I'm planning on building some speakers for my workshop. I will be using the SB20PFC30 8in woofer. As for the mid-high, I'm hoping I could get away with TC9FD only. Judging by the datasheet, it doesn't do well off axis in the high range, but for these kind of speakers I don't think it matters.
As for the circuitry:
- active crossovers and bi-amplification: I already have boards for crossovers and amps, plus everything will be in the enclosure
- passive crossovers: no experience with them whatsoever; I could do measurements and plug them in that crossover app, but it's more complicated; the advantage is that I already have the amp for them
- if a 3-way is recommended, the woofer will be powered separately and the mid + tweeter will have a passive crossover
Winisd simulation:
Sketch:
As for the circuitry:
- active crossovers and bi-amplification: I already have boards for crossovers and amps, plus everything will be in the enclosure
- passive crossovers: no experience with them whatsoever; I could do measurements and plug them in that crossover app, but it's more complicated; the advantage is that I already have the amp for them
- if a 3-way is recommended, the woofer will be powered separately and the mid + tweeter will have a passive crossover
Winisd simulation:
Sketch:
I've had the same idea, perhaps check out this thread:
10F/8424 & RS225-8 FAST / WAW Ref Monitor
I'm suggesting you have an optic at that thread because the TC-9 has similar specs to the Scan-Speak used in that project, and by all accounts, very good sound. Apparently its main drawback is relatively low efficiency; this may not matter so much in your application.
The SB20PFC, a paper woofer, has very similar specs and frequency response to the RS225, at least, up to about 1,000Hz. It may well measure differently to the Dayton once installed in a cabinet of course.
A member on stereo net au forum has done a similar project with those drivers, sealed and crossing over at second order at 600Hz, happy with the sound. Unfortunately, the XO values are n/a.
I've done an Xsim model, but it's only a sim - without measurements - which I could post if you're interested.
Geoff
10F/8424 & RS225-8 FAST / WAW Ref Monitor
I'm suggesting you have an optic at that thread because the TC-9 has similar specs to the Scan-Speak used in that project, and by all accounts, very good sound. Apparently its main drawback is relatively low efficiency; this may not matter so much in your application.
The SB20PFC, a paper woofer, has very similar specs and frequency response to the RS225, at least, up to about 1,000Hz. It may well measure differently to the Dayton once installed in a cabinet of course.
A member on stereo net au forum has done a similar project with those drivers, sealed and crossing over at second order at 600Hz, happy with the sound. Unfortunately, the XO values are n/a.
I've done an Xsim model, but it's only a sim - without measurements - which I could post if you're interested.
Geoff
I'd use the TG9, it's more efficient & thus a better match for the SB; as for the crossover, with this type of design, if you crossover at the baffle step frequency you can get away with a simple 1st order crossover, i.e. 2 components, maybe 3 if you feel the need for level matching
Agreed, the TG9 is a better option: looks cool, too.
If you haven't seen them, here are some box alignment ideas for the SB20 from the Madisound web site:
"Some possible box alignments:
Sealed box of 0.5 to 0.75 cubic foot for a 3dB down of about 70Hz
Vented box of 0.75 cubic foot with 2" Ø vent by 5" long for f3 of 55Hz
Vented box of 1.0 cubic foot with 2" Ø vent by 4" long for f3 of 49Hz ((i.e. about 30 litres, your model)
Vented box of 1.25 cubic feet with 2" Ø vent by 3.5" long for f3 of 45Hz
Vented box of 1.5 cubic feet with 2" Ø vent by 3" long for f3 of 42Hz"
I've used the SB16pfc in a two way, sounds really nice so the SB20 should be good too. But the frame shape is a pain....
HTH
Geoff
If you haven't seen them, here are some box alignment ideas for the SB20 from the Madisound web site:
"Some possible box alignments:
Sealed box of 0.5 to 0.75 cubic foot for a 3dB down of about 70Hz
Vented box of 0.75 cubic foot with 2" Ø vent by 5" long for f3 of 55Hz
Vented box of 1.0 cubic foot with 2" Ø vent by 4" long for f3 of 49Hz ((i.e. about 30 litres, your model)
Vented box of 1.25 cubic feet with 2" Ø vent by 3.5" long for f3 of 45Hz
Vented box of 1.5 cubic feet with 2" Ø vent by 3" long for f3 of 42Hz"
I've used the SB16pfc in a two way, sounds really nice so the SB20 should be good too. But the frame shape is a pain....
HTH
Geoff
Last edited:
Not sure if I've done this the right way, but if the baffle is 30cm wide it looks like the baffle step frequency would be 383 Hz, (115/0.3), not sure if that's correct.
Once I confirm the baffle step I could tweak my sim.
Geoff
Last edited:
Pretty much on par with what winisd suggests
That should be worth a try! According to my sketch, the baffle step frequency should be around 426Hz.
That looks really nice, I've read that thread a while ago. But the drivers are about 3 times the price of the SB+TG.
Last edited:
The thread includes some discussion about using the Vifa instead of the ScanSpeak in this project, and other threads on this board compare the drivers.
The SB20 has very similar specs and frequency response to the RS225, although its Xmax is 4.5mm compared with 7mm for the Dayton, so the bass at higher volumes won't be quite as good. But it's 1/3 the price....
My point was that with some crossover tweaking, you might be able to do something similar to the RS225/Scanspeak design, for a fraction of the price.
Of course, I haven't built it so I don't know whether it would work.
Geoff
The SB20 has very similar specs and frequency response to the RS225, although its Xmax is 4.5mm compared with 7mm for the Dayton, so the bass at higher volumes won't be quite as good. But it's 1/3 the price....
My point was that with some crossover tweaking, you might be able to do something similar to the RS225/Scanspeak design, for a fraction of the price.
Of course, I haven't built it so I don't know whether it would work.
Geoff
I know, the RS225 seems like a fantastic driver. I was actually considering it for my main audio system, but I went with a sub for the low end instead.
I actually played around with xsim and the plots from the datasheets. Not sure how accurate it is, but it doesn't look that bad to me.
A few things about your sim:
you should never put a resistor in series with a woofer: it is a serious fire hazard. Instead, I suggest you control the woofer output by increasing the value of the inductor;
what's the crossover point in your sim, please?
what data inputs did you use - measured, or taken from the manufacturer's spec sheet or other sources like an existing project;
I'd be tempted to use at least a second order on the TG9 to roll it off more steeply, so that it doesn't try to play bass.
I'm messing about with a possible XO for these drivers, but it's only based on the data sheets. I'm not quite happy with it but will post it when it looks OK.
Geoff
you should never put a resistor in series with a woofer: it is a serious fire hazard. Instead, I suggest you control the woofer output by increasing the value of the inductor;
what's the crossover point in your sim, please?
what data inputs did you use - measured, or taken from the manufacturer's spec sheet or other sources like an existing project;
I'd be tempted to use at least a second order on the TG9 to roll it off more steeply, so that it doesn't try to play bass.
I'm messing about with a possible XO for these drivers, but it's only based on the data sheets. I'm not quite happy with it but will post it when it looks OK.
Geoff
Last edited:
Worth checking port velocity with an 8” woofer and a 2” port you may be marginal, sub 20m/s is what you are aiming for at you max input watts, these can be got from WinISD, also check cone excursion at max amp watts to make sure you don’t exceed.
With these component values, about 5.5kHz
I used FPGraphTrace to take snapshots of the graphs from the datasheets, as I don't have the drivers yet
Winisd always gives good results even with ridiculous port diameters. As far as I remember, this combination gives around 5m/s. I don't know how much to trust it though.
I also use fptrace, and SB supplied data is generally regarded as accurate, bearing in mind that it's not measured in your cabinet. Unfortunately I don't have measuring equipment.
I'd be crossing this woofer at a maximum of 1.5KHz, preferably below 1000 Hz. I don't really understand all the technical reasons and physics, but if you do a search on 'where to cross 8" woofer" you will find some discussions with people far more informed than me: Design-Criteria
You can see from the product specs below that the woofer starts to 'beam' at that point (1500Hz). That is, the high frequencies start to drop off as the measurement mike moves away from in front of the driver.
5,000 Hz is really tweeter or full range (TC9, TG9) territory. The SB16 beams from about 1,800 or so, my project using that driver crossed it at about 2200 and it sounds good. The graphs don't show it but I expect that both woofers would show some harshness and distortion if pushed with too high a crossover point.
I'm still working on my sim: as I said above, I've also been thinking about a similar project and will post the sim when I'm happy. At the moment I'm using a second order with a 3mH inductor on the SB20 and a third order on the TG-9 starting with a 16 microfarad cap on the TG9, plus shaping/contouring parts on both drivers. Crossover point is about 800 Hz.
I tried using the same crossover as the RS225/ScanSpeak project but it didn't look very good despite the similarity of the product specs.
I'm very much a newbie at this and I really hope other people with more knowledge and expertise than I will respond to your questions.
Geoff
I'd be crossing this woofer at a maximum of 1.5KHz, preferably below 1000 Hz. I don't really understand all the technical reasons and physics, but if you do a search on 'where to cross 8" woofer" you will find some discussions with people far more informed than me: Design-Criteria
You can see from the product specs below that the woofer starts to 'beam' at that point (1500Hz). That is, the high frequencies start to drop off as the measurement mike moves away from in front of the driver.
5,000 Hz is really tweeter or full range (TC9, TG9) territory. The SB16 beams from about 1,800 or so, my project using that driver crossed it at about 2200 and it sounds good. The graphs don't show it but I expect that both woofers would show some harshness and distortion if pushed with too high a crossover point.
I'm still working on my sim: as I said above, I've also been thinking about a similar project and will post the sim when I'm happy. At the moment I'm using a second order with a 3mH inductor on the SB20 and a third order on the TG-9 starting with a 16 microfarad cap on the TG9, plus shaping/contouring parts on both drivers. Crossover point is about 800 Hz.
I tried using the same crossover as the RS225/ScanSpeak project but it didn't look very good despite the similarity of the product specs.
I'm very much a newbie at this and I really hope other people with more knowledge and expertise than I will respond to your questions.
Geoff
Last edited:
Yes, that is true. As I said, passive crossovers are "I have no idea what I'm doing" territory. For me it's just easier to throw in an active crossover and two chip amps, albeit a bit more work with wiring and power supplies, but much easier when it comes to design.
That could indeed be a simpler way to go; unfortunately I don't have the equipment to do that.
I'm trying a 4mH inductor and 20 microfarad 'shunt' cap on the woofer and a 16 microfarad cap with a 2mH and 4 ohm resistor as a shunt to shape the response of the TG9:
Still needs some work - and by someone who knows more than I do.
I'm surprised that there aren't more projects which use the SB20: the stereo net au is the only one I've seen. The crossover point in that project was 600Hz, which I think is more appropriate than what I've come up with. The SBPFC series are very well made and the new versions have a round frame which makes them easier to mount. NB: cost of the SB20 in Oz: $52, cost of the RS225: $130. TG-9 $35, Scan Speak about $150: you can see why I've thought about this project.
Geoff
I'm trying a 4mH inductor and 20 microfarad 'shunt' cap on the woofer and a 16 microfarad cap with a 2mH and 4 ohm resistor as a shunt to shape the response of the TG9:
Still needs some work - and by someone who knows more than I do.
I'm surprised that there aren't more projects which use the SB20: the stereo net au is the only one I've seen. The crossover point in that project was 600Hz, which I think is more appropriate than what I've come up with. The SBPFC series are very well made and the new versions have a round frame which makes them easier to mount. NB: cost of the SB20 in Oz: $52, cost of the RS225: $130. TG-9 $35, Scan Speak about $150: you can see why I've thought about this project.
Geoff
Last edited:
Active filter simulation
The MFB is a bandpass filter to raise the notch that appears at that frequency. Although not sure if it's needed.
The MFB is a bandpass filter to raise the notch that appears at that frequency. Although not sure if it's needed.
Hi,
I Can confirm that SB20 sounds great! It has a Great kick. Yes, it is not high-end but nothing to complain about. one of few 8” that is very Well behaved in the Roll of.
If going active, and place it close to wall, and boost it some, you Can put it in a 12-15 liter closed cabinet.
You might be missing some in the Upper tweeter range, with the Vifa TC9. Another Way is crossing SB20 with a tweeter.
There is a Danish product named Lemus Home Classic 1500 using this woofer. Have not heard it yet, but Im confident that it Sounds great.
I Can confirm that SB20 sounds great! It has a Great kick. Yes, it is not high-end but nothing to complain about. one of few 8” that is very Well behaved in the Roll of.
If going active, and place it close to wall, and boost it some, you Can put it in a 12-15 liter closed cabinet.
You might be missing some in the Upper tweeter range, with the Vifa TC9. Another Way is crossing SB20 with a tweeter.
There is a Danish product named Lemus Home Classic 1500 using this woofer. Have not heard it yet, but Im confident that it Sounds great.
You could look at the Vifa TC7-04, which plays a little louder and higher than the TC-9; it is also highly regarded. I've simmed a crossover for the TC7 and SB20, I was happier with that sim than the one with the TC9. Main drawback of that sim is high parts values.
I didn't know the SB20 had been used in the Lemus design, it would be interesting to know what tweeter is used there. The SB20 is also used in some Australian retail speakers.
Geoff
I didn't know the SB20 had been used in the Lemus design, it would be interesting to know what tweeter is used there. The SB20 is also used in some Australian retail speakers.
Geoff
I think the tweeter in the Lemus speaker is also an SB. One of them with a small faceplate.
I have heard good things about the newer SB fullrange.
The SB65 variant are also liked by many. I prefer Vifa TC9. SB65 sound more like a true tweeter. Depends on the application.
I have heard good things about the newer SB fullrange.
The SB65 variant are also liked by many. I prefer Vifa TC9. SB65 sound more like a true tweeter. Depends on the application.