I have far less experience than Steve with DMLs, but I have played with them for about a year and a half now. I've also done a little bit with small full rangers, having built a pair of quarter wave TL type units. Experience with small full range drivers taught me that at the end of the day, they just cannot move enough air for the bottom end. I like orchestral music and they just didn't deliver. However, they are working well for me as rear speakers in a 5.1 system.
Moving on to DMLs, I came to them after several instructive but failed attempts at OB speakers - mainly due to inappropriate drivers. I knew I wanted a combination of OB bass and DML mid/top. I tried 25mm EPS, but didn't like it. I fancied trying plywood but knew that birch ply was perhaps too heavy. I hit on 3mm poplar ply as having half the mass of birch ply.
A 396 x 565mm poplar ply panel has worked well for me, although it is lacking at the extreme top end frequencies. My panels are attached to a frame via self-adhesive neoprene 'straps'. They run from around 300hz upwards. The bottom end is served by Monacor 15 inch PA drivers in a winged OB setup. In my smallish listening room, bass is strong down to 40hz.
Crossover and EQ duties fall to a miniDSP HD. Measurements are taken with a calibrated miniDSP UMIK-1.
Having had a mishap with one of the exciters, I am now taking the opportunity to evaluate EPS and compare it with my ply panels. As you will see from my various posts, I am having issues with higher frequencies. I think this may be due to the double sided tape I'm using. I'll know later today whether I'm right. What I have determined, however, is that EPS seems more capable of going all the way to 20khz than my ply panels. The image below shows an EQed plot from a 350 x 396mm panel, taken with the mic about 30cm from the panel.
Attachments
Your welcome. I found it through the citations of one of those other articles. Have a nice day!
I see my first exciters made it to my mail box yesterday. I picked up some ASX05408-HD-R from Digikey.
Steve,
My daughter rags me about my messy room all the time. After looking at yours, I don't feel so bad! Brothers from another mother, I guess!
My daughter rags me about my messy room all the time. After looking at yours, I don't feel so bad! Brothers from another mother, I guess!
Lordtarquin.
This is a quick picture of the vh grade at 2inches, I've also zoomed in a bit to show more detail.
The plot shows plus and minus 2.5db over the range of 125hz to 8k, it would have carried on up to 20k if I had not cut a large hole in the middle of the exciter coil area ,which is what is causing the drop.
It does come back up at the 20k mark though, which shows the exciter is outputting up to this frequency.
Hopefully you should be able to do this without large amounts of EQ ?
Steve.
This is a quick picture of the vh grade at 2inches, I've also zoomed in a bit to show more detail.
The plot shows plus and minus 2.5db over the range of 125hz to 8k, it would have carried on up to 20k if I had not cut a large hole in the middle of the exciter coil area ,which is what is causing the drop.
It does come back up at the 20k mark though, which shows the exciter is outputting up to this frequency.
Hopefully you should be able to do this without large amounts of EQ ?
Steve.
Attachments
Steve
Love the room. Mine seems to be a paragon of tidiness in comparison...😉
I have now done a plot of the smaller EPS 70 panel with the glued exciter in place. It's not a precise match, but it closely follows the DA frequency chart in terms of where the peaks and troughs are.
In fact my implementation of the DA DAEX25FHE*4 is a little bit better than their published FR chart with a slightly less deep upper treble trough and starting at 7K rather than DA's published 5k.
Whilst it's fixable with EQ, it does suggest that there's a fundamental weakness in the exciter in the upper treble region. Perhaps my mishap with the blown exciter has been a blessing in disguise! With EQ, it reaches 20khz easily on the EPS panel - which it cannot on the poplar ply one.
Time to try to source some new exciters to add to the Monacors that hopefully will arrive next week. Some more high grade EPS is also needed.
Love the room. Mine seems to be a paragon of tidiness in comparison...😉
I have now done a plot of the smaller EPS 70 panel with the glued exciter in place. It's not a precise match, but it closely follows the DA frequency chart in terms of where the peaks and troughs are.
In fact my implementation of the DA DAEX25FHE*4 is a little bit better than their published FR chart with a slightly less deep upper treble trough and starting at 7K rather than DA's published 5k.
Whilst it's fixable with EQ, it does suggest that there's a fundamental weakness in the exciter in the upper treble region. Perhaps my mishap with the blown exciter has been a blessing in disguise! With EQ, it reaches 20khz easily on the EPS panel - which it cannot on the poplar ply one.
Time to try to source some new exciters to add to the Monacors that hopefully will arrive next week. Some more high grade EPS is also needed.
what is TLS.. also your panels are having holes in the middle and not fully flat surface.. what is the purpose of it, so far seen most of the DML to be flat surfaces
Can we use the panels full range without any cross over and subwoofer support or does the DMLs need low end support of subwoofer.
Last edited:
Lordtarquin.
DA use foam core for their tests, so I would expect a hf roll off.
I thought the 25mm unit would have a better response?
Although the hf is still there but at a lower level ?
Have you measured the coils just in case of damage?
Steve.
DA use foam core for their tests, so I would expect a hf roll off.
I thought the 25mm unit would have a better response?
Although the hf is still there but at a lower level ?
Have you measured the coils just in case of damage?
Steve.
Jaxboy.
You can see why I was chucked out of the living room.
She who must be obeyed could not stand all the wires anymore 😀 .
Steve.
You can see why I was chucked out of the living room.
She who must be obeyed could not stand all the wires anymore 😀 .
Steve.
sarathssca
TLS - transmission line speaker. Output from the bass driver goes through a specially constructed interior pathway and vents at the bottom. The phase of the output port then matches what's coming from the front.
My personal experience is in using DMLs with bass support. By high-passing the DML exciter, much of the bass energy is removed, thus permitting more output for the desired range (from 320hz in my case) without bass frequencies potentially overdriving the exciter voice coil.
Bass frequencies require the shifting of a lot of air, so a bending wave panel (a DML) needs either to be big, or use particular materials. Take a look at Spedge's canvas panel experiments.
TLS - transmission line speaker. Output from the bass driver goes through a specially constructed interior pathway and vents at the bottom. The phase of the output port then matches what's coming from the front.
My personal experience is in using DMLs with bass support. By high-passing the DML exciter, much of the bass energy is removed, thus permitting more output for the desired range (from 320hz in my case) without bass frequencies potentially overdriving the exciter voice coil.
Bass frequencies require the shifting of a lot of air, so a bending wave panel (a DML) needs either to be big, or use particular materials. Take a look at Spedge's canvas panel experiments.
Steve
I've given up trying to understand these exciters! They have been like it from day 1.
sarathssca
I forgot to add that yes, the panels can be used full range. I only really alluded to this in my reply. For domestically 'sensitive' situations, I suspect a smaller panel plus bass unit may have a higher PAF (Partner Acceptance Factor). Indeed, so-called FAST setups with bass driver(s) plus full range driver for mid/tweeter duties is how I started down this road. But once I'd replaced the mid-tweeter with a DML, there's no going back.
I've given up trying to understand these exciters! They have been like it from day 1.
sarathssca
I forgot to add that yes, the panels can be used full range. I only really alluded to this in my reply. For domestically 'sensitive' situations, I suspect a smaller panel plus bass unit may have a higher PAF (Partner Acceptance Factor). Indeed, so-called FAST setups with bass driver(s) plus full range driver for mid/tweeter duties is how I started down this road. But once I'd replaced the mid-tweeter with a DML, there's no going back.
Hello Yura,
Very nice sounding panels. Could you please share more speakers build information?
What panel materials you used?
What are dimensions of panels?
What kind of exiter do you used?
/DMLTux
I use a combination of DML, TLS and a sub. The Nautaloss I is a very compact (11.5x7x9") 5 foot fibonacci spiral transmission line. Mine uses a 6.5" mid-woofer that plays 20 Hz pretty strongly. The DML is 12x16"canvas that is choked at 40Hz, but due to way it is mounted, only goes down strongly to about 80Hz. The combination gives me a sound that picks up the nuances in the mid-range, plus plenty of thunder. I'm very pleased.
Sarath.
Thanks Lordtarquin for answering the tls question ,I didn't spot that.
I also usually use the TLS up to about 300hz ,they sound really good in this area.
But I usually try to find how low the panels can safely go without distress ,which varies from panel to panel type.
I then, while music is playing ,move the cut off point on the panels up and down in the say 90hz to about the 300hz region ,to find the best possible sound.
I do like to run the panels as full range as possible for more of that dml sound.
This works for me and helps sort out the room suck outs,a win win situation 😀
As for the holes on my panels, these are test panels that have been experimented on ,over many years, the vh grade eps has been sawn up so it has no parallel sides ,painted with vinyl silk to minimise self noise ,a cone shape has been sanded into the surface to allow more intimate sounds to come through the eps, the hole was to try and minimise the exciter cavity problem and reduce the the distortions within the coil area,(it would seem I am the only one who has a problem with this ? ),and a lot more tests I can't remember, it's basically at the end of its life now,but good for comparisons,it still sounds good 😀
Building a panel can be as easy or as hard as you like ,come to that ,as cheep or as expensive as you like, some of my best panels were free 😀
I have paid for fancy materials,but they usually disappoint🙁
Steve.
Thanks Lordtarquin for answering the tls question ,I didn't spot that.
I also usually use the TLS up to about 300hz ,they sound really good in this area.
But I usually try to find how low the panels can safely go without distress ,which varies from panel to panel type.
I then, while music is playing ,move the cut off point on the panels up and down in the say 90hz to about the 300hz region ,to find the best possible sound.
I do like to run the panels as full range as possible for more of that dml sound.
This works for me and helps sort out the room suck outs,a win win situation 😀
As for the holes on my panels, these are test panels that have been experimented on ,over many years, the vh grade eps has been sawn up so it has no parallel sides ,painted with vinyl silk to minimise self noise ,a cone shape has been sanded into the surface to allow more intimate sounds to come through the eps, the hole was to try and minimise the exciter cavity problem and reduce the the distortions within the coil area,(it would seem I am the only one who has a problem with this ? ),and a lot more tests I can't remember, it's basically at the end of its life now,but good for comparisons,it still sounds good 😀
Building a panel can be as easy or as hard as you like ,come to that ,as cheep or as expensive as you like, some of my best panels were free 😀
I have paid for fancy materials,but they usually disappoint🙁
Steve.
Cheapvega.
The obvious answer is the art canvas panel, I think I paid about £8 for a pack of 4 30cmx43cm panels , bargain.
10 2mm 4x6inch panels cost about £5 .
They are easy to put together for a good sound.
This panel could also be improved on.
Although this panel goes down to 40hz I personally would not use it down to this frequency or any other panel for that matter .
But if it works for you and others that's fine.
I run my tls up to 300hz so my cheapest option would be the small card panel .
Rolling them off at about 120hz depending on size and stresses?
These are free as I have used cornflake boxes or pizza boxes , the thickness is usually in the half mm region.
I glue them together to make 1mm card ply panels, pva ,card ,pva,card,pva.
You could also just us a 1 or 1.5 mm card of course if you have them lying around.
I try to keep the panel as light as possible, similar to a paper cone speaker.
Plenty of detail with low colouration and it can go loud as long as you don't run it too low.
These are the two main types of panels I will be using with my eps panels, the veneer panal is very similar to the card panel in sound and looks, this could also be an option, but obviously I'd have to spend some money.
Steve.
The obvious answer is the art canvas panel, I think I paid about £8 for a pack of 4 30cmx43cm panels , bargain.
10 2mm 4x6inch panels cost about £5 .
They are easy to put together for a good sound.
This panel could also be improved on.
Although this panel goes down to 40hz I personally would not use it down to this frequency or any other panel for that matter .
But if it works for you and others that's fine.
I run my tls up to 300hz so my cheapest option would be the small card panel .
Rolling them off at about 120hz depending on size and stresses?
These are free as I have used cornflake boxes or pizza boxes , the thickness is usually in the half mm region.
I glue them together to make 1mm card ply panels, pva ,card ,pva,card,pva.
You could also just us a 1 or 1.5 mm card of course if you have them lying around.
I try to keep the panel as light as possible, similar to a paper cone speaker.
Plenty of detail with low colouration and it can go loud as long as you don't run it too low.
These are the two main types of panels I will be using with my eps panels, the veneer panal is very similar to the card panel in sound and looks, this could also be an option, but obviously I'd have to spend some money.
Steve.
I've just gone mad and ordered some epoxy resin to see how a thin coating will sound applied to different surfaces?
Will this improve on the pva or will it turn out to be too rigid ?
Hopefully I can get a thin enough coating.
We shall see .
I have not planned this ,it was an impulse buy,but one I have been thinking of for some time😕
For years now,I have also been thinking about using one of the more powerful piezo units to mount on the front of an eps panel,the only problem is I will need a piezo amp plus power supply.
The idea is to get away from using coils for driving the panel , and also the piezo can be placed on the front surface ,so that the primary surface would be facing the listener, not at the back of the panel.
I've had my finger on the buy button for some years now,maybe my next birthday or
Steve.
Will this improve on the pva or will it turn out to be too rigid ?
Hopefully I can get a thin enough coating.
We shall see .
I have not planned this ,it was an impulse buy,but one I have been thinking of for some time😕
For years now,I have also been thinking about using one of the more powerful piezo units to mount on the front of an eps panel,the only problem is I will need a piezo amp plus power supply.
The idea is to get away from using coils for driving the panel , and also the piezo can be placed on the front surface ,so that the primary surface would be facing the listener, not at the back of the panel.
I've had my finger on the buy button for some years now,maybe my next birthday or
Steve.
Luckily I returned to the forum after a long absence to find this post.
I got a lot from this paper, particularly because I've also been thinking about travelling vs standing waves in an effort to flatten out extreme peaks in the spectrum. Love the graphic where its very clear that as damping increases, the wave pattern goes from standing to travelling wave.
My own experiments looking at the resonance peaks in a DML have taught me that the biggest resonance peaks do come from first reflections coming back toward the exciter, as you would expect - from corners and edges. Rounding the corners, and putting cusps on the edges opposite the exciter help to eliminate these resonances and increase the distance travelled before a standing wave can occur.
The paper also shows another reason why a vertically oriented narrow panel (ie tall panel) is probably best. It shows that as you move from 1x1 mode up to 1x7 mode, the dispersion characteristic gets worse along the axis with the large wavenumber. Thus for the tall DML not only is base accentuated because the edge modes are maximised (due to that shape having maximal edge length), but also because most of the uneven angular dispersion then occurs in the vertical direction rather than the horizontal, and has less effect on the listening window.
I believe a lot more can be done with the shape of the whole panels, or edge shaping, to optimise performance. I'm currently playing with elliptical shape to attempt to maximise distance travelled before the second reflection, but I think log spiral and parabolic also could prove worthwhile.
Speaking of shape, another very interesting phenomenon I've come across recently which I've not seen on the forums (not saying much 'cos I haven't read all the forum posts) is that of the 'acoustic black hole' (ABH). An ABH is an accurately machined parabolic taper applied to an edge, or cut from the middle of a panel tapered down as thin as possible, and usually with a viscoelastic damping material applied. This is a method to prevent reflection which could profitably be employed in DML speaker design.
At the moment I am half way through reading this article by B Zenker linked from frieda' post.
This answers Lordtarquin's problems with his and my exciters.
This is not all the problems associated with coil exciters but so far only deals with coil resonances?
I was so excited when reading this I had to post straight away 😀
This is all part of what I have been saying for years!
He might get on to other problems further on but I think he is just concentrating on coil resonance?
Very interesting.
Steve.
(PDF) Upper Frequency Limit of Flat Panel Loudspeakers - Evaluation of the Voice Coil Break-Up Modes
This answers Lordtarquin's problems with his and my exciters.
This is not all the problems associated with coil exciters but so far only deals with coil resonances?
I was so excited when reading this I had to post straight away 😀
This is all part of what I have been saying for years!
He might get on to other problems further on but I think he is just concentrating on coil resonance?
Very interesting.
Steve.
(PDF) Upper Frequency Limit of Flat Panel Loudspeakers - Evaluation of the Voice Coil Break-Up Modes