It was my go-to choice when the client wanted/afford small, high SQ drivers; otherwise normally used inexpensive ceiling PA coax.
Sorry
I personally added a ribbon tw cheap with two capacitors 1+0.22 and It immediately got along well with the full range
I personally added a ribbon tw cheap with two capacitors 1+0.22 and It immediately got along well with the full range
You are right in ways. Lol. Out of curiousity I bought the SB to test. Right off the bat yes the highs was smoother as compared to my Coral Flat 8A but what
the SB looses out is the tone, its sounds kind of thick & muff when compared to the Corals. This Im sure is due to the thicker cone of the SB. My test is on OB maybe the SB will sound better in a box who knows. Glad that I've heard the sound of SB as it give me the inspiration & drive to tweaked & improve my Corals.
This I have to thank all who share their experiences here in Diyaudio as I tested out the weaks that I've read, good learning experience for me. Adding phase plug was a good step forward but the big surprise was the ping pong ball trick, improve clarity even more. Oh thanks to Nandappe for his magnet tip these including other tweaks have totally brought the Flat 8a to a really high level of SQ
My bad- I didn't know the frame is plastic.
Could try decoupling the driver. Rope caulking, plumbers putty, silicone, damp the frame etc
http://www.meadowlarksings.com/common features.htm
Attachments
I added simple ribbon tw buy on aliexpress with only capacitors 1uf+0.22(1.22uf without inductance). Cable duelund. Is much better for my tqwt(ddvp20) with sb20. Sorry my english...
I’d really appreciate any opinions on a design I’m considering for the SB20frpc30.
First off, I haven’t built a new speaker in maybe 4 years, but I do have a good amount of speaker design and building experience and I usually follow just about all best practices. I extensively model, measure and tweak all of my designs and try pretty hard not to push drivers out of their comfort zone. I exclusively build multi-way speakers anymore, but I’ve built a handful of single driver designs, and I usually find them to be somewhat limited and flawed (sorry!), so I’m not exactly looking for best practices advice so much as will it work pretty well kind of advice. I don’t expect this to be a reference type of design, but I don’t want to build in extra deficiencies either. I don’t really listen that “critically” anymore, and these will be placed to the side of the listening area in my living room, which actually sounds very good, but obviously isn’t great for imaging. . I have a separate media room for a more traditional setup.
Ok, so I’m thinking of a 1.5 way with the SB20PFCR30-8 in a pretty undersized sealed or aperiodic alignment for both drivers. About 30L total. Possibly using the large cap trick to flatten the high Q bump around 100hz. My goal is not deep bass extension, but instead solid mid bass with an f6 between 60-80hz. These will be crossed to a subwoofer. I want to go with a 1.5 way to deal with baffle step and keep the sensitivity high. And double up on cone area and power handling of course. The SB20PFCR30-8 should give me 4-5db baffle step instead of the full 6db when using identical drivers. Right?
My biggest question is, does undersizing the enclosure negatively affect the midrange quality? Does relieving the back pressure with a leaky enclosure fix this if so? Or is this really just a bass quality issue?
Also, with the somewhat different specs of the two drivers, can they be placed in a shared space or do I need to divide them? I’m thinking of a classic looking design similar to this https://www.fyneaudio.com/product/vintage-classic-viii-sm/ with the aperiodic vents exiting through the lower grill cloth. That’d be a lot easier with a shared cavity.
That’s about it. Feel free to weigh in on any other issues with the design and I apologize for the novel!
First off, I haven’t built a new speaker in maybe 4 years, but I do have a good amount of speaker design and building experience and I usually follow just about all best practices. I extensively model, measure and tweak all of my designs and try pretty hard not to push drivers out of their comfort zone. I exclusively build multi-way speakers anymore, but I’ve built a handful of single driver designs, and I usually find them to be somewhat limited and flawed (sorry!), so I’m not exactly looking for best practices advice so much as will it work pretty well kind of advice. I don’t expect this to be a reference type of design, but I don’t want to build in extra deficiencies either. I don’t really listen that “critically” anymore, and these will be placed to the side of the listening area in my living room, which actually sounds very good, but obviously isn’t great for imaging. . I have a separate media room for a more traditional setup.
Ok, so I’m thinking of a 1.5 way with the SB20PFCR30-8 in a pretty undersized sealed or aperiodic alignment for both drivers. About 30L total. Possibly using the large cap trick to flatten the high Q bump around 100hz. My goal is not deep bass extension, but instead solid mid bass with an f6 between 60-80hz. These will be crossed to a subwoofer. I want to go with a 1.5 way to deal with baffle step and keep the sensitivity high. And double up on cone area and power handling of course. The SB20PFCR30-8 should give me 4-5db baffle step instead of the full 6db when using identical drivers. Right?
My biggest question is, does undersizing the enclosure negatively affect the midrange quality? Does relieving the back pressure with a leaky enclosure fix this if so? Or is this really just a bass quality issue?
Also, with the somewhat different specs of the two drivers, can they be placed in a shared space or do I need to divide them? I’m thinking of a classic looking design similar to this https://www.fyneaudio.com/product/vintage-classic-viii-sm/ with the aperiodic vents exiting through the lower grill cloth. That’d be a lot easier with a shared cavity.
That’s about it. Feel free to weigh in on any other issues with the design and I apologize for the novel!
Most of your ideas are spot on.
SB20FRPC will blend very well with SB20PFCR in an Aperiodic box as a 1.5 way speaker, it is almost as simple as a single driver but with the extra boost at the low end and no loss of overall sensitivity.
Your proposed volume of 30L is a bit small. Also putting the two drivers in a shared volume is not a good idea in my book. The extra pressure at LF will modulate the precious midrange and no one wants to lose clarity in that area. Two equal volumes, each one with their own aperiodic vent will be magic with these drivers.
Also no need for the "large cap trick" as both drivers have a rather low Qts.
I quickly modelled your idea in Basta with some small changes and it is quite promising.
At each volume set to 20L, Qb 3, LPF on the woof 3.6mH at around 250Hz, the overall sensitivity is around 92dB, with f6 at 48Hz.
I include a screenshot of the Basta simulation.
SB20FRPC will blend very well with SB20PFCR in an Aperiodic box as a 1.5 way speaker, it is almost as simple as a single driver but with the extra boost at the low end and no loss of overall sensitivity.
Your proposed volume of 30L is a bit small. Also putting the two drivers in a shared volume is not a good idea in my book. The extra pressure at LF will modulate the precious midrange and no one wants to lose clarity in that area. Two equal volumes, each one with their own aperiodic vent will be magic with these drivers.
Also no need for the "large cap trick" as both drivers have a rather low Qts.
I quickly modelled your idea in Basta with some small changes and it is quite promising.
At each volume set to 20L, Qb 3, LPF on the woof 3.6mH at around 250Hz, the overall sensitivity is around 92dB, with f6 at 48Hz.
I include a screenshot of the Basta simulation.
Fantastic, thanks for taking the time to do that! I’ve only brought them into unibox at this point, but I’ll model them more completely before I move forward. The Qts isn’t “that” low really, .45 for the SB20FRPC and probably over .40 for the woofer after resistance of the coil is included. But the point is taken, I doubt I’d need the cap when quasi aperiodic loading, if that’s even advisable?
Are you saying that the added LF pressure of the woofer would be the problem with the shared space? Since the SB20FRPC is run without a crossover it’ll already be creating its own LF pressure.
I’ve never used Basta, is Qb equivalent to Qa or Ql? Or is it box Q (unlikely)?
Are you saying that the added LF pressure of the woofer would be the problem with the shared space? Since the SB20FRPC is run without a crossover it’ll already be creating its own LF pressure.
I’ve never used Basta, is Qb equivalent to Qa or Ql? Or is it box Q (unlikely)?
Btw, I get Q of 1.05 for the SB20FRPC and .9 for the SB20PFCR with some added resistance for the inductor. That’s 15L each with no fill and before introducing any leakage or port resistance. Isn’t that just about perfect as a starting point for an aperiodic enclosure? FYI, I won’t be attempting to completely flatten the impedance peak, just shaping the drivers’ roll off for a final Q of .7-.8
You are most welcome.
I like Basta and have been using it for a number of years.
It was created by a member of this forum, Svante. Not sure if he's still active.
Basta is very intuitive, very accurate in its predictions, caters for almost anything, except TL, MLTL and such.
Disappointingly for me it doesn't do series XO, maybe because the author stopped developing it further.
Check the links:
https://www.tolvan.com/index.php?page=/basta/basta.php
Qb in Basta is the same as Ql in other programs, represents the losses and damping of the system.
https://www.tolvan.com/basta/Tutorials.htm#_Toc198915347
I like Basta and have been using it for a number of years.
It was created by a member of this forum, Svante. Not sure if he's still active.
Basta is very intuitive, very accurate in its predictions, caters for almost anything, except TL, MLTL and such.
Disappointingly for me it doesn't do series XO, maybe because the author stopped developing it further.
Check the links:
https://www.tolvan.com/index.php?page=/basta/basta.php
Qb in Basta is the same as Ql in other programs, represents the losses and damping of the system.
https://www.tolvan.com/basta/Tutorials.htm#_Toc198915347
Actually, I’d heard of Basta, but now I remember that I was following along his thread when he was developing it. I probably did try it back then. The screen grab you posted looked a lot like The Edge to me and now I know why!
I always use a variety of free programs for the modeling, mainly Jeff Bagby’s spreadsheets plus Unibox and WinISD.
Does anyone else want to weigh in? Stanislav says 30l is too small for the two drivers (15L each) and while it sounds too small to me too, my modeling shows box Q of about 1.0 before any stuffing, which as I understand it, is about perfect for a starting point for an aperiodic enclosure. I’ve never built one though.
Edit: actually I’ve done a couple of stuffed TL mid enclosures, but they weren’t undersized.
Edit: actually I’ve done a couple of stuffed TL mid enclosures, but they weren’t undersized.
I would ignore what Unibox and WinISD tell me and model the whole thing in Basta!
Basta is extremely flexible, can work with several volumes (systems) at the same time and give you immediate idea what's happening.
As you want aperiodic loading you can set your volume, set Qb = 3 and see the result. Vary the volume(s) by 4 - 5L up and down and observe the response.
The theory is a nice thing to know and take into account but the practice can be quite different at times.
There is no reason why an aperiodic box needs to be undersized. I guess this comes from the old days when Dynaudio were active in aperiodic designs with some of their high Qts drivers.
I personally stopped paying too much attention to Qt values as predicted long time ago. I mostly observe the impedance as that will tell me a lot about what to expect. When I build the actual body, being that a box or an OB, I find Basta nearly spot on in its prediction.
Basta is extremely flexible, can work with several volumes (systems) at the same time and give you immediate idea what's happening.
As you want aperiodic loading you can set your volume, set Qb = 3 and see the result. Vary the volume(s) by 4 - 5L up and down and observe the response.
The theory is a nice thing to know and take into account but the practice can be quite different at times.
There is no reason why an aperiodic box needs to be undersized. I guess this comes from the old days when Dynaudio were active in aperiodic designs with some of their high Qts drivers.
I personally stopped paying too much attention to Qt values as predicted long time ago. I mostly observe the impedance as that will tell me a lot about what to expect. When I build the actual body, being that a box or an OB, I find Basta nearly spot on in its prediction.
These days I would use VituixCAD instead of Basta.
Although Basta's interface is a bit easier and quicker to work with.
It's true that WinISD can be a bit off, this is mostly due to certain Q parameters of the port and cabinet, as well as port end correction that are way to optimistic.
(which I change by default btw)
However, WinISD is much easier and quicker to compare multiple drivers or ideas at once.
Keep in mind that the new version has some major bugs with active filters.
So it could be useful to just use all three of them in a work flow.
First to do some quick preliminary research, while using VituixCAD to finish the final design.
For a quick peak at certain baffle design and bafflestep, I use Edge as well for example.
Although Basta's interface is a bit easier and quicker to work with.
It's true that WinISD can be a bit off, this is mostly due to certain Q parameters of the port and cabinet, as well as port end correction that are way to optimistic.
(which I change by default btw)
However, WinISD is much easier and quicker to compare multiple drivers or ideas at once.
Keep in mind that the new version has some major bugs with active filters.
So it could be useful to just use all three of them in a work flow.
First to do some quick preliminary research, while using VituixCAD to finish the final design.
For a quick peak at certain baffle design and bafflestep, I use Edge as well for example.
I’ll give Basta a try again. I may even still have a copy in my speaker programs folder. I was using Jeff Bagby’s box modeling program last night and I could more clearly see what was going on with the responses and I think you’re right about a bit larger enclosure. Even just a couple liters per driver gives a cleaner looking knee.
I actually like unibox best for quickly checking a driver’s box needs and bass extension. And as b_force states, WinISD is pretty good for comparing multiple drivers and the effects of bass EQ. For baffle step and diffraction with standard shaped baffle JB’s room and boundary simulator is pretty useful and easy to use.
I actually like unibox best for quickly checking a driver’s box needs and bass extension. And as b_force states, WinISD is pretty good for comparing multiple drivers and the effects of bass EQ. For baffle step and diffraction with standard shaped baffle JB’s room and boundary simulator is pretty useful and easy to use.
I never intended to derail this thread with a discussion about modeling software, but since it’s gone that way… Stanislav, I did indeed have a copy of Basta! on my computer and based on the saved work sessions, I’ve actually used it sometime back!
I input the two drivers here and I’m having a couple of issues that I’m hoping you have easy answers for.
First of all, the plotted sensitivity is basically 10db low. The driver sensitives are correct and the source voltage is 2.83V.
Secondly, the phase is rotating excessively, to the point that it’s basically useless, even with no crossover applied.
Third, 3.5mH does not roll the woofer off the same as yours did. It’s closer to 700hz and there’s a phase mismatch between about 700-2k.
Do you have any idea what I’m doing wrong? I’ve gone through every possible setting that I can find. Thanks
I input the two drivers here and I’m having a couple of issues that I’m hoping you have easy answers for.
First of all, the plotted sensitivity is basically 10db low. The driver sensitives are correct and the source voltage is 2.83V.
Secondly, the phase is rotating excessively, to the point that it’s basically useless, even with no crossover applied.
Third, 3.5mH does not roll the woofer off the same as yours did. It’s closer to 700hz and there’s a phase mismatch between about 700-2k.
Do you have any idea what I’m doing wrong? I’ve gone through every possible setting that I can find. Thanks