought that the LF should act like a open baffle panel with a slow 6db roll off per octave but your wood panels don't measure way.  The XPS panels however seem to.  Maybe the XPS and wood generate LF differently?!?

After 30 years of building I'm trading in my SPL meter and tone generator for OmniMic!  Picked a V1 for less than $120 so I'm pretty pumped.  Hopefully measurements and a little more research will tell us more definitive what's going on with our beloved panels.

Think I said this earlier but I intend to treat these like a line array.  I will build the 2 panel array and use EQ to get a reasonably flat FR.  Hope 2 panels will be able to manage the demands EQ will be put on them.  At least get us closer.  The small panels sound fabulous as they are so do think we are not to far off.

Hey, what are you referring to about Z panels?   Are you talking 2 per side in stereo???  Do tell my man...

[OB_Newbie]

Forgot to say... ordered the Ultra version of the Dayton Audio exciters.

The thought is the the more compliant spider is responsible for the lower Fs and be more suitable for light weight panel materials.  I got them and mounted to small panels but only got a quick listen before the left panel started to make some bad vibrations and then finally died.  Was a very short listen but I think they are going to sound better on XPS than the Thruster model.

Sounded more detailed right away with no break in.  Another impression I got was that it sounded quieter.   A highs signal to noise ratio.  But it was such a short listen so have to wait for the replacement to arrive before I get a better idea of the performance but can at least say it looks promising on light weight panels anyway.  The steel spider will probably be better on ply as it can exert more force on the heavier and more rigid panel.

[Odal3]

Keep us posted on the Ultra exciters. Curious to see how they work out.

Calling them Z-panels in honor of Zygadr.

I will probably go smaller panels too make them more "acceptable" in the room - the 5 x 2.5 feels a bit large in the room.  Check out this thesis for some good theoretical background (Not really journal level research material but it is for sure a much easier read than many of the other more mathematical papers): http://publications.lib.chalmers.se/records/fulltext/154618.pdf

Some comments from this:
Fig 5.5 explains that the dip and peak we see on the LF, which per this is due to the boundary conditions - how the panel is held/clamped, etc. Looks like we need to figure out how to hold it with "fully roller clamps"

Fig 4.19 shows why the panels is so much easier to integrate in a normal room than a conventional speaker - perhaps one of the reasons we like it so much.

[OB_Newbie]

Keep us posted on the Ultra exciters. Curious to see how they work out.

Calling them Z-panels in honor of Zygadr.

I will probably go smaller panels too make them more "acceptable" in the room - the 5 x 2.5 feels a bit large in the room.  Check out this thesis for some good theoretical background (Not really journal level research material but it is for sure a much easier read than many of the other more mathematical papers): http://publications.lib.chalmers.se/records/fulltext/154618.pdf

Some comments from this:
Fig 5.5 explains that the dip and peak we see on the LF, which per this is due to the boundary conditions - how the panel is held/clamped, etc. Looks like we need to figure out how to hold it with "fully roller clamps"

Fig 4.19 shows why the panels is so much easier to integrate in a normal room than a conventional speaker - perhaps one of the reasons we like it so much.

Hey Odal,
Z-panels ... I like that!! 

Got the replacement exciter yesterday but the last 2 nights have been crazy so have not mounted it to the panel yet.

Interesting read.  Both the size and clamping method appear responsible for the peaks and dips!?!?   

To me, it looks like EQ is the only real practical method to smooth the response at this time.  But then again, narrow peaks and valleys are difficult to hear.  Looking at the simulations they are very narrow though the actual measured response is not as sharp nor narrow so EQ would not be to extreme.

Also, I did not see and explanation as to what "fully rolled" clamping method is... where is that discussed?  Does anyone understand what that means?

 

[sedge]

I too have no idea what roller is ,the only things I found interesting was the optimum panel hight and width of .95 although not sure if this is much help as the down side of this paper was that the test panels he was using , were having trouble trying to better the sound of cheep computer speakers ! So no tips on how to make a better sounding panel .

OB_ newbie
Only used two 12inch panels for one side of the stereo  setup ,but thought it worked well,the two seemed to compliment each other,but never followed it up ,always seem to get side tracked,interesting to see if you think stacking the panels is an improvement.
Steve

[OB_Newbie]

I read the paper with such excitement... listening tests of dynamic speakers against DML panels.  FINALLY, a real controlled test.  Only to find out it was against a cheap Apple computer speaker.  But then again, the DML panel was teeny tiny and thin as hell (20cm long, 17 cm wide and 1.2 mm thick).  The listening portion so disappointing...

But there was a number of things to glean out of it.  Still sifting through info...

I too was shocked at the panel ratios.  Square is bad but damn near square is optimal?!?  Say what?  Do we have more supporting evidence? Seems the NXT has also mentioned the golden ratio!?!?

I'm pumped about the array... I think the extra power handling and ability to EQ might cleanup some things that have plagued performance at least with XPS.  I bought thin birch ply to try as well... just need more exciters.

Want to order them desperately but I just received my replacement Dayton Ultra exciter and been busy and never got it mounted to the panel.  But I can say...even after a very quick listen, the Ultras appear to best the Thrusters in sound quality right from the opening signal with NO break in period!!!

With me placing an order.  If anyone oversees (Steve???) wants to do a group buy with me I'd be happy to ship them to you.  Have to think it would be cheaper this way.  They are well packaged and can be mailed in a protective envelop with no problems.  How much can that cost to mail?!?  Can't be much more than a single exciter to most places...

 

[sedge]

OB_newbie
On the parts express site if you look in( understanding exciters ) you will see they suggest the width of the panel should be less than 4/5 the hight and the exciter should be placed 2/5 of the hight and width to stop the build up of standing waves.
Years back I did a buyout of 50 el cheapo exciters which are working well for me ,I have been toying with the idea of getting some high output piezo thin film exciters to get over the problems of the spider,but haven't made up my mind yet ,Rob started this idea over on diy audio many years ago but they were very low powered then so things could be better now ,maybe?
So thanks for the offer but at the moment I'm happy with the exciters I have.
Steve

[OB_Newbie]

No problem, thought I'd ask if anyone wanted to try the new high power and clean sounding Thrusters.

If anyone else listening in wants to try any exciters pop me an PM. 

Cheers! 

[Odal3]

Sorry to get you too excited  , I was mainly reading the theory and the analytical modelling. A master thesis is normally a good summary of the theory but results of studies should be taken with a grain of salt.

The roller clamp is referring to a theoretical boundary condition used in evaluating forces and vibration in beams, plates, structures, etc. Not really sure either how to (if even possible) to implement it in practice, but here is at least a visual: https://ecourses.ou.edu/cgi-bin/ebook.cgi?doc=&topic=st&chap_sec=05.1&page=theory

(there are also roller clamps used for pipes, hoses in the medical field, etc. but I believe that is a different thing)

But the effect of clamping of the edges is real. That is how I equalize (=smooth) the LF and mid on my panels. It reduces the excessive vibrations and tightens up the bass and mid.

If you have some spring clamps at home, try put some on the edges, and measure it with and without to see the difference. If you place them right it will to some degree smooth out the peaks and dips from room and panel. It doesn't remove the issue completely but makes it better. Try for example to put one at each side of the panel around the middle- for example horizontal line from the exciter. Other positions to smooth it out is to put two clamps on opposite corners (top left and bottom right, or perhaps all four corners). Another is the top (and/or bottom) edge vertical line from the exciter.Make sure to clamp it right on the edge. Not sure if this will work with the XPS/EPS panels since it may crush them, but for the wood panels it's great. The clamping especially helped when suspending the panels in the air.

OB_Newbie - if you get a lot of vibraiton on the thinner birch board as I did, (but the boards I used were larger) you may need to add some support to make them sing. So if you're not happy with the bass, try for example to attach a thin and short wood stick on one side only of the exciter. I used one of the free paint wood stirring sticks, and this really tightened up the bass. I haven't tried it yet on the 1/4 since they sound good anyway but for the 1/8 birch or 1/2 XPS it definitely helped.

[sedge]

Oral3
I have used clamping and gluing coins to the panel edges to fill in dips in the response ,I use the rta and pinch the panel edges between my thumb and forefinger ,if you slide your fingers round the panel you can quickly and easily see the response changing ,I have thought that these places could be good mounting points but you have to be sure that adding the extra weight or objects is not compromising the sound.
Steve

[exspec]

Very sorry to hear about the passing of Zygadr... my condolences to his friends and family.

So I am still considering materials for my midbass panels (200hz-2000hz or so) and have come across this:

http://www.sterlingpromotions.ca/GatorLite-Board.html

Has anyone tried the GatorLite?? It seems like an interesting product, but recall regular Gatorboard was not popular long term. I am having trouble on committing to a sheet of something to use. Funds are tight, so cannot afford to play around as much as I would like. Going to run to Home Depot this week to grab some paint for a job, and was going to take a peek at what they had on hand.

[OB_Newbie]

Sorry to get you too excited  , I was mainly reading the theory and the analytical modelling. A master thesis is normally a good summary of the theory but results of studies should be taken with a grain of salt.

The roller clamp is referring to a theoretical boundary condition used in evaluating forces and vibration in beams, plates, structures, etc. Not really sure either how to (if even possible) to implement it in practice, but here is at least a visual: https://ecourses.ou.edu/cgi-bin/ebook.cgi?doc=&topic=st&chap_sec=05.1&page=theory

(there are also roller clamps used for pipes, hoses in the medical field, etc. but I believe that is a different thing)

But the effect of clamping of the edges is real. That is how I equalize (=smooth) the LF and mid on my panels. It reduces the excessive vibrations and tightens up the bass and mid.

If you have some spring clamps at home, try put some on the edges, and measure it with and without to see the difference. If you place them right it will to some degree smooth out the peaks and dips from room and panel. It doesn't remove the issue completely but makes it better. Try for example to put one at each side of the panel around the middle- for example horizontal line from the exciter. Other positions to smooth it out is to put two clamps on opposite corners (top left and bottom right, or perhaps all four corners). Another is the top (and/or bottom) edge vertical line from the exciter.Make sure to clamp it right on the edge. Not sure if this will work with the XPS/EPS panels since it may crush them, but for the wood panels it's great. The clamping especially helped when suspending the panels in the air.

OB_Newbie - if you get a lot of vibraiton on the thinner birch board as I did, (but the boards I used were larger) you may need to add some support to make them sing. So if you're not happy with the bass, try for example to attach a thin and short wood stick on one side only of the exciter. I used one of the free paint wood stirring sticks, and this really tightened up the bass. I haven't tried it yet on the 1/4 since they sound good anyway but for the 1/8 birch or 1/2 XPS it definitely helped.

Hey Odal!  Thanks for sharing the paper... was still a very interesting read.

I have balsa that I'm going to use to straighten the panel.  Want just enough to straighten and bring a bit of rigidity to the panel.  I intended to use the balsa at the top and bottom of the panels just backed off the edges slightly... not perfectly flush with the edge.

Have new Thrusters on the way. 

Been busy over the last week+ but hope to get back to the panels soon... killing me!

[OB_Newbie]

Very sorry to hear about the passing of Zygadr... my condolences to his friends and family.

So I am still considering materials for my midbass panels (200hz-2000hz or so) and have come across this:

http://www.sterlingpromotions.ca/GatorLite-Board.html

Has anyone tried the GatorLite?? It seems like an interesting product, but recall regular Gatorboard was not popular long term. I am having trouble on committing to a sheet of something to use. Funds are tight, so cannot afford to play around as much as I would like. Going to run to Home Depot this week to grab some paint for a job, and was going to take a peek at what they had on hand.

Hey exspec,
I think the XPS/EPS is the way to go.  Without having to handle high freq. duties a 1 inch thick panel would do a good job for you.  A 4ft. x 8ft. sheet is under $14 and they will cut it to size.  White glue is cheap.  The Dayton "Ultra" exciters sound even better than the "Thruster" but I have had problems with them... they are more fragile it seems.

I think Sedge had a good recommendation.  You should listen to the panels full-range before implementing a tweeter.  I think you will REALLY like the dipolar radiation of the panels.

[OB_Newbie]

More DML reading.  I have not read it yet but thought I'd share ... 

http://www.tnw.tudelft.nl/fileadmin/Faculteit/TNW/Over_de_faculteit/Afdelingen/Imaging_Science_and_Technology/Research/Research_Groups/Acoustical_Imaging_and_Sound_Control/Publications/M.Sc._thesis/doc/Wilfred_van_Rooijen_200104.pdf

[OB_Newbie]

DML panel experience with Baltic birch plywood 
Capturing first impressions for comparison to 1 in. thick XPS panel material of the same size and Dayton Audio exciter.  Its understood that this ply panel is far from optimized but approaching this as the control sample so that I have an understanding of the core panel properties; the platform to begin making improvements from. 

DML configuration
Panel Material : 24 in. x 30 in. 1/8 in thick Baltic birch plywood panel purchased at Rockler
Exciter : Dayton Audio Dayton Audio DAEX32EP-4 aka "Thruster".
Suspension : Sitting atop antique high-back chairs leaning against one back post and resting against a damped rubber-band.  Bottom of panels on thin socks.

Listening Impressions
As expected, Bass is improved.  Smoother and more extended.  Listening to "Misguided Angel" on Cowboy Junkies - Studio CD, as the track opens you can *very clearly* hear the heat register and the bass player to Margo's right keeping time with his foot which is picked up by the mic on the old wood floors of this 100+ year old church.  I have used this track to tune my Open Baffle bass panels and with these birch panels there is ample bass extension and output where a sub woofer typically reigns. While I resist measuring, tweaking and otherwise setting out to improve the sound immediate I hit it with an RTA and see the panel resonance frequency and resulting peek centered at 32Hz.  Impressive for a 24 in. x 32 in ply panel.  The peek is aggressive and must be dealt with; not boomy or "one note" bass but just a simple over emphasis and possibly audible panel resonances but as the exciters break in I am hearing reduced panel noise!?  The panel's physically excursion at the edges are very visible and "large" for a DML panel.  Interesting.

High freq. response is somewhat muted and expected based on others listening experience.  Immediately noticeable as a lack of clarity but as I listen more I am adjusting to the tonal balance and its still very enjoyable to listen to.  A little EQ would make the panel more balanced and listenable.  At first I thought they were lacking in micro detail compared to an XPS panel and while that might be true, I feel that they may be extracting more detail such and guitar string resonances.  Hall ambiance seems to be improved on certain recordings as well but it is too early to make and definitive statements on such small details and with the exciters still needing to be run in. 

With very apparent work to do on the panels suspension to smooth out the bass resonance it will likely change the high freq. response so hard to make any more meaningful observations in regard to frequency extremes.  But it does appear that the high frequencies will need EQ to be balanced and provide a technically solid DML panel solution.  Not that I could not listen to these panels as they sit today; the sound very nice to my ears and there is something immensely enjoyable about listening to a panel full-range without anything else in the signal path.  But given different tastes and music with ample high frequency content there would be protests in regard to its HF extension.  Not as extended as XPS but still not terribly rolled off or as obvious as Odal's measurements.  Maybe small panels help in this regard.

General Observations
- The panel is not efficient (say 4-6dB down) as XPS.  The exciters are warm to the touch but not hot by any means.
- 1/8 in thick panel is very thin which leads to panel self-noise and "drum" resonances as heard on larger XPS panels; all related to material rigidity and panel size.  On Muddy Waters "Folk Singer", "You Gunna Need My Help" where his  booming voice goes from quiet passages to very loud vocals cause significant self-noise.  More noise than what I experienced with the 2 ft. x 4 ft. XPS panels.  Arg.     
- Even with the 32Hz panel resonance, they seem more "quiet" to me. A lightweight XPS panel might be more detailed in some instances but may impart a very, very slight fuzziness to the presentation.  The ply by comparison is more  damped and results in more clarity.  Might be lacking some ultimate detail however due to this character it appears that there are other details that are allowed to emerge from the recording coming from a quieter panel... if that makes sense.  I need more time with these panels before I can say for sure what I *think* I'm hearing.  What happens after the panels are suspended may change this character but there is something different about the general presentation of the ply panel.
- These could be made to look much better than XPS panels if the bass can get sorted and the high freq. EQed (or possibly a change in panel size).  This alone has me interested in building a set of finish panels as these are going to be in the living room.

Path Forward
I like the ply panels and think they have potential and totally understand why others really liked them.  Very natural and smooth sounding. 

Curious of your thoughts on some early options I might try:

• Add balsa strips to the top and bottom edges just off the edge to stiffen the panel and prevent the panel from making such large excursions.  This ought to tighten the bass, reduce excursions helping the 32Hz panel resonance.   
  

• After reading the Master’s Thesis in the Master’s programme in Sound and Vibration (http://publications.lib.chalmers.se/records/fulltext/154618.pdf) by KUONAN LI, we can see that the suspension method can smooth out the jagged response we are getting from XPS.  So I am wondering about something similar to JGale's panel frame with automotive carpet and panels attached with Velcro or something similar to Podium's sound pads but on the panel edges.  Possibly suspend the panel with a single top/center dowel with the panel very lightly touching a thin, softly padded frame raised at various locations on the panel.  Either lightly resting on portions of the raised,padded frame or attached via Velcro is a small number of locations.  Tuning the panel so to speak.
  

• With such bass extension, possibly reducing the panel size to not only reduce panel noise but also increase HF response?  These panels seem much smoother and if we can get more HF extension they could be a viable option that will look better in the room.  Just curious, did anyone try a smaller size ply panel?  I'm still surprised at the low freq. response and either I'm adjusting to these panels and getting use to their tonal balance but these are not terribly rolled off and sound really nice.  A smaller panel (maybe halving these 24x30 panels) should increase HF extension and with the bass response I'm getting now they might only need support below 80-100Hz.  Damn, that would pretty impressive for such a small panel! 
  

Any plywood panel listeners out there?  Maybe the newer, more powerful Thrusters make plywood a more approachable material?  Any observations or thoughts would be appreciated! 

Cheers!
Rich

[Odal3]

Nice write up. Thanks for sharing the comparison. I keep coming back to the birch boards all the time as well. If you get it up the bottom from the "floor", you may tame the LF peak. Bottom needs to either be on loosely fitted spikes standing on something soft. Or even better suspend it.

How loud do you play to get the exciters warm? Even when I play my 1/4 thick boards loud, they are still cool after several hours of playing. I temporarily abandon the 1/8 due to the excessive movement and self noise you describe - so curious to see if you can successfully stiffen them. But, if you don't mind loosing some db output, the 1/4 sounds even better to me. They have zero self noise even at large panel sizes.

[Squibby]

deepest condolonces to Zyggy's family and friends....his input and vision will be sorely missed.

@OB_Newbie........I've got a single birch plywood full range speaker set up in my living room. it is being fed with a stereo signal (summed to mono) so that i only require one panel (omnidirectional and all that). it does sound extremely natural and clean, particularly with acoustic, vocals and any music which is comprised of traditional instruments (ie no synthesized or electronic). it is freely suspended using rubberized paracord and the plate on the back which is attached to the wallbracket is rubberized over its entire surface to try an reduce vibrations and reflections.

it still lacks bass on certain tracks but i'm reluctant to add a sub as this defeats the whole point of the panel and the objective we have all being trying to reach with these speakers. my thoughts tend to lean towards using slightly different panel materials and increasing the sizes.

i have a couple of thoughts regarding material and am anxious to try a composite of spruce/maple veneer on an end grain balsa core or spruce/maple veneer on carbon fibre. in theory both of these would be sufficiently rigid but lightweight to provide an ideal panel (although the carbon fibre may case issues with self-damping). bracing is also another option i have yet to explore in practice but should improve certain frequencies (HF mostly) depending on the type of bracing used (perhaps small aluminium profiles or something else which should sound a little more "tinny" compared to the warmth of the wood).

I have in my gallery a few pictures of the prototype panel in its infancy. i will take a few tonight of the panel on my wall and post them tomorrow should anyone be interested.

all the best,

Squibby.