Hello Christian.
Once again, the heavier the panel, the more of a pounding the exciter brace will get.
Out pf ballance exciters will also contribute noise from side wobble ,a good reason to mount the exciter body to a brace.
Not forgetting the lead out wires.
I have heard mechanical noise from a low frequency driver when pushed to its limits.
Steve.
Once again, the heavier the panel, the more of a pounding the exciter brace will get.
Out pf ballance exciters will also contribute noise from side wobble ,a good reason to mount the exciter body to a brace.
Not forgetting the lead out wires.
I have heard mechanical noise from a low frequency driver when pushed to its limits.
Steve.
Hello XRK,
The TC9FD seems to be an appreciated full range. Do you know threads of application of this speaker showing some directivity plots?
Christian
I finished building one panel using a piece of 24x22x1 (inch) XPS. I currently don't have any measurement capability (working on that) so at present all I have are my subjective listening impressions. And: I'm disappointed. The bass is weak, and vocals exhibit some pretty apparent (to me) wide variations in level vs. frequency. I thought the bass would need a sub to help out, so no great surprise there. But the mids are not to my liking at all.
I haven't given up because I (belatedly) discovered a lot of negative posts regarding XPS. Locally I can get Coroplast and various thicknesses of polycarbonate twin-wall to play with, but getting thinner ~3mm Baltic plywood is a different matter. I've found a couple of domestic sources but shipping is an issue -- for cost and arriving-in-good-shape considerations.
Here are a couple of photos showing my current efforts --
That's the back of the panel. I went with a frame that constrains all 4 sides. The panel was attached to the frame using butyl double-sticky tape, based on some info I found in this thread.
The laptop in the background now has a working (linux) version of REW installed on it. No calibrated microphone yet...
That's a photo showing the front of the panel. I used Liquitex blue acrylic spray paint to improve the WAF and also to (hopefully) improve some of the panel's acoustic characteristics. Too bad that didn't work out, I like the color. So does the wife! The acrylic spray paint tends to come out kind of "splatty" but, since the acrylic dries more slowly, it levels out OK.
To the left you can see my somewhat-disassembled SAE MK 3 power amplifier. It decided to misbehave during my testing so now I get to troubleshoot it. Fortunately, I have the service manual because I replaced all the electrolytic capacitors not too long ago. And being a double-E engineer with 40 years in the field of IC failure analysis doesn't hurt 🙂.
I probably will get to first use REW to check out my amp. What FUN I'm going to have! By today's standards the amp's distortion characteristics aren't stellar so I can use my external USB sound card (a Behringer UCA202) to test and adjust the amp's idle current so it meets its SAE specifications.
I haven't given up because I (belatedly) discovered a lot of negative posts regarding XPS. Locally I can get Coroplast and various thicknesses of polycarbonate twin-wall to play with, but getting thinner ~3mm Baltic plywood is a different matter. I've found a couple of domestic sources but shipping is an issue -- for cost and arriving-in-good-shape considerations.
Here are a couple of photos showing my current efforts --
That's the back of the panel. I went with a frame that constrains all 4 sides. The panel was attached to the frame using butyl double-sticky tape, based on some info I found in this thread.
The laptop in the background now has a working (linux) version of REW installed on it. No calibrated microphone yet...
That's a photo showing the front of the panel. I used Liquitex blue acrylic spray paint to improve the WAF and also to (hopefully) improve some of the panel's acoustic characteristics. Too bad that didn't work out, I like the color. So does the wife! The acrylic spray paint tends to come out kind of "splatty" but, since the acrylic dries more slowly, it levels out OK.
To the left you can see my somewhat-disassembled SAE MK 3 power amplifier. It decided to misbehave during my testing so now I get to troubleshoot it. Fortunately, I have the service manual because I replaced all the electrolytic capacitors not too long ago. And being a double-E engineer with 40 years in the field of IC failure analysis doesn't hurt 🙂.
I probably will get to first use REW to check out my amp. What FUN I'm going to have! By today's standards the amp's distortion characteristics aren't stellar so I can use my external USB sound card (a Behringer UCA202) to test and adjust the amp's idle current so it meets its SAE specifications.
It’s a very popular driver for DIY. You will find a lot of projects on DIYA that use it.
Here are some measurements. Directivity is really mostly a function of diameter of the driver.
https://hificompass.com/en/speakers/measurements/peerless/peerless-tc9fd18-08
Mark,
One thing you can almost certainly find pretty easily (and I think is worth trying) is a 5-ply 1/4" plywood at Lowe's or Home Depot. One version is Revolution Ply (or Revply). And you may find others in the area where they sell 2'x4' plywood sheets. I like the "Birch" I can sometimes find there. I say birch in quotes because I suspect the only thing that's actually birch is the very thin surface veneers. You may have to pick through a lot of panels to find nice, flat panel (or two), however. I like the sound quality of such plywoods much better than XPS (or EPS), or coroplast, but be prepared for the sensitivity of plywood to be much, much lower than XPS.
I wouldn't work too hard or spend too much $ on Baltic birch, as it's on the heavy side, which means even lower sensitivity.
Eric
Hello Mark
Thank you for sharing with pictures what you built.
More than 3 yares ago now, XPS 20mm was also my first material when I started with DML and I was also disappointed by the sound (even if it is from it I started to appreciate how a DML fills the room). It was PVA coated with an white acrylic in front for a better looking.
I tested a 9mm Depron which is also an XPS, PVA coated. It didn't work.
Last week because I had those panels laying in my workshop and I was looking for large panels for tests, I made directivity measurements on them. I should add that soon in the paper about directivity.
To come to the reason of my reply, I agree with Eric (@Veleric ) about plywood. After XPS, I made panels with 3mm poplar plywood from a local DIY store. The results are much better even without EQ. They have been working for more than 2 years without modifications.
I won't go to baltic wood because of the weight (and probably the price and the difficulty to source it)
Seeing your pictures, I wonder if the panel is not too constraint on the frame. Maybe Eric or others could comment. I would have made it also more rectangular (same width but increase the height using the space to the fllor) but I don't know your constraints. Here I have to deal with the idea to put some woofer below the panel and not having something too high.
Maybe you could also use REW as generator for example in 1/3 of octave pink noise to check there is no parasitic noise from the structure, the connection, the tube. I use a floor lamp for testing. It is prone to resonate (it is taller than yours). As the DML are quite heavy, many vibrations go to the structure.
I hope you will find more satisfactions with other materials.
One advice from Steve (@spedge ) is not to go too fast in the assembly... what I should said I haven't really followed for now (Hello Steve!), or let say not fully in his way but I know he is right. Glue an exciter on the almost raw material, suspend it with tape to listen (and/or to measure). Then decide and go to the next steps. Maybe several sheets of the material will be needed.
Happy to see somebody using Linux!
Christian
Christian.
About a month ago, before I dismantled everything in my music room, my cousin came for a surprise visit .
He came to listen to my panels, he did not realise I was still into audio.
I had sent him some of my videos I had sent you, just for a laugh 😃
I just played the odd set of panels that were hanging up at the time.
Can't remember which, I think one might have been the proplex panel 🤔
He really liked my panels and when comparing the sound to my headphones he said he preferred the sound of the panels 😎
I have only demonstrated my panels to a few audio friends, so it was nice to get a fresh listener's impressions.
Steve.
About a month ago, before I dismantled everything in my music room, my cousin came for a surprise visit .
He came to listen to my panels, he did not realise I was still into audio.
I had sent him some of my videos I had sent you, just for a laugh 😃
I just played the odd set of panels that were hanging up at the time.
Can't remember which, I think one might have been the proplex panel 🤔
He really liked my panels and when comparing the sound to my headphones he said he preferred the sound of the panels 😎
I have only demonstrated my panels to a few audio friends, so it was nice to get a fresh listener's impressions.
Steve.
Freedom666.
XPS is already heavily damped.
From experience.
Adding kitchen foil and silicone will only worsen the problem.
I have used 5mm XPS which sounds OK when coated in a thin coat of epoxy resin.
This gives the XPS the high frequencies that it needs for full high frequency response.
I have in the past, posted music and frequency response plots for this panel type, with and without epoxy coating.
Steve.
XPS is already heavily damped.
From experience.
Adding kitchen foil and silicone will only worsen the problem.
I have used 5mm XPS which sounds OK when coated in a thin coat of epoxy resin.
This gives the XPS the high frequencies that it needs for full high frequency response.
I have in the past, posted music and frequency response plots for this panel type, with and without epoxy coating.
Steve.
Dave,
Can you please expound on this issue of lobes a bit more? Consider the following, and for the sake of simplicity, ignoring the impact that the exciter itself has on the mode shapes:
Would, for example, the (2,2) mode, having 4 antinodes, have 4 lobes that each radiate into the 4 quadrants of the hemispherical half space in front of the panel? And directly along the vertical and horizontal axes, would those lobes no longer be apparent due to cancellation by the neighboring antiphase nodes?
And further would for example the (1,3) mode radiate with 3 lobes that would all be observed along one axis, but appear as a monopole along the other axis?
That also brings up the question of how you calculate the directivity plot. You may have explained this in the videos, but I don't recall, sorry. In the directivity plot, does the SPL at say 45 degrees represent the SPL at a single point in the horizontal plane but at 45 deg to the left (or right)? Or is it the average across all angles in the vertical plane, rather than just 0 deg in the vertical? Or something else?
Thanks,
Eric
Yes, all of this is correct!
Right now I'm doing only the horizontal plane at 1m, from -45deg to 45 deg. There's no vertical plane in there. Could always be added in, of course.
Additionally, I've been measuring some free plates - the PETTaLS simulation seems to work well for small free plates either supported by the exciter, whose magnet is attached to something, or hung by strings (this would be a free exciter magnet) - see picture below. This was for a small (17cm x 17cm) gatorboard panel.
When I move to larger panels, it's pretty clear that the strings are imposing a new kind of boundary condition, and I don't want to risk the exciter by having it be the only thing supporting the panel, so I was only measuring with strings. The impedance comes out looking halfway between a totally free panel and a panel with 2 rigid supports in its upper corners, but these supports only affect the low-frequency modes. I'm going to have to figure out a way to model "2 strings" as a boundary condition eventually.
As much as I like working on these advanced topics, I'm worried that I'm getting too far afield from the original intention of why I was developing PETTaLS (it was originally just for my grad students to make graphs and pictures easily!). I'd really like this to be something that somebody who is completely new to this process can use to help them understand what modes are, why the positioning of the exciter is important, why different exciters produce different results, why you want to not use a super heavy plate, etc. So... I need to find the right balance of features that will be useful to the experts (e.g. everyone in this forum) and to the newbies. If I have too many options about coincidence frequency, modal overlap, etc. it might be too confusing to new people. Just thinking out loud!
Dave .
Can't you have two versions, one for simple people like me and one for the likes of Christian 😃
Steve.
Can't you have two versions, one for simple people like me and one for the likes of Christian 😃
Steve.
More probably "Eric, Christian and others..."... I have even not reactivated my preferred one which is to have the directivity plot of an open back DML! But I do it!
Christian
I was thinking about this exciter.
Although it is not very good for my panels.
I does have its uses.
Thick heavy materials can be driven very well up to 10k.
It can drive stud walls, which is very similar to my rigid ply panel but with thicker materials.
This could be mounted on a solid wall for those who do not have a stud wall, with a couple of inches of sound proofing materials within its frame.
I may give this a try at some point and see how it sounds.
I am not expecting anything exceptional but it could be good for surround sound or some small pa work.
Ordinary exciters probably would not do this job as well as this type of exciter.
Steve.
Hello dave, again.
Regarding your two string mounting points.
I use my fingers to find positions on my panels that can flatten or fill in dips in the lower frequency response below about 300hz or so.
This is easy on my very small 6x9inch panels but not so easy on the larger panels.
I usually place weights in these positions.
It would be handy if these positions could be predicted in your software for larger panels.
These positions can also be used as mounting points for the panel.
This sorts out two problems in one go.
Is this possible.
Steve.
Regarding your two string mounting points.
I use my fingers to find positions on my panels that can flatten or fill in dips in the lower frequency response below about 300hz or so.
This is easy on my very small 6x9inch panels but not so easy on the larger panels.
I usually place weights in these positions.
It would be handy if these positions could be predicted in your software for larger panels.
These positions can also be used as mounting points for the panel.
This sorts out two problems in one go.
Is this possible.
Steve.
Thank you all for your comments/questions. Regarding the XPS I used, I was unable to find anything thinner than 1". I had my suspicions that it wouldn't work out all that well, one of the reasons I just built one panel to start with.
I used the butyl-rubber on all 4 edges based on some measurements I saw in this thread......but now I can't find that info. This thread needs an index 😀.
The Lowe's near me has 5mm plywood underlayment. It's called RevolutionPly. They also have 2x4 birch plywood. It's a bit thicker, about 5.6mm. IIRC the underlayment has gotten good reviews for use as DML panels.
I want to examine a product called "hardboard tempered panel" that Home Depot stocks. A little more than 3mm in thickness, it looks a lot like masonite and is claimed to be made from eucalyptus fiber. It may be too dense though because it's a compressed & glued material.
I used the butyl-rubber on all 4 edges based on some measurements I saw in this thread......but now I can't find that info. This thread needs an index 😀.
The Lowe's near me has 5mm plywood underlayment. It's called RevolutionPly. They also have 2x4 birch plywood. It's a bit thicker, about 5.6mm. IIRC the underlayment has gotten good reviews for use as DML panels.
I want to examine a product called "hardboard tempered panel" that Home Depot stocks. A little more than 3mm in thickness, it looks a lot like masonite and is claimed to be made from eucalyptus fiber. It may be too dense though because it's a compressed & glued material.
Mark51.
My local wickes store used to have 3mm MDF which worked very well as a dml panel.
I made two 6ft x 2ft panels ,one MDF and the other 3.4 mm ply.
They sounded similar but the MDF had a more damped sound with less coloration than the ply.
The rigidly of the panel is not a problem as long as you make it large enough.
I would show a picture but this site is having problems with my phone again.
Steve.
My local wickes store used to have 3mm MDF which worked very well as a dml panel.
I made two 6ft x 2ft panels ,one MDF and the other 3.4 mm ply.
They sounded similar but the MDF had a more damped sound with less coloration than the ply.
The rigidly of the panel is not a problem as long as you make it large enough.
I would show a picture but this site is having problems with my phone again.
Steve.
Both of these are worth trying. The speakers I have been listening to for the last 5 years or so as my main speakers are made from Revolution ply. Below is a pic of one of them, before I attached the stands to hold them up. I had more pics in a previous post showing the construction details, but they were linked to photobucket and have been lost.
And I think possibly my favorite speaker that I have made so far is made from Lowes (or HD?) birch plywood. Although I think the panels that they refer to as birch plywood can actually vary considerably from time to time and store to store, so it may be a bit of a crapshoot. If you find some you like get a few panels because you may never be able to find exactly the same thing again!
Eric