A Study of DMLs as a Full Range Speaker

Believe it or not the first thing I do when starting a new DML project is to see which side of the raw EPS panel is my preferred side as they do not sound the same to me. When doing this I use high quality vocal tracks with a lot of hi hat cymbals like Dian Krall....... When listening one side will sound shoutier/louder this is the front side, while the less shouty side is the back side.

This is one reason why most of you wont be able to follow my designs if you cant tell the difference between the two sides. Unless maybe you can measure it?

Or it could mean I am just crazy but thats what I've been doing for the past 9 years or so. :ROFLMAO:
 
Yes you can just use the exciters as intended and just stick it to any surface. Its just that over time the VC can start to sag depending on the size of the exciters and the smaller exciters wont sag as much due to having a lighter magnet.

They do a model where the part that is stuck on remains and the rest can be changed.
Now I see - at least in a variation of the model I'm interested in using - that threaded ring, implying you can adhere that to the excited surface and then when you need to change the exciter, just screw it off and screw on a new one. Again I see the implication being "just let it hang" as if that's a-ok.

I'd sure like to know what the manufacturer is / has been thinking, regarding this.
 
https://www.foambymail.com/polystyrene-foam-sheet.html

If not mistaken it goes as high as 5lbs. but not many make this high density.
I looked around a lot, and 50 kg/m3 is the highest I have found, which should be the same as 3lbs.
According to this article it can range between 12 and 50kg:
https://www.bpf.co.uk/plastipedia/polymers/expanded-and-extruded-polystyrene-eps-xps.aspx

EPS packaging in the top range of density is very rare. I got one sheet when I bought a fridge that might be similar to my 25kg/m3 plates, but that is a special application where a thin sheet with high compressive strength is needed. All other EPS packaging I have come accross seems to be 20kg/m3 or less. I find it extremly unlikely that someone would use top grade EPS for fruit.
 
  • Like
Reactions: 1 user
Now I see - at least in a variation of the model I'm interested in using - that threaded ring, implying you can adhere that to the excited surface and then when you need to change the exciter, just screw it off and screw on a new one. Again I see the implication being "just let it hang" as if that's a-ok.

I'd sure like to know what the manufacturer is / has been thinking, regarding this.
The ideal scenario is for the voice coil of the exciter to be centered in the magnetic pole gap, and there is a margin of error for its positioning.
IF the voice coil suspension is stiff enough to support the magnet, then it is not necessary for additional support to the magnet. But this applies only to small magnets. This is also why ferrite magnets are not used in exciters. They're just too heavy for a decent BL.

Neodymium magnets can produce much higher BL from a much lower mass, and this is why they are used in exciters. The limit for sizr/wattage/FR, is dictated by the inductance of the voice coil, the weight of the magnet, and the stiffness of the suspension around the coil. So until there's a better/cheaper material available than neodymium, we are stuck with nuts and bolts to support the concomitant higher wattage/heavier motors.
 
  • Like
Reactions: 1 user
Also I swore I read an article on EPS that said they go as high as 5lbs, I just cant find it anymore.

Then around 2 years ago I found some EPS square 12inchX12inch sheets 1/4inch thickness on Ebay that said 5lbs density. So I quickly snatched those up before someone else bought them as that was the only ones left and I wanted to see what 5lbs density was all about.

Compared to my 1/2inch 30g density EPS these 1/4 inch 5lbs density is around twice as dense as the 30g. In fact its so dense it reminds me of those thin balsa hobby woods they sell in craft shops. Now there is a down fall though because 5lbs eps is so dense you lose sensitivity just like with wood materials. Its also less lively but a bit more accurate sounding. For me I prefer a livelier sound with better sensitivity which is why I prefer the 30g 1/2inch EPS sheets over the 5lbs density.

Its like the Goldie locks and the three bears analogy. The material shouldnt be too soft or to hard/dense but JUST RIGHT. :LOL:

I just found this after making this post and this chart goes up to 4lbs. So I am sure I didnt imagine 5lbs. :D

https://omnexus.specialchem.com/selection-guide/expanded-polystyrene-eps-foam-insulation
 
Last edited:
  • Like
Reactions: 1 users
Ok, interesting, I have only seen XPS in those densities, but glad it doesn't seem like I have missed out :)

I also found that lower density gives better sensitivity but below 20-25kg/m3 high frequency reproduction suffers due to lack of compressive strength, so I think the sweetspot is as light as possible while still getting enough HF. For a tigher sound I found that at least from the samples I had, GPS is a bit better than regular EPS of the same density, and changing size and suspension helps as well, all without sacrificing efficiency.
 
  • Like
Reactions: 1 user
Are you sure it is 3 lbs?
+ @Audiofrenzy
I am not really fan of these units! I guess 3lbs is 3 pound per cubic feet (pcf) so 48kgm³ which a lot for EPS no? I have in mind that 25kg/m³ is a good target for DML.
Common EPS range is 12 to 30 (35?)kg/m³. See a previous post #7207

Audiofrenzy : an electronic kitchen balance is more precise than necessary to evaluate the density. Possible?

Christian
 
I am not sure it doesnt say the density but this is the company the EPS container in my photo is from.

https://www.aptcollc.com/

My other EPS says 30g.....the EPS container seems to be similar density to the 30g.
Taking an example on this site : 5F box. Internal volume is 28,6l, External volume is 40l. Weight is 525g. So density = 46kg/m³
Audiofrenzy, you are the luckier DML DIYer having found real HD EPS.
Like what some data help in the understanding.
If we push a bit with figures this gives about 15MPa as Young modulus.
In 12.5mm thick, areal mass µ = 0.570kg/m²
Too late... and too tired for now to go further with figures. We'll see next time.

Audiofrenzy, do you add some PVA on it ?

Christian
 
Recently tested EPS (fruit box) seems to be not good in the high frequency
Is there any recommended material above 2K~20K frequency?
Thank you so much
What is the density of your EPS. For now I haven't got HF from the EPS I have (less than 20kg/m³). From post #9417 from Steve and just before from Audiofrenzy, also I think from Leog (25kg/m³), HD EPS is needed.
Alternative but with a lower efficiency is poplar plywood.
 
  • Like
Reactions: 1 user
Looking up Poron 92, I see there are many densities available, what density did Eric use, or did I miss that reference?
Eucy
Eucy,
Sorry not to respond earlier. I'm just catching up! It was 15 lb/ft3 version, 0.25" thick. Poron Extra Soft Slow Rebound 4790-92-15250 to be specific. In that thickness it only comes in 12 and 15 lb as far as I know, and I doubt it makes much difference.
Eric
 
@homeswinghome
The source for the frequency demarcation for low and high frequencies.
https://www.politesi.polimi.it/bitstream/10589/89767/3/Moschini_2014.pdf;vibroacoustic
Around page 49 it discusses it.

The paper that shows a change in upper frequency that you’re asking about is
https://www.extrica.com/article/10418/pdf

Ah, I just reread what you were asking. The attenuation for high frequency on damped edge panels is just what I see in the SPL graph.
If I leave the edges free I always have a higher spl in the upper frequencies compared to when I have the edges covered in rubber.
Jamie,
Thanks for those links. I think maybe you mean around page 140 in the first paper, if I am not mistaken.
Eric
 
I have been reading through a lot of scholarly articles on material design for acoustic properties.


High frequency modes are either affected by natural resonances in the material, or coincidence resonance where different frequencies amplify or attenuate each other because of having mode overlaps.

Reading the technical literature can help a lot, even if it's not directly addressed to DML performance.

Regarding the term "coincidence resonance" it is not a term I know. I understand well what is meant by coincidence frequency, but not coincidence resonance. Can you explain?

And in another post you mentioned "coincidence frequency overlap" which is another concept I am not familiar with. Can you explain or share a reference for those?

Thanks,
Eric
 
Recently tested EPS (fruit box) seems to be not good in the high frequency
Is there any recommended material above 2K~20K frequency?
Thank you so much
If you have the time and facilities to experiment, then the 'orthogonal spring' (Jamie's idea for the name) seems to promise encouraging results. I used EPS with Kraft paper skins, and the driver glued to a polycarb disc in a hole in the middle of the panel. It delivered wider bandwidth and flatter response, and retained all of the DML advantages. But the downside is low efficiency.
You will have to do a lot of experimentation.

Another way to tackle the problem is to use very thin EPS sheet, say 3mm thick, and to skin it with a very strong material like aluminum foil. Use epoxy resin for adhesive for even higher stiffness.
This combination will give you a higher overall density than EPS by itself (I think I got 70kg/m3), and therefore lower efficiency. It seems there's no way to get high bandwidth AND high efficiency together unless you use multiple exciters.
 
Last edited:
  • Like
Reactions: 1 user