Alvipet.
yes, I was coming to this.
different panel materials react differently but generally a very small hole with a little damping in, can be very beneficial .
of course you could just cut out the centre and use a phase plug, but this would rob you of the HF from the coil area.
And of course my idea of cutting out the centre of the coil and replacing with some sort of dome works well too.
I usually make a hole no bigger than about 2mm if not using a dome.
Depending on the panel material, I use various fillings .
Heavier panels tend to use heavier ,bigger blobs.
Very light panels such as eps, I tend to use very thin light strips of used dish sponge foam, which is softer.
And must not be jammed in the hole!!!
I was surprised that I had to use the blu-tack and not the light foam for the 1mm corrugated cardboard as I thought it was a very light panel.
So it is a combination of the small hole and the type of damping used that smoothes the response.
But it is always a good idea to test using the foam first , if you think you have a very light panel material , just to make sure you keep the mass down.
Too much mass will make the panel sound softer or smoother, but will rob the panel of some of its dynamics.
So be warned.
Steve.
yes, I was coming to this.
different panel materials react differently but generally a very small hole with a little damping in, can be very beneficial .
of course you could just cut out the centre and use a phase plug, but this would rob you of the HF from the coil area.
And of course my idea of cutting out the centre of the coil and replacing with some sort of dome works well too.
I usually make a hole no bigger than about 2mm if not using a dome.
Depending on the panel material, I use various fillings .
Heavier panels tend to use heavier ,bigger blobs.
Very light panels such as eps, I tend to use very thin light strips of used dish sponge foam, which is softer.
And must not be jammed in the hole!!!
I was surprised that I had to use the blu-tack and not the light foam for the 1mm corrugated cardboard as I thought it was a very light panel.
So it is a combination of the small hole and the type of damping used that smoothes the response.
But it is always a good idea to test using the foam first , if you think you have a very light panel material , just to make sure you keep the mass down.
Too much mass will make the panel sound softer or smoother, but will rob the panel of some of its dynamics.
So be warned.
Steve.
Alvipet.
Yes I think I tried using masking tape at one point .
If you are using a hard panel type like ply , you could try my method of using a 1inch max strip of printer paper .
I'll post a picture , it's a lot easier.
You have to get it in the right position which can be very fiddly .
But the edge of the strip will be somewhere near the middle depending on panel and exciter ?
I have to look at the response on the RTA in real time to make sure it's in the right place.
Using my ears so close to the panel does not work.
I just quickly stuck this on to show the general position , it does not have to be tight up against the panel , in fact I used to put spacers to hold the paper off the surface slightly.
I did at one time just have a square inch of paper held at one side of the exciter With masking tape ,which also worked well ,I think ?
it's not just that it can flatten the frequency response, but mainly it improves the air and dynamics a little.
The paper becomes a bit like a passive radiator as well as blocking some of the cancellations in the central area ?
Steve.
Yes I think I tried using masking tape at one point .
If you are using a hard panel type like ply , you could try my method of using a 1inch max strip of printer paper .
I'll post a picture , it's a lot easier.
You have to get it in the right position which can be very fiddly .
But the edge of the strip will be somewhere near the middle depending on panel and exciter ?
I have to look at the response on the RTA in real time to make sure it's in the right place.
Using my ears so close to the panel does not work.
I just quickly stuck this on to show the general position , it does not have to be tight up against the panel , in fact I used to put spacers to hold the paper off the surface slightly.
I did at one time just have a square inch of paper held at one side of the exciter With masking tape ,which also worked well ,I think ?
it's not just that it can flatten the frequency response, but mainly it improves the air and dynamics a little.
The paper becomes a bit like a passive radiator as well as blocking some of the cancellations in the central area ?
Steve.
Attachments
Steve
While you explained the techniques with central mass, I was thinking about the drawing below of a Manger driver. I thought a mass was attached at the center but in fact it is more sophisticated with a kind of central suspension but seems the mass effect is here.
About Tectonic fixation, here is below what I have in my notes. See at the beginning of this Tectonic video "deeper in DML"
The EQ is also documented. Around the topic we are speaking about : the high shelf at 7kHz (most of our panels have a loss here) and the dip at 12kHz where you propose a mass as counter measure.
I would be interested by more information about added mass. It is something I have never seen or my attention was somewhere else...
Christian
Steve,
Don't you have cavity emission by the hole like what goes out the rear vent?
Christian
Well, the exciters are right next to each other, and the microphone was 1 meter away, so I think it's fair to say it was equally in front of both exciters.
My experience with two exciters is that sometimes I can find two spots where the two exciters tend to complement each other, and the FR is flatter with two than with one. Other times, I have difficulty finding a spot for the second exciter that makes any improvement. Certainly there are locations for the second exciter that make the FR worse, too.
I made these speakers some time ago, but as best as I can recall, in this particular case there was a slightly flatter response with both than with one.
For me, the most important thing is to understand that you are unlikley to find that the best spot for one (or two) exciters by following someone's "rule of thumb". You have to take the time to test different positions (or pairs of positions) to find what actually does work best, for that particular panel, in that particular mounting, with those particular exciters.
Eric
Christian,
I've always assumed they actually have some kind of elastic support in the center 60% or so of all four edges, in addition to those 4 points of firmer support. If you look closely at the animation in Steve's link, you will see some rectangles underneath the central portion of the two edges that are visible. It's hard to know how hard or soft they are, but the fact that there are significant deflections along the edges in thos regions suggests the support is pretty soft, like foam would be.
As Steve pointed out, it kind of looks like the panel "bottoms out" along at least one edge, but I'm pretty sure that's just an effect of the animation, in which the deflections are grossly exaggerated to help show how the panel deforms.
Eric
Wow, I think I'm gonna have to read through this whole thing, and I have skimmed through some, but I have a question regarding mounting...
I see these hung vertically a lot, and I'd like to try building these for an atmos home theater. But that brings up the question, if they are best mounted hanging by string/wire, how would you implement that on the ceiling?
My thought was, why hang it at all? Is it ok to mount the exciter to the wall firmly, maybe even on something very heavy? I notice some models have screw holes; What are they for if the panel attaches to the VC?
I see these hung vertically a lot, and I'd like to try building these for an atmos home theater. But that brings up the question, if they are best mounted hanging by string/wire, how would you implement that on the ceiling?
My thought was, why hang it at all? Is it ok to mount the exciter to the wall firmly, maybe even on something very heavy? I notice some models have screw holes; What are they for if the panel attaches to the VC?
Veleric.
I think we are all aware of what happens if two tweeters are placed on the same baffle , you will get comb filtering between the two .
with the tweeters you will get the comb filtering in the air off axis ,which is bad enough.
but with dml the comb filtering is on and in the panel itself .
it is easy to see , by watching the the waves moving up and down the frequency response as you move a microphone between and around the two exciters .
There is no best place , only maybe a less worst place to place the the two exciters.
They may work together at low frequencies but as the frequencies get higher they will start to fight each other.
And then you will get to the problems of when it is transmitted into the air and the off axis responses ?
I bought 40 exciter and wanted to make a podium clone, but when I found out what two exciters did to each other , I decided to stick with just the one
I also found that one exciter sounded far better than two, the one exciter sounded more coherent ,to my ears ,of course.
You can put more watts into 2 or 4 exciters ,but with my light and efficient panels and 10watt exciter power was never a problem.
Steve.
I think we are all aware of what happens if two tweeters are placed on the same baffle , you will get comb filtering between the two .
with the tweeters you will get the comb filtering in the air off axis ,which is bad enough.
but with dml the comb filtering is on and in the panel itself .
it is easy to see , by watching the the waves moving up and down the frequency response as you move a microphone between and around the two exciters .
There is no best place , only maybe a less worst place to place the the two exciters.
They may work together at low frequencies but as the frequencies get higher they will start to fight each other.
And then you will get to the problems of when it is transmitted into the air and the off axis responses ?
I bought 40 exciter and wanted to make a podium clone, but when I found out what two exciters did to each other , I decided to stick with just the one
I also found that one exciter sounded far better than two, the one exciter sounded more coherent ,to my ears ,of course.
You can put more watts into 2 or 4 exciters ,but with my light and efficient panels and 10watt exciter power was never a problem.
Steve.
Pontos.
on page 152 I made three recordings of an hdn8 exciter on a stud wall, and a few pictures.
there is no reason screwing into a ceiling stud would be a problem, except for someone sleeping up stairs !!
the volume on my tablet was very low , but a computer and headphones should do better.
this was very loud when I recorded it in my room.
so turn the volume up and enjoy.
Steve.
on page 152 I made three recordings of an hdn8 exciter on a stud wall, and a few pictures.
there is no reason screwing into a ceiling stud would be a problem, except for someone sleeping up stairs !!
the volume on my tablet was very low , but a computer and headphones should do better.
this was very loud when I recorded it in my room.
so turn the volume up and enjoy.
Steve.
I have some XPS panels hung from my ceiling.... some acrylics will be replacing them soon
XPS works great though the acrylics will have lower extension. So we will see what happens with rattling (current hanging method uses interlocked hooks)
But yea Pontos it should be no problem to hang them. If your ceiling is high enough you may even be able to play with hanging height to optimize response. I have to mount close to the ceiling but I would love to hang like 1m off.
XPS works great though the acrylics will have lower extension. So we will see what happens with rattling (current hanging method uses interlocked hooks)
But yea Pontos it should be no problem to hang them. If your ceiling is high enough you may even be able to play with hanging height to optimize response. I have to mount close to the ceiling but I would love to hang like 1m off.
Cheapvega and pontos.
Actually it never occurred to me to put the hdn8 on the ceiling and turn the ceiling in to a dml speaker .
But if you have a void with no sleeping quarters or neighbours, it makes sense.
I don't think I made measurements of the room response but as I said in the recording ,the response was from below 40hz to about 18k and the wall actually went loud, which I was not expecting.
And attaching to the actual stud sounded the best ,to my ears !
This would have a very small footprint and probably not need for a sub ?
Not sure how it would sound on the other side of the ceiling in a roof void (invisible) and how to stop any buzzing from loose ceilings .
My walls are not perfect but I didn't notice any buzzing
Holding the exciter up against the stud wall worked far better than I thought it would and had a pretty decent sound.
I could try and take measurements if anyone is interested ?
I'm sure there are better exciters or shakers out there that would do a better job .
I must admit I am a little tempted to put the two hdn8 drivers on the ceiling in my room ,just for fun 😁
I was going to throw them in the bin ,but maybe they could have a use ?
Steve.
Actually it never occurred to me to put the hdn8 on the ceiling and turn the ceiling in to a dml speaker .
But if you have a void with no sleeping quarters or neighbours, it makes sense.
I don't think I made measurements of the room response but as I said in the recording ,the response was from below 40hz to about 18k and the wall actually went loud, which I was not expecting.
And attaching to the actual stud sounded the best ,to my ears !
This would have a very small footprint and probably not need for a sub ?
Not sure how it would sound on the other side of the ceiling in a roof void (invisible) and how to stop any buzzing from loose ceilings .
My walls are not perfect but I didn't notice any buzzing
Holding the exciter up against the stud wall worked far better than I thought it would and had a pretty decent sound.
I could try and take measurements if anyone is interested ?
I'm sure there are better exciters or shakers out there that would do a better job .
I must admit I am a little tempted to put the two hdn8 drivers on the ceiling in my room ,just for fun 😁
I was going to throw them in the bin ,but maybe they could have a use ?
Steve.
Ok curiosity got the better of me.
I screwed the hdn8 onto my ceiling , I now have quite a big hole where the exciter screwed into the plaster but was not long enough to go into the stud wood !!
I managed to get these measurements before it fell off 😕 leaving the hole😠
The first photo is with the microphone about 30cm away from the exciter , I waved the microphone around about 30cm to get an average response.
The second photo was with me walking around the room with the microphone in front of me at about chest hight.
There was nothing below 40hz everything in the plot below 40hz is wind noise as I was moving around.
Interestingly the third photo shows the 600hz noise I noticed ,while the exciter was lying on the side !
This is a massive peak, but I found that if I held it tightly between my fingers ,the peak disappears.
This exciter is obviously ment to be screwed into something solid ,which is why I was getting such bad results with lighter material.
On page 152 you can see pictures of the exciter with a shed load of blu-tack on the coil area , this is still there , trying unsuccessfully to stop this noise.
also you will see the pictures of the two halves ,as it fell apart because of very little glue holding it together
it has a very flimsy casing and the screw is too short ..
I did try holding it onto the door which lowered the frequency to 20hz but the frequency response was appalling.
Steve.
I screwed the hdn8 onto my ceiling , I now have quite a big hole where the exciter screwed into the plaster but was not long enough to go into the stud wood !!
I managed to get these measurements before it fell off 😕 leaving the hole😠
The first photo is with the microphone about 30cm away from the exciter , I waved the microphone around about 30cm to get an average response.
The second photo was with me walking around the room with the microphone in front of me at about chest hight.
There was nothing below 40hz everything in the plot below 40hz is wind noise as I was moving around.
Interestingly the third photo shows the 600hz noise I noticed ,while the exciter was lying on the side !
This is a massive peak, but I found that if I held it tightly between my fingers ,the peak disappears.
This exciter is obviously ment to be screwed into something solid ,which is why I was getting such bad results with lighter material.
On page 152 you can see pictures of the exciter with a shed load of blu-tack on the coil area , this is still there , trying unsuccessfully to stop this noise.
also you will see the pictures of the two halves ,as it fell apart because of very little glue holding it together
it has a very flimsy casing and the screw is too short ..
I did try holding it onto the door which lowered the frequency to 20hz but the frequency response was appalling.
Steve.
Attachments
That's really cool Spedge, and I'm really surprised that works so well, wow. But that's actually the complete opposite of my thinking. I'm building an isolated room to keep sound IN.
My concern with hanging from the ceiling is that the string forms 4 solid attachment points, whereas hanging on a wall it does not (although don't they still dampen the panel somewhat?). Even if that works, the response between matching panels on the wall would not be the same. Honestly, I don't like the idea of hanging vertically by string either because, with the heavy magnet free to float, doesn't its inertia dampen the response?
Has anyone hung a DML panel by ONLY the back of the exciter and no wires or strings? That seems like a much better way to mount it to me.
I'm thinking of attaching the back magnet of a large DML panel to a bare stud (probably about 6-12" away) with some form of dampening attachment to purposely NOT vibrate the wall, on the inside of a double thick drywalled wall with the drywall on the outside hopefully blocking most of the sound.
I'd like to build a HT room that doubles as an anechoic chamber for speaker testing and instead of treating a rectangular echo chamber with dampening/diffusion, I'd be treating an anechoic room with purely reflective treatment (maybe the DML panels themselves?) that can be easily removed for testing. So the inside of the room will be round with AT fabric defining the walls with rockwool and foam traps behind it. Reading earlier in the thread that dampening behind the panel really opens up the performance of the panel is perfect for my application. I'm basically building an isolated room in my basement with the ceiling supported by the walls and double thick drywall on the outside and plenty of dampening inside to keep the sound in.
A pic speaks volumes (pun intended):
My concern with hanging from the ceiling is that the string forms 4 solid attachment points, whereas hanging on a wall it does not (although don't they still dampen the panel somewhat?). Even if that works, the response between matching panels on the wall would not be the same. Honestly, I don't like the idea of hanging vertically by string either because, with the heavy magnet free to float, doesn't its inertia dampen the response?
Has anyone hung a DML panel by ONLY the back of the exciter and no wires or strings? That seems like a much better way to mount it to me.
I'm thinking of attaching the back magnet of a large DML panel to a bare stud (probably about 6-12" away) with some form of dampening attachment to purposely NOT vibrate the wall, on the inside of a double thick drywalled wall with the drywall on the outside hopefully blocking most of the sound.
I'd like to build a HT room that doubles as an anechoic chamber for speaker testing and instead of treating a rectangular echo chamber with dampening/diffusion, I'd be treating an anechoic room with purely reflective treatment (maybe the DML panels themselves?) that can be easily removed for testing. So the inside of the room will be round with AT fabric defining the walls with rockwool and foam traps behind it. Reading earlier in the thread that dampening behind the panel really opens up the performance of the panel is perfect for my application. I'm basically building an isolated room in my basement with the ceiling supported by the walls and double thick drywall on the outside and plenty of dampening inside to keep the sound in.
A pic speaks volumes (pun intended):
Eric,
I think we see the same things : the blue colored panel in #4505 is extracted from a Tectonic video and around are my notes. The 4 supports keep locally the panel in its resting plan. The rest of the edges seems moving in the axis perpendicular to the plan and rotates. The rectangle in the center are visible in the drawing. What is their role : spring or damper. Probably both. I assume more damper to limit the wave reflections (no proof of that)
Christian
Pontos,
In your post 4510 you have mentionned an atmos home theater. Have you though about the compatibility of the Atmos HT with "omnidirectional" speakers like DML. I just had a look what is Atmos : it is based on the reflexion of upward directed loudspeakers on the ceiling. I assume the DML would be at the reflection point? Have you already made some tests that show a possible feasibility?
Atmos CAN be upward directed speakers, but it can also be speakers installed in the ceiling. It all depends on your settings in the receiver. It's intended to be in the ceiling, but they make the upward firing thing for people who don't want to deal with that.
Here's the Dolby recommendations for various speaker setups. I'm going for the 7.1.4 overhead setup.
And no, I haven't actually done anything with it yet. I'm just getting started learning what I can. But I want to start experimenting with a wide variety of speakers of which the DML sound the most promising and easy to me. I've designed and built my own speakers, crossovers, and subs before but only BRs.
Last edited:
Thank you Pontos.
I think it is in this paper 21st-NXT/Azima where the DML properties are listed that it is mentioned they are good candidates for multi channel system.
I see just before a reply from Cheapvega who is working on ceiling speakers. That's good. If I remember the last exchanges, we (or at least I) was wondering about the effect of wall (=ceiling) close to the back of the panel... not clear for me. The only sure thing is there is an effect on the response.
I started with DML about 1 year ago coming like you from more standard design. The way is not easy, there is no promise of results to match our expectations but it is worth a try.
Christian