Here's a good site that has a variety of sound files for the same recoding.
2L High Resolution Music .:. free TEST BENCH
Here's another one for various style:
Free High Resolution Audio | Find HD music
Geez X!
Talk about some mad scientist looking table!
I'd get lost in the wiring! ha ha!
I wouldn't know where the power button is on that contraption!
Talk about some mad scientist looking table!
I'd get lost in the wiring! ha ha!
I wouldn't know where the power button is on that contraption!
Looking at those, and thinking about modes of vibration, made me wonder if anyone has tried something like a nautilus shell. The shape of which is interesting because it's (discretely) self-similar. A golden spiral, or any logarithmic spiral, is continuously self-similar. That is, the modes of vibration would be the same regardless of the frequency. Well, at least for an infinitely large spiral, but I guess you could make one large enough for a reasonably low frequency.
Sorry, don't want to clutter the thread, but a good recording (for evaluating drum timbre and attacks) that also sounds great on YT:
YouTube
Thanks for the link 2C! Sounds great! No worries, I was just reading a review that surprised me. So maybe clutter here - but maybe applicable if you are looking for a good sounding affordable amp. Not everyday that “peasants can dine with a prince” (prince being XA25 and J2):
AkitikA GT-102 power amplifier Page 2 | Stereophile.com
AkitikA GT-102 power amplifier Page 2 | Stereophile.com
Here's the Warnaka patent, for those who may be interested... Filed March 8 1962.
https://patentimages.storage.googleapis.com/d7/c4/c9/cbd405fcdd75ae/US3247925.pdf
Yes, it would be kinda like Gatorboard. And I'm not sure if that's good or bad, or both!!! The difference is that the gatorboard skins are a wood/resin composite that is denser (i.e. heavier) than wood but not as stiff. So the wood veneer over foamcore should be lighter and stiffer than gatorboard. I'm not sure if this is better or worse.
I have noticed also that gatorboard lacks HF. Funny things is, so does just about everything else I've tried, with the exception of plywood. The explanation escapes me.
I've tried XPS (1 inch and 1/2 inch), gatorboard (10 mm and 5 mm), end grain balsa, foamcore, various ceiling tiles, PVC foam, polycarbonate twinwall (8mm), and many plywoods. The plywoods give decent output up to about 8 or 10 khz, everything else seems to drop off around 4 khz. I would love to understand why. My best estimate for the critical frequency of 1/4 inch plywood is about 3-4 kHz, and I get virtually the same for 1 inch XPS. So I'm having trouble explaining the difference based on that.
I feel like I'm still pretty far from understanding what the optimum properties are for a panel. Sure, it needs to be light and stiff, but beyond that I'm not so sure what's ideal, or if there is such a thing. Plywood gives great range, but lacks efficiency. XPS has great efficiency but lacks range. How do you get both? One simple approach is to combine them (i.e wood veneer over foamcore). But yeah, that is very much like gatorboard so may or may not be better. I have to admit I've started wondering if simply using the high efficiency panels like PS foam combined with DSP to get the range isn't the best approach. But I'm not going there just yet.
Eric
Veleric admittedly I am grasping at straws here but the speed of sound in wood ( various types) is between 3300 M/s and 4400 M/s whereas polystyrene is circa 900 M/s. I can’t find if foaming polystyrene increases or decreases that figure but as we are dealing with surface waves I can’t see discontinuities like bubbles increasing the speed. Just a hunch that might be something to do with the high frequency attenuation of PS v Birch ply
Excellent find! As old as it is, the more current theory papers I've been reading seem to prove out what Warnaka describes and the equations are the same and the explanations given are a little easier to understand as well. Short and sweet patent. And I like how he shows the drawing of the panel being framed and pined at the corners elastically...just as you have described earlier. I can see why NXT references this patent in a lot of their other patents.
The acoustic attenuation rate of foamed PS is quite high. It is used in cavity walls for sound control so my guess is a thicker panel will have a greater difference front to back than a thinner panel, which we can test. I have found a couple reference papers for sound attenuation in wood and another for a range of materials used in sound deadening, including foamed PS. I will read them tomorrow and post links if they are useful.
Currently I think there it may be a combination of the different attenuation rates and the different speed of sound in ply V PS that are causing the high frequency transmission differences that Veleric noted. Ply has a higher speed of sound and a lower attenuation rate than PS which might answer Veleric’s question.
Last edited:
So the one paper you posted a week or so ago I've been reading over and over and it seems to stress that the critical Freq of a panel is all important to the overall FR smoothness and especially to a decent high end. I don't fully understand the theory, but it seems that if you want to improve the HF's you need to have the crit F above your highest frequency you want. Thus if you want to go to 20Khz then your critical F needs to be at least 20K or higher. The higher the better as less distortion/noises/whatnots get passed down to the lower freq's. I've done some calcs on like EPS that showed a Crit.F. of around 6.5K which theory says will not support higher freq's very well and potentially cause large peaks and dips above it. Of course you can try to take advantage of the high peaks and dips by EQ's or using separate tweeters etc...but I believe if we start with the right material to begin with with a high crit. F we will have a much better fighting chance for proper high end.
And of course you need a larger panel as well to support the additional modes that will also tend to smooth out the HF's and make them more prominent.
Then you add in the fact that the exciter may tend to have a naturally sloping downward FR at the high end and you end up with double trouble. For instance, take the cheap $8 24W 25mm high flux Dayton exciter everyone loves to use. If you look at its example FR on 1" XPS, you clearly see the very obvious downward slope on the high end. I don't think its realistic to think you can get good high end from that exciter even though it has a lot of other things going for it...I like the exciter because its cheap, handles power well and is perfect for experimenting. Beyond that, it sounds pretty good, but I wouldn't bet a really nice DML system on it.
Now there are other exciters that do have a really nice extended response on the same 1" XPS. For instance one of the Dayton $5 19mm excitors has a really nice extended response on the high end and even the low end...but it only handles 5 watts...I would think that maybe using a few of these (in parallel/series) on a panel along with a higher powered exciter, one of the 20-40 watt models may give you the high end you want as well as the mids and lows.
Thats the best I can figure out for now... so for me I'm currently looking for a panel material with a high crit F...the one paper you mentioned used a panel of 4mm corrugate polypropylene (aka, corroflute, plastic sign making material, "plastic corrugated cardboard"). They claim it had an crit F of 23Khz! and an F0 of 110hz. And they claimed it had good extended response to 20K and the manufacturer of the panel (PackardBell) listed the FR as up to 18K. So I'm looking at that material potentially as the research paper used an ~6"x8.5" panel of the stuff to get the crit F of 23K... I have not yet run through the numbers on 4mm plastic corrugate to see if the Crit F is really that high, but if it is, it may be an ideal material to experiment with because you could also get an extremely low F0 from a not too unreasonable sized panel...i.e., less than the 4'x8' sheets you can get them cheaply in... at least if my calcs are correct...
At any rate, you can see what road I'm heading down...maybe in the next month or so I'll be able to experiment some...unless someone beats me to it.
Last edited:
I played with this about 4 or 5 years ago. The best result I got was with 5mm bamboo plywood. HF was extended well beyond 10k as I recall, but it depended on choice of exciter. The one that worked best was from RS Components. I need to look for the details. I did post response measurements and details to both the Audio circle and Parts Express groups at the time. I had the ply panels cut and supplied by Ponoko.com.
I finally decided to actually make a pair of DML's. I've been playing around for months with different panel materials, sizes, exciters, frames and splines, but still didn't have a pair to just sit and listen to. I realized that I was getting into analysis paralysis, so decided it was okay just to build something, even if it was far from optimized. So I did.
The panels are Revolution Ply underlayment plywood about .200 inches thick, cut to about 16" x 23". They are mounted in a frame and supported by EPDM foam that's 12" and 9" long on the long and short sides respectively. Each has two Dayton Audio DAEX25FHE-4 24 watt exciters.
been listening to them last night and today and I'm really encouraged. They sound better than I expected for a first attempt.
I made some REW measurements today too. They need a little EQ but less than I expected. Will share the results later when I figure out how!
The panels are Revolution Ply underlayment plywood about .200 inches thick, cut to about 16" x 23". They are mounted in a frame and supported by EPDM foam that's 12" and 9" long on the long and short sides respectively. Each has two Dayton Audio DAEX25FHE-4 24 watt exciters.
been listening to them last night and today and I'm really encouraged. They sound better than I expected for a first attempt.
I made some REW measurements today too. They need a little EQ but less than I expected. Will share the results later when I figure out how!
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
My images don't seem to work, let me try again:
An externally hosted image should be here but it was not working when we last tested it.
