Hello,
I have a very limited experience : my first tests in DML were with XPS. The one I had was pink and as the looking was also part of the expectation, I tested some painting. To go step by step, I started with PVA coating then those acrylic white painting for ceiling with some water added. Not really happy of the result : it was not a clean white, the increase in weight was here and the process not really reliable so I decided then to test plywood. Let see the other feedback but currently I haven't reach a good balance color/added mass with XPS... which doesn't mean it is not possible!
Christian
PS : for the exciter, it was an area PVA coated and sanded without painting.
I'd stick with the 50/50 PVA
Paint has fillers in it which you don't need and which will add weight and deaden the sound
It's probably time for reflection and maybe a change of emphasis. The topic of this thread is a study of dmls as a full range speaker, but after almost 500 (!) pages it's pretty clear that most accept the limitations of bass performance in a dml panel and the need or at least desirability of using a pistonic bass unit to supplement the dml panel. Maybe we should direct more attention to deriving the best method of achieving this as a design aim.But have we stifled our inspiration?
I believe there's merit in pursuing a multi driver panel form of bass unit, coupled with a mid/high range dml panel.
Thoughts??
Eucy
Andre.
The response down to 80hz is no problem, but is a response up to 20k a problem?
The obvious choice would be a xps panel if a restricted hf is needed?
The problem is I do not have a guitar to test the sound.
He has a fishman guitar amp which does not have an aux out, but he does do his mixing at home on his home stereo.
So maybe I could make speaker panels for his home mixing?
Or am I wasting my time and his ?
Steve.
The response down to 80hz is no problem, but is a response up to 20k a problem?
The obvious choice would be a xps panel if a restricted hf is needed?
The problem is I do not have a guitar to test the sound.
He has a fishman guitar amp which does not have an aux out, but he does do his mixing at home on his home stereo.
So maybe I could make speaker panels for his home mixing?
Or am I wasting my time and his ?
Steve.
Christian, Those resistor values are very high. I think you might be picking up noise from there. I would try keeping resistor values to below 1k, preferably below 100r if you can, and drive them from the headphone output for a better drive voltages (the headphone output should be able to drive 32ohms.) The higher your measuring system impedances are, the more susceptible they are to interference.
Having said that, I think I've just realised why I was getting weird readings below 10hz on my system!! 🤦♂️ I suspect it's because the headphone output must be high-pass filtered, and my resistor values might be too low! (I have not investigated yet.)
If you do test lower value resistors on your test rig, and then get the same low-frequency wobbles on the impedance curves as I did, then we will know for sure what the problem is.
It looks like there has to be a trade-off between noise compatibility and low frequency capability.
Yes, you don't want much response below, say 150hz nor above 5khz. If you did gave him a guitar speaker with too much signal at bass or treble it would not sound good at all.
For a guitar you would want a band-pass filter cutting off at 6db/oct below 200hz, and the same above, say 4khz. Gentle. No more than 6db/octave. You would play with damping to get this right... various thicknesses of double-sided tape between the driver and the panel to reduce the HF, and play with panel size to get the bass end right.
I don't know about XPS, I have not found a piece to test yet. But you do need something that will respond well up to a gentle peak at 3khz, and then drop off slowly. Check the Celestion curve somewhere above. That's ideal for guitar.
If you want to make monitor speakers for mix-down, then it's imperative to get the response as flat as possible with as great a bandwidth as possible. Any peaks or dips in the monitors will be badly exposed when he plays back his mix on different systems, in his car or at church or at a gig. (For example, weak bass in his monitors will result in too much bass in his mix on another system; Or too much treble in his monitors will result in weak treble in his mix on another system; etc etc. His monitor sound will be inverted on another system)
Instrument speakers can be tailored and scooped and peaked, but studio monitors must preferably be as ruler-flat as possible. Hi-Fi can be somewhere between the two.
Last edited:
Eucy,
So you mean a DML subwoofer to go with a DML "full range", right? That is, an all-DML WAW (Woofer Assisted Wideband)?
Do you think a DML subwoofer really would offer much that a pistonic sub wouldn't?
Andre doesn't seem to think so.
To be clear, I'm on the fence about it myself. I do like the idea of a DML sub just from the challenge of doing it. I did play around with some ideas myself, but I wasn't really happy with anything and dropped it, at least for the time being.
You mentioned a "multi-driver" bass unit. Did you mean multiple exciters on the same "subwoofer panel"? Why do you assume it would need multiple exciters instead of just one?
I'm not disagreeing with your suggestion, I'm just curious to understand your thinking a little better.
Eric
Hello Eucy,
I think that implicitly most of us are in this idea of a "wide range" helped by some mean in the bass more than a true "full range". At least it is my goal. It is the title of the thread which is not fully in phase with that probably?
Christian
André,
Is there a misunderstanding because of the coma in the screen shot? The set up of my laptop being "French" the decimal separator is a coma instead of a dote but strangely, the language is English. Confusing? Below a schematic of my set up. The serial resistor is 10Ohms. I have voltage dividers to protect the line input. The power amp internal supply is a classical transformer, the 5V power supply for the soundcard is a switching one.
The today investigations shown the RCA to jack cord connecting the line in is probably badly shielded. I replaced it by a better one but it is not the main cause. More investigations needed.
The SoundCard having an headphone output, I could simplify the circuit... if this output is accessible to REW under linux! Or use an extra USB supplied DAC that I use generally for FR measurements as source with an USB mic.
Eric,
I'm thinking more of a stiff pane driven principally in pistonic mode using multiple small drivers like the BMR engines, several examples of which have been shown here in previous posts.
Perhaps there may be a possibility of some controlled bending wave action as well but there's no escaping the fact that area+excursion = bass, so by using a panel of say 300x600 and 4 motors with an excursion of +/_ 5mm (approx BMR limits), it should be possible to get a decent result. I'm positing the drivers being arranged in a stretched diamond pattern purely because if bending does occur, this pattern will best match mode 1,1 in a rectangular panel.
Panel shape- maybe just simple truncated corners ala Audiofrenzy (as shown in one of his rare non-carpet shots -LOL)
Rear wave cancellation issues should be able to be controlled with a shallow enclosure. Above this I envisage a mid/high range panel of say 300x300 separated by an isolation strip.
For starters I'd avoid crossovers, and run the bass panel full range to see what happens.
Opinions/suggestions welcomed
Eucy
Last edited:
Hi Christian -
Well, the intention of the study was clear, the realization is also clear, and falls short of the intention. I accept the limitation, and want to find the most elegant, compatible way to achieve satisfactory bass without resorting to a traditional sub.
That's my intention, it remains to be seen if it can be realised
There's an old ditty from my childhood which has stood me in good stead over the years:
"Somebody said that it couldn't be done
but he with a chuckle replied
that maybe it couldn't, but he would be one
who wouldn't say so till he'd tried
so he buckled right in with a trace of a grin
on his face, if he worried he hid it
and he started to sing as he tackled the thing
that couldn't be done, and he did it!"
(PS - it's probably caused me a lot of grief as well
)Eucy
Eucy,
Why assume that the 300x600 bass panel couldn't handle the high frequency as well? Perhaps you are assuming that the BMR drivers wouldn't go high enough, is that it? If so, why not add another HF, small excursion driver to the same panel, which may or may not have a high pass filter?
I'm not sure this would work, but, for argument's sake, why assume that it wouldn't?
Eric
To build any type of subwoofer you will need a low pass filter. The most convenient way to do it is with a sub plate amp.
https://www.parts-express.com/speaker-components/subwoofer-plate-amplifiers
https://www.parts-express.com/speaker-components/subwoofer-plate-amplifiers
Eric,
Maybe - what type of HF driver did you envisage?
I was more thinking that by separating them, each panel may be designed to favour the desired frequency range- a fast light panel for the mid/highs, and a bass panel which may be heavier/stiffer (or selectively so) as needed
Eucy
Subwoofer-a loudspeaker component designed to reproduce very low bass frequencies.
In any driver especially full range drivers the mid to high frequencies spl output will usually be much higher then the lower frequencies therefore the higher frequencies will drown out the lower frequencies if not filtered out.
Adding multiple drivers will increase lower frequency output but without a filter it will also increase the already high output of the higher frequencies.
If you look at this BMR chart again right around 100hz its starts to drop like a rock like most exciters.
https://www.parts-express.com/pedocs/specs/297-2156--tectonic-tebm65c20f-8-spec-sheet.pdf
In any driver especially full range drivers the mid to high frequencies spl output will usually be much higher then the lower frequencies therefore the higher frequencies will drown out the lower frequencies if not filtered out.
Adding multiple drivers will increase lower frequency output but without a filter it will also increase the already high output of the higher frequencies.
If you look at this BMR chart again right around 100hz its starts to drop like a rock like most exciters.
https://www.parts-express.com/pedocs/specs/297-2156--tectonic-tebm65c20f-8-spec-sheet.pdf
Yep - no dispute there but surely it's also due to the small BMR diaphragm size, and by using 4 or so motors on a large diaphragm, it will boost those low frequencies to an acceptable level
Time will tell
Sure a larger diaphragm will boost those low frequencies to a acceptable level but what do you think will happen to the (already high) higher frequencies if not filtered out?
You said it yourself in post 9,495 make 2 separate panels in there favored desired frequency range. One for bass and one for mids and highs. The best way to do this for a sub/bass panel is with a low pass crossover aka sub plate amp. This is the reason I dont understand why you feel that it would work better without any crossover?
You said it yourself in post 9,495 make 2 separate panels in there favored desired frequency range. One for bass and one for mids and highs. The best way to do this for a sub/bass panel is with a low pass crossover aka sub plate amp. This is the reason I dont understand why you feel that it would work better without any crossover?
Af...I just want to try it - KISS principle - maybe the extra mass of the larger diaphragm will self damp/filter the higher frequencies to an acceptable level- same basic moving mass principle which limits the HF performance of a traditional woofer.