Back from Norway on Sunday, this week I didn't finished the speakers 
I thought it would be funny to try different materials for the baffle, so I took MDF, birch ply, different wood sheets and made some baffles. I made some listenings too and now I think to have the optimum, for me at least: a sandwich, with 4 mm poplar ply, 10 mm beech ply and again 4 mm poplar ply. I can't explain why, however. Maybe because of the combination of soft and rigid/hard woods, but I really don't know. And it may or may not work with other speakers. But it's funny
I thought it would be funny to try different materials for the baffle, so I took MDF, birch ply, different wood sheets and made some baffles. I made some listenings too and now I think to have the optimum, for me at least: a sandwich, with 4 mm poplar ply, 10 mm beech ply and again 4 mm poplar ply. I can't explain why, however. Maybe because of the combination of soft and rigid/hard woods, but I really don't know. And it may or may not work with other speakers. But it's funny
Hi,
today I made some measurements with the asymmetrical baffle in comparison with the centered driver, here the results:
Centered driver:
Asymmetrical baffle:
The asymmetrical baffle seems to have a more even frequency response between 500 Hz and 2 kHz, but above this range it begins to make waves and here the centered driver works better.
Stan made a sim for the off-center driver, I made another one with slightly different parameters, the difference in driver position is 1 cm, so not really relevant - but the centered driver shows a better frequency response. And - it sounds better, especially in the middle range more lively. Where is the error, if there is one at all? It's "always said" that a centered FR is actually a no go, although Pa's SLOB and many other designs are also built that way. Should I stop worrying about this? - and listen to my ears. Here is the picture of the asymmetrical baffle (just a trial version):
today I made some measurements with the asymmetrical baffle in comparison with the centered driver, here the results:
Centered driver:
Asymmetrical baffle:
The asymmetrical baffle seems to have a more even frequency response between 500 Hz and 2 kHz, but above this range it begins to make waves and here the centered driver works better.
Stan made a sim for the off-center driver, I made another one with slightly different parameters, the difference in driver position is 1 cm, so not really relevant - but the centered driver shows a better frequency response. And - it sounds better, especially in the middle range more lively. Where is the error, if there is one at all? It's "always said" that a centered FR is actually a no go, although Pa's SLOB and many other designs are also built that way. Should I stop worrying about this? - and listen to my ears. Here is the picture of the asymmetrical baffle (just a trial version):
Hi Istvan,
Great to see you tinkering with it, the fun, and doubt is endless.
Yes, "stop worrying and learn to love the bomb".
Otherwise if you get the woof up closer to the FR you might get an even better response from both variants.
Something like the image below
Cheers
Stan
Great to see you tinkering with it, the fun, and doubt is endless.
Yes, "stop worrying and learn to love the bomb".
Otherwise if you get the woof up closer to the FR you might get an even better response from both variants.
Something like the image below
Cheers
Stan
Attachments
Hi Stan,
many thanks!
Oh yes, it's so much fun, even the worrying
Thanks for the idea, I could remove the red panel (which IS intended to make fine tunings, so it wasn't a bad idea ).
Maybe I can insert it under the woofer, which will reduce the floor effect and maybe flatten the 250 Hz peak.
The fun continues
many thanks!
Oh yes, it's so much fun, even the worrying
Thanks for the idea, I could remove the red panel (which IS intended to make fine tunings, so it wasn't a bad idea
).Maybe I can insert it under the woofer, which will reduce the floor effect and maybe flatten the 250 Hz peak.
The fun continues
So, after the reconfiguration of the baffle I got the following frequency responses:
Asymmetrical baffle:
Centered FR:
With centered FR and higher position of the woofer I loose about 2-3 dB between 55 and 150 Hz, and the peak at 200 Hz is also 2-3 dB lower, but it's there (it's the room, I think). But I also get a nasty 500 Hz peak which isn't there with the woofer in the original lower position. Mids and treble also look better with the original woofer position and centered driver.
Conclusion: never change a winning team where "winning" not necessary means "the best", but the best possible with this hardware instead. Maybe. I think. But I'm not sure. Oh come on
Asymmetrical baffle:
Centered FR:
With centered FR and higher position of the woofer I loose about 2-3 dB between 55 and 150 Hz, and the peak at 200 Hz is also 2-3 dB lower, but it's there (it's the room, I think). But I also get a nasty 500 Hz peak which isn't there with the woofer in the original lower position. Mids and treble also look better with the original woofer position and centered driver.
Conclusion: never change a winning team
where "winning" not necessary means "the best", but the best possible with this hardware instead. Maybe. I think. But I'm not sure. Oh come on
Hi,
despite the heat wave, I wanted to try something...
Sim and measurements have shown that the treble is a little wavy and that the FR has deficits in the upper treble range. So I attached two GRS PT2522-4 in a custom waveguide with a cap, while losing what felt like 2 liters of water...
And... Oh yes... Wow. The treble has become much better, there is also a nice sheen, without any sharpness, the high hats are wonderful, and the mid-range seems to benefit too, it has become more airy, the previously rather slightly dark sound character is now "brighter", just beautiful.
Measurement will follow... after the cold front next week🥵
despite the heat wave, I wanted to try something...
Sim and measurements have shown that the treble is a little wavy and that the FR has deficits in the upper treble range. So I attached two GRS PT2522-4 in a custom waveguide with a cap, while losing what felt like 2 liters of water...
And... Oh yes... Wow. The treble has become much better, there is also a nice sheen, without any sharpness, the high hats are wonderful, and the mid-range seems to benefit too, it has become more airy, the previously rather slightly dark sound character is now "brighter", just beautiful.
Measurement will follow... after the cold front next week🥵
Last edited:
I could´nt wait, so here are the measurements, with some surprise and aha-effect...
First measurement - listening place, tweeter on top, the FR baffle with the tweeter tilted forward about 5° (which means an upright position, because the baffle is normally tilted back at 5°): the highs are falling down; the sheen what I desrcibed yesterday could come from the more flattened frequency response.
Second measurement - same as the first, but baffle in original position: less loss in the treble range, but the loss is still there
Third measurement - listening place, the original configuration without added tweeter: flat response from 400 Hz
Fourth measurement - original configuration without added tweeter at 1 m distance: slightly rising treble, but still good, the 230 Hz peak is less pronounced
Fifth measurement - original configuration without added tweeter at 0.5 m distance: similar to the 1 m mesurement, the 230 Hz peak is almost gone (it MUST be room induced), the treble still looks fine
Conclusions:
1. The additional tweeter seems to make more trouble than advantages, so it will be eliminated - this was planned as a two way system, and was´nt intended to hook up a further driver.
2. I must optimize the room.
Questions:
1. How can I get the 0.5-1 m frequency response of the +400 Hz range at the normal listening distance? Is it room related too?
2. Is the 230 Hz peak really room related, or is there some discrepancy at the xo point?
First measurement - listening place, tweeter on top, the FR baffle with the tweeter tilted forward about 5° (which means an upright position, because the baffle is normally tilted back at 5°): the highs are falling down; the sheen what I desrcibed yesterday could come from the more flattened frequency response.
Second measurement - same as the first, but baffle in original position: less loss in the treble range, but the loss is still there
Third measurement - listening place, the original configuration without added tweeter: flat response from 400 Hz
Fourth measurement - original configuration without added tweeter at 1 m distance: slightly rising treble, but still good, the 230 Hz peak is less pronounced
Fifth measurement - original configuration without added tweeter at 0.5 m distance: similar to the 1 m mesurement, the 230 Hz peak is almost gone (it MUST be room induced), the treble still looks fine
Conclusions:
1. The additional tweeter seems to make more trouble than advantages, so it will be eliminated - this was planned as a two way system, and was´nt intended to hook up a further driver.
2. I must optimize the room.
Questions:
1. How can I get the 0.5-1 m frequency response of the +400 Hz range at the normal listening distance? Is it room related too?
2. Is the 230 Hz peak really room related, or is there some discrepancy at the xo point?
Hi,
although very busy, I was extremely curious how the speakers would sound with the new serial xo.
Basically it's Stan's design, which I modified slightly and soldered together quick & dirty earlier. I liked the sound and decided back then to finish building the xo with premium components (as previously reported).
Well, the time has come.
Mundorf copper foil inductor, Dayton and Mundorf audio grade resistors, ClarityCap, gold terminal, Elecaudio wiring - yolo
The result: permanent goosebumps. A hitherto unknown to me, incredibly clear, detailed sound, wide, deep and highly fanned out sound, immense joy of playing and speed. Suddenly I don't even think about the need for a tweeter. The lows aren't bloated - no measurements taken yet but I would bet that the 230 Hz peak is gone; still a lot of bass pressure. Mind you - the xo that was soldered together as a test at the time sounded different, even if the basic characteristics were there. Now everything is a few steps higher. I'm deeply impressed, I didn't expect something like that. Many thanks, @Stanislav
Have I arrived? Oh well. But... diy is diy is diy... something will come to mind
Maybe my Harsh-like xo with premium components. Or some premium driver. Baffle experiments. Heaven knows...
The journey goes on...
although very busy, I was extremely curious how the speakers would sound with the new serial xo.
Basically it's Stan's design, which I modified slightly and soldered together quick & dirty earlier. I liked the sound and decided back then to finish building the xo with premium components (as previously reported).
Well, the time has come.
Mundorf copper foil inductor, Dayton and Mundorf audio grade resistors, ClarityCap, gold terminal, Elecaudio wiring - yolo
The result: permanent goosebumps. A hitherto unknown to me, incredibly clear, detailed sound, wide, deep and highly fanned out sound, immense joy of playing and speed. Suddenly I don't even think about the need for a tweeter. The lows aren't bloated - no measurements taken yet but I would bet that the 230 Hz peak is gone; still a lot of bass pressure. Mind you - the xo that was soldered together as a test at the time sounded different, even if the basic characteristics were there. Now everything is a few steps higher. I'm deeply impressed, I didn't expect something like that. Many thanks, @Stanislav
Have I arrived? Oh well. But... diy is diy is diy... something will come to mind
Maybe my Harsh-like xo with premium components. Or some premium driver. Baffle experiments. Heaven knows...
The journey goes on...
Last edited:
Hi again,
after long listening sessions I had the impression that the upper mids are hot, screaming, sometimes sharp, but in this range the frequency response was fine... So I covered the back of the FR baffle with 3 mm Noico and more important with eggcrate foam. Further, I put a felt strip into the cutout.
Well... sharpness is gone, the upper mids are clearer and have more space. Cool
after long listening sessions I had the impression that the upper mids are hot, screaming, sometimes sharp, but in this range the frequency response was fine... So I covered the back of the FR baffle with 3 mm Noico and more important with eggcrate foam. Further, I put a felt strip into the cutout.
Well... sharpness is gone, the upper mids are clearer and have more space. Cool
Hi... oh well...Have I arrived? Oh well. But... diy is diy is diy... something will come to mind
I didn't think it would happen so quickly. The next step - an FR a step or two higher, the TangBand W8-2145. Sim made and it looks good, it's a drop-in replacement for the SB Acoustics, so why not? The drivers and the additional xo parts (the peak in the higher treble "must" be flattened) should arrive this week, so there will be something to report again... it's so much fun
Are you using any felt or other isolation material between the baffles and legs? I have to eminence alpha 15a and its vibrating very much and which is causing some issues with FR.Hi again,
after long listening sessions I had the impression that the upper mids are hot, screaming, sometimes sharp, but in this range the frequency response was fine... So I covered the back of the FR baffle with 3 mm Noico and more important with eggcrate foam. Further, I put a felt strip into the cutout.
Well... sharpness is gone, the upper mids are clearer and have more space. Cool
View attachment 1211603
You would need to support the whole speaker body to the back, like fence posts and their side supports, in order to stop it from vibrating.
Just relying on thick baffles , thick upright supports and some buffer between baffles is not going to eliminate the vibration.
Imagine for a minute that 1.2 - 1.5 m tall, 0.4 - 0.5 m wide structure and its mass.
Rigidity rules here.
