Thanks DMLBES. I have plenty of PVA glue available.
I was indeed thinking of 50-100cm (sorry, I always tend to think metric these days!) and attaching as you suggest. I need to decide whether to recess the panel in the frame a little. I've gone for the Dayton Audio 25mm framed job. The build seems sturdier that some of their others and there are some useful mounting holes for the spine support (more like ribs really as they will be in a horizontal orientation).
I was indeed thinking of 50-100cm (sorry, I always tend to think metric these days!) and attaching as you suggest. I need to decide whether to recess the panel in the frame a little. I've gone for the Dayton Audio 25mm framed job. The build seems sturdier that some of their others and there are some useful mounting holes for the spine support (more like ribs really as they will be in a horizontal orientation).
I have the Dayton Audio DAEX25FHE-4 exciter with the four 'ears' for possible mounting / securing purposes. I also have ultra high density EPS board for the panel. Is there any reason why I should not use some M3 bolts & nuts to secure the exciter to the panel (in addition to the glue fixing) as opposed to using some kind of spine support attached to the frame?
If I read the exciter positioning thinking correctly, the exciter will be best positioned in an area where there are more resonant modes (NXT.......rubbish??....THINK AGAIN!). This, therefore, begs the question as to whether the exciter and panel are going to give their best either when the exciter is firmly secured just to the panel, or to both the panel and the frame via a spinal support.
If I read the exciter positioning thinking correctly, the exciter will be best positioned in an area where there are more resonant modes (NXT.......rubbish??....THINK AGAIN!). This, therefore, begs the question as to whether the exciter and panel are going to give their best either when the exciter is firmly secured just to the panel, or to both the panel and the frame via a spinal support.
Others have tried mountimng the exciter directly to the panel and have generally been disappointed. As well, most all commercial DML's have mounted or supported the exciter to a frame/spline/spine/bracket/what-have-you. Some also prefer , the excitor not mounted at all except via the glue ring on the front. Again it all depends on so many factors....weight of exciter, type of spider it uses, how the panel is mounted, etc... Also, that particular exciter can be mounted to spine vis the hole in the center of the magnet assembly on back...I believe its an M3 bolt hole as well. the only issue there is that hole may also provide some magnet cooling...as I said, lots of choices to experiment with...no right answers, just compromises based on your design... and lots of listening and measuring...
Thanks Geosand, that's very helpful. The existing holes on this exciter should permit easy mounting to a spine type arrangement. I guess one train of thought is if the exciter is appropriately secured to the frame, then it is more likely to excite the required resonant points sufficiently. I'm going for a flexible approach to panel mounting in a frame to allow for experiments.
Indeed, definitely something to be aware of. I have various things I can experiment with. I hope to prepare the panels tomorrow and build the frames on Monday. Then it's a case of lots of listening and measuring. REW and the miniDSP software is likely to see a lot of use over the next few days. The joys of retirement...
Two reasons not to use M3 bolts and nuts to secure the exciter to panel. 1.Anything other then the moving voice coil touching the panel material will hinder (result in less) vibrations, the tighter you make the bolts the less vibrations.
The diaphragm of a DML panel is no different then the diaphragm of a conventional cone driver you wouldn't screw nuts and bolts to a conventional drivers cone would you?
2. Anything other then the voice coil touching the panel will result in a change of sound. If you put metal bolts on a DML panel it will resonate the nuts and bolts. The same way if one uses metal washers for weights it will resonate the metal washer and change the sound.
A frame and a spine is needed for a (good sounding) DML panel the same way a conventional cone driver needs a (frame) basket to hold everything together. A exciter is just in its raw state with just a magnet and voice coil. If you add a metal basket on a exciter with surround it becomes similar to a BMR driver. I don't understand how people cant see it as its a basic foundation for any driver as it prevents voice coil sag and gives the exciter something to push off of.
The frame with surround holds and supports the panel materials weight while the spline supports and holds the magnets weight.
What would you say are the main areas that make a good exciter?
This is a couple of renders of the design I have started. It has a screw on back cap which you could also use to fit a stand, allowing the panel to stay in the air and not touching any surfaces. I currently have the contact point being a solid circular disc after reading through some of this thread.
If there is anything you would like to see, let me know. FD61 is my top priority right now, and with some luck I will be for a long time, but when I am winding down for the night I like to continue designing, and so this is what I have started.
Paul
This is a couple of renders of the design I have started. It has a screw on back cap which you could also use to fit a stand, allowing the panel to stay in the air and not touching any surfaces. I currently have the contact point being a solid circular disc after reading through some of this thread.
If there is anything you would like to see, let me know. FD61 is my top priority right now, and with some luck I will be for a long time, but when I am winding down for the night I like to continue designing, and so this is what I have started.
Paul
Ahh retirement...being poor with nothing to do ;-)
I'm working on my own little experiment right now. (No, it won't be an audiophile grade speaker, its just an experimental platform to test from before I decide to build the real thing). Have ordered the same exciters as you...should be here in a week or so...now I just need to get a cheap calibrated Mic to use with my old Win 10 Tablet that I've just updated and am installing REW and ARTA on...so I can take measurements (or at least learn how too, lol.) At any rate, once I'm done I'll post my project here with all the details and pics and such and for XRK (AKA Mr Foam) I'll even give it a name and some backstory...
I cant tell from the pics but is the contact point at least 2-3mm higher then the surrounded edges? If not at higher excursions the diaphragm material could hit the outer edges depending on how much excursion your exciter moves.
There are many things my wife thinks I "should" be doing though!
But yes, lack of money until my state pension kicks in in just under 2 years does indeed mean the DIY route for a lot of my HiFi. Unfortunately, my other hobby is photography!
I have a cheap Behringer measuring mic plus their generic profile which seems OK enough. I can also use my AQ Dragonfly Red to help with getting a more reliable reading, thus bypassing the laptop sound thingy.
Thanks for the additional insights DMBLES. It does make sense. I have seen what I think are are your analogies between the various DML elements and traditional cone drivers and it is all very logical.
Only one extra hobby!?
Maybe that's why my wife is always saying something about all my "toys" in the house!
Thanks for the message DMLBES
The actual part that makes contact ends up about 4mm out from the frame to avoid contact and instead of it being a ring that attaches to the surface it is a solid circle to give a larger contact patch to the panel. I could possibly make this contact part a screw on piece so the contact patch could be changed but not sure how much difference it would make and if it would be worth implementing.
Power of the motor is 1.7Tm @ 4 Ohm. This is a bit on the lower end I know but makes good use of some magnets I have laying about. Compliance I will not know till I properly go and test it but we can probably have anything from around 0.01mm/N to 0.30mm/N with the current design but changing material and thickness of flexing components.
Paul
Is there a particular source/type of EPS you like? I've tried 1 inch Foamular 250 XPS. It is the most efficient panel I've tried, but doesn't sound as "accurate" to me as Sure-ply plywood.
How much difference does coating with PVA really make? I have not tried that yet, but it would be easy to do. Do both XPS and EPS benefit equally from a PVA coating?
Thanks,
Eric
There seems to be a wide variety of cores and skins and thicknesses available. What construction did you use?
Eric
Highest possible compression/density EPS. 1inch formula XPS is too thick, 5-10mm (1/4-1/2) is ideal thickness. Plywood is more accurate because its stiffer but less efficient because its heavier. EPS/XPS are one of the most efficient of materials because pound for pound its one of the lightest materials. Like I said before there are compromises to each material.
Using PVA glue to make the panel stiffer has a slight difference. Also its not just one technique that will make a really big difference but a combination of multiple techniques used together in ones design that will make a significant difference in sound.
EPS and XPS are different as they are not equal to begin with so using PVA glue will have slightly different benefits. EPS is beads fused together while XPS is one whole unit. XPS is actually clearer with more detail but EPS has better warmth and better bass most likely due to EPS being a tad bit more flexable then XPS. If you look at EPS you can see the beads stuck together while XPS has no beads.
If you like the sound of wood then by all means use it as it is all personal preference but at the expense of losing efficiency.
EPS/XPS has a more diffused sound then wood. Wood has a less diffuse sound because its more accurate due to it being stiffer/harder.
Honeycomb core composite with skin IMO combines the qualities of EPS/XPS and wood making it efficient as well as accurate but at the cost of being very expensive and harder to implicate as you need to know how to do certain steps to perfection. I for one don't have the funds to venture into that costly territory.
Great summary from DMLBES. I tested Carbonfibre, Beech Ply, and Foam-board. Carbonfibre was the most neutral and accurate, Beech ply slightly more coloured but close to Carbonfibre, Foam-board was louder but least neutral. I have yet to test EXP foam.
I cross over to subwoofers at 100hz so the efficiency issue with ply is less significant. It’s all a compromise and fun to experiment with. I dream of being able to afford huge sheets of Carbonfibre ...
I cross over to subwoofers at 100hz so the efficiency issue with ply is less significant. It’s all a compromise and fun to experiment with. I dream of being able to afford huge sheets of Carbonfibre ...