I was doing the tests from my desktop, which has a bit noisy fan. Also, I noticed I have some interference coming from the amp when maxing the amplitude. Will try to clean it up for future tests.
I like the idea of trying to come up with some simple common set of tests to evaluate panel performance. I think it is tricky though since we are using varying equipment that usually is not calibrated. But will be helpful even if we cannot really reference them against each other directly. Instead of dBSPL I guess using amplifier output voltage as reference could make sense?
Been listening a bit to my sanded and treated Neopor with 2x 50w, and it also at lower levels I do prefer the sound. Not sure yet how much is the fact that I'm using two exciters or the PVA treatment, but it is clearly sounding the best so far of any of my experiments.
Ordered 8x DAEX25FHE-4...I do think that adding the Zobel network to the 50w exciters might bring the HF up to acceptable levels, and will give that a go, but nice to have a reference and I should be able to drive up to 8 exciters with my amp. Not sure if it is a good idea with neither that much power or that many exciters on one plate, but like to have the room for experimentation.
Ordered 8x DAEX25FHE-4...I do think that adding the Zobel network to the 50w exciters might bring the HF up to acceptable levels, and will give that a go, but nice to have a reference and I should be able to drive up to 8 exciters with my amp. Not sure if it is a good idea with neither that much power or that many exciters on one plate, but like to have the room for experimentation.
Sarathssca,
The cd sounds like a very good alternative, as it's light, has a hole in the middle like Steve likes and are plentiful. I just threw away several dozen, as I have all their songs on my computer plus a back-up hard drive. By the way, ALWAYS have a back-up storage system for your music! When I got divorced (best thing she ever did for me!), all my songs were on the computer, which she kept. Instantly, no more library of songs I had spent years accumulating! Or pictures, too!
The cd sounds like a very good alternative, as it's light, has a hole in the middle like Steve likes and are plentiful. I just threw away several dozen, as I have all their songs on my computer plus a back-up hard drive. By the way, ALWAYS have a back-up storage system for your music! When I got divorced (best thing she ever did for me!), all my songs were on the computer, which she kept. Instantly, no more library of songs I had spent years accumulating! Or pictures, too!
Sarathssca,
In my previous post when I mentioned poplar, birch, it is more in the idea of density than quality. Poplar plywood is lighter than birch one. My source of plywood is the DIY store next to my house... As they propose poplar one, as it is the lighter they have, I have used it. I have not detected problems that could be linked to the plywood quality. I think birch come from posts of other members. It exists also thinner than 3mm in aeromodel shops (more expensive).
But I think almost all the plywood should work. Some posts spoke about "underlayment plywood" (sorry if the wording is a bit inexact...).
For the canvas, I used real balsa where Spedge had in fact a light plywood (basswood? sold for balsa) and I think Jackboy used more standard plywood as in the video on Youtube. Refer to my history file where there are some details (not so many...).
I understand it is not easy to make choice in all of that... We don't have all the same constraints of space, the same taste. The results might be different also because we don't have the same material sourcing.
Clearly, my current plywood panel is probably not optimized (45x120cm) but it has my preference... knowing my canvas pair due to the use of balsa is a bit aside the original one. Other builder might have a different opinion..
Christian
As we don't have the same exciters or materials, the difference of efficiency of those elements will remain. In fact I am wondering if the smartphones are not the best tool. There are many SPL meter apps, their mic are quite good and the chain of gain seems under control.
Does somebody have experience about the smartphone as SPL meter? Difference from one model to an other one?
+ @sarathssca
As I am in a phase of testing..; by curiosity... and if it inspires somebody...here is a quick test with a CD.
The CD was double taped on its whole surface on the panel (I wonder if it is what was on you tube, we could imagine it is just glued near the center with a little spacing like a whizzer cone)
- blue is the panel as it is (XPS9mm, 20x30mm). As in other posts I use it because it is available and easy to modify
- green is with the CD on the front face
- brown is the CD alone (no XPS)
This is not a canvas (I don't want make damage for now to my canvas pair).
For now I won't adopt or reject the idea... Not as damaging as I would have thought... To keep in the vast drawer of ideas?
Christian
Apps have to be calibrated to be accurate at all. They don't know anything about things like the sensitivity of the mic on your device. Even if the app would have measurments of every available model and used that data, phones will vary individually.
And the mics are usually quite low spec, incapable of handing high SPL, so often useless to gain information, especially at levels where distortion becomes an issue.
Hi Christian,
Thanks for testing with the CD, but it seems it was better without it in the graphs above..
I think in that youtube video, that person just added some small disc/washer on the panel surface where exciter was there behind, and attached the CD to the washer, so that CD surface was not touching the panel surface. But not 100% clear how it was attached to the panel, as not more detailed info given in that video of how it was attached to the panel. And the purpose he was saying that the CD was for treble region like a tweeter.
@hkguy6
I would like to come back o the attachment method you described hkguy6 in the post #4750
Here is a quick test where I added a metallic washer between the exciter voice coil and the panel. As usual those days, it is a XPS 9mm 20x30cm not really special, only for tests purpose by comparison.
The washer is 27mm diameter with a 15mm hole, 2.7mm thick so about 8.4g (evaluation, no balance...) which is a lot compare to the voice coil mass 1.6g!
Looking at the spectrogram, the picture is not so good. Some frequencies remain a long time.
The technique seems promising. Here the idea was to push a bit far in weight to see if there is FR limitation... and this washer was just under my eyes close to the exciter diameter. The screw is also missing in this test.
So who will propose an easy to make light enough solution ?
Christian
I would like to come back o the attachment method you described hkguy6 in the post #4750
Here is a quick test where I added a metallic washer between the exciter voice coil and the panel. As usual those days, it is a XPS 9mm 20x30cm not really special, only for tests purpose by comparison.
The washer is 27mm diameter with a 15mm hole, 2.7mm thick so about 8.4g (evaluation, no balance...) which is a lot compare to the voice coil mass 1.6g!
- blue is the XPS panel as reference
- green is with the washer added
Looking at the spectrogram, the picture is not so good. Some frequencies remain a long time.
The technique seems promising. Here the idea was to push a bit far in weight to see if there is FR limitation... and this washer was just under my eyes close to the exciter diameter. The screw is also missing in this test.
So who will propose an easy to make light enough solution ?
Christian
Thank you Steve.
Commenting the length of the video and the energy cost of the bytes as probably nothing to do in my comment...
Anyway, at least 2 interesting points :
- the 4 exciters arrangement increase the SPL up to a certain frequency. The good point is that the effect goes quite high in frequency despite the distance between the exciters. 2dB remains @20k.
- the FR with 2 exciters seems not different from 1 exciter which goes in the way shared here.
By the way, the behavior of this acrylic is unbelievable with its rising FR (+6 to 8dB @20k)
Hi Leob, I'm very interested in your results testing the zobel. This builder of full range speakers is very enthusiastic about the benefits, contrary to what many others say. Please see this link: Zobel Network and the Full Range Driver
Everyone,
this forum is really active and I will have a hard time to reply to each of you.
So, I'll try to distribute my answers/ comments in several posts.
Thank you, Eric.
Well, to be honest, I follow this forum for quite some time now, with pauses, and I decided to trust the findings of its members and go with common grounds and additional information that one can find in research papers. In the end the plywood sheets I used are 11$ CDN, therefore it should be not too much pain to redo a panel if something goes completely wrong.
It is not really complex to build. Just do some reading about the different guitar bracing types and pick one. It took me an afternoon to cut and glue the strips for my panels, then let them dry over night and then take a concave hand planer or Dremel and give the strips their basic shape.
My experience in respect to multiple exciters is similar. However, I only had my framed panels laying flat on a table when I dabbled with positioning of three exciters.
What I found is that one can highly change the sound characteristics through the distance amongst the exciters and I also found that the closer they were to each other the better the sound. However, there was one combination that I was not able to reproduce where the sound had a lot of air and soft mids which I really liked. Here, the exciters where also only roughly 1 inch apart on a 3.2x4 ft panel, but I settles with a super tight placement in the end.
Sebastian
this forum is really active and I will have a hard time to reply to each of you.
So, I'll try to distribute my answers/ comments in several posts.
Thank you, Eric.
Well, to be honest, I follow this forum for quite some time now, with pauses, and I decided to trust the findings of its members and go with common grounds and additional information that one can find in research papers. In the end the plywood sheets I used are 11$ CDN, therefore it should be not too much pain to redo a panel if something goes completely wrong.
It is not really complex to build. Just do some reading about the different guitar bracing types and pick one. It took me an afternoon to cut and glue the strips for my panels, then let them dry over night and then take a concave hand planer or Dremel and give the strips their basic shape.
My experience in respect to multiple exciters is similar. However, I only had my framed panels laying flat on a table when I dabbled with positioning of three exciters.
What I found is that one can highly change the sound characteristics through the distance amongst the exciters and I also found that the closer they were to each other the better the sound. However, there was one combination that I was not able to reproduce where the sound had a lot of air and soft mids which I really liked. Here, the exciters where also only roughly 1 inch apart on a 3.2x4 ft panel, but I settles with a super tight placement in the end.
Sebastian
Hello Christian,
awesome! Thank you for that test!
I think your findings make kind of sense.
I tried to find a tool so that it is a little easier for me to rethink my considerations in respect to the most suitable multi-exciter wiring.
When you dabble with different series / parallel / series-parallel / parallel-series setups and enter your amps power and minimum ohm spec you can see how different wiring may affect to wattage exciter with different ohmage will see. I am especially referring to your green / red graph.
https://soundcertified.com/speaker-ohms-calculator/
In respect to the green/blue graph I once found the following paper which used the Feonic FXPRO Audio Drive for testing.
https://eejournal.ktu.lt/index.php/elt/article/download/11232/5962
They state in their conclusion that " The application of two actuators for the excitation of panel increases the emission of sound energy into the forepart hemisphere."
This might explain that especially the mid-high frequency range is a little higher with 2 exciters despite the same wattage applied.
Lastly, it turns out that polyurethane was a bad idea. I had it here and I thought I save some time and do not get something like epoxy. I had no idea that it would be so soft. Even after a week of drying it was still elastic and that is the opposite of what my intention was. So, just today, I removed it again. Luckily, I was able to peel it of like dead skin. It will take at least until the weekend until I will be able to fill the cavities again.
Sebastian
No dream or magic in electricity! The Zobel circuit won't recreate what is happening with the voice coil inductance. It might be a help for the amplifier if this one has difficulty driving the loudspeaker in high frequency. In my tweak of cheap class D amp, I add such a circuit at the output of the amplifier to be sure in any case in HF (far above 20k) it is loaded. So medicine for the amp yes, medicine for the loudspeaker in our case no.
Hello Christian,
since I started, I somehow primarily focused on the panel design and I did not spend any time on exciter wiring, impedance, wattage,... considerations. Last weekend, I finally sat down and did some reading on that topic, because I still did not know what to do with my 3 exciters at hand.
In short ,... I was completely wrong, or better underestimated the complexity.
My previous speakers were always single fullrange driver speakers and therefore I never had to bother too much about that part of our hobby.
After some reading I realize that my tube amp's output wattage is too low and that I will never find an exciter wiring solution that would allow me to combine the lower sensitivity 40Watts exciters with the high sensitivity 20 watt exciter.
So, I decided to order a second set of DAEX25SHF-4 so that I would run 2 of them in series on each panel. My tubeamp provides the highest output at 8Ohm and there are not too many exciters available for that. So series of two 4 Ohm exciters it is.
That leaves me with the other exciters, which I can now use to do some testing with the bracing without fear to stress/ruin the 3M too much.
The idea behind the bracing and what I found out so far.
When you look at this graph then you will notice that the most flat spl is actually represented by an acoustic guitar. Hence it is a very neutral medium for sound reproduction, it seems.
What I also learned from looking at this graph is that I'll use 80ies rock, as well as percussion with chimes to test my panels and not the mostly seen female focal.
And then there is the following graph.
I think a guitar (no overtones considered) plays from 80Hz to ~1.5-2khz?
So most of the muddy-/honky-/cuppy-/hollowness should be "tunable" by bracing, because that is what guitar builders do via bracing.
That also explains to me why some people see value in adding metal disks or cups to their dml systems to raise the top end of the spectrum.
My experience with bracing so far is that when I got the plywood sheet from the store and tapped it in different spots, I got more or less the same tone and it was not even very loud. After I added the bracing, the panel wasn't just only much stiffer, but it created on each spot a completely different tone with a lot of attack. the panels already felt like an instrument and it was also much louder but at the same time provided more sustain, which we do not really want here. Hence my fully foam suspended frame. The suspended panels still have the areas with the different tones, but the sustain is mostly gone.
Today I asked my wife to stand in front of the panel and listen while I moved the exciter around and we both actually found serveral other spots apart from the 2/5-3/5 region that sound maybe even much better.
Because I do not want to make wild claims, I ordered a measuring mic and I will attempt to make spls for different positions with one exciter. This will take 2 weeks, but it would be great to see how this compares to the members' experiences.
So, we will see...
Sebastian
Seen a few reports claiming that it can indeed improve the HF response of a driver. And it does make sense somehow....if the amp sees a lower impedance, it should deliver more power. If the response is unaffected, doesn't the amp give stronger output at lower impedance when using a Zobel, or does the output just get wasted?
Basically, the amp is a voltage generator. The power is the result of this voltage into the load impedance. For an exciter and a Zobel network, 2 devices are feed with in parallel with the same volatge. The exciter, no change for it. The zobel network where the resistor sees more and more part of the voltage when the frequency increases due to the reduction of the capacitor impedance with frequency. At the end, you have the answer... the energy is wasted. It might be some amps happy with that because they see a kind of constant impedance. But giving back HF level, no.
I have ordered the components, so will be able to test it soon.
I'm not an electronics engineer, so likely got some things wrong, but there seems to be two common applications for Zobel network. When used on the amp output, something like a 0.1uF cap is used as an RF snubber.
Other application is as part of crossover circuit, to make the filter behave better for the woofer. There you match impedance with the driver, but you don't actually feed the HF content to the driver.
So in neither of those applications you should expect a difference in resulting FR. But when using with a full range exciter, the only report I seen with someone who tested it said it did affect the response.
Christian.
At first I did try using different handheld microphones directly to my computer , but the electrical noise was very bad.
I then remembered that I used to make recordings outside in the garden listening to pond fountain and birds singing.
I had also made recordings of my wife bell ringing at the local church which sounded very good.
I even made some recordings of an organist at st Mary Abchurch London which sounded very good.
using the phone sounded better than using separate microphones.
The frequency response was also better , with a fairly flat response from 50hz to 10k , which dropped sharply below and above this .
I just found the sound very was natural from the phone, so thought I would give it a go for recording my panels.
usually I record ,with the phone lying face up on a rolled up rug (it works).
If I remember correctly ,I should move the phone a little towards the right panel to equalise the channel balance, obviously I did not do this on my last quick recording.
I'm having to move the phone much closer to the speakers than I would like ,as all the junk in my room is affecting the sound.
The closer you get to the panels the closer you get to the LF drivers, which starts to dominate the sound, so that adjustments need to be made.
The closer you get , the panel also starts to sound exactly like a pair of headphones, I can't tell the difference.
I might try using a piece of card in the middle of the phone to see if it makes a difference, when I get things sorted out ?
Steve.