Jack,
okay i just checked out that xda unit on the site. wow, i understand now what you are referring to. for a 100.00 ea. i couldn't build a nicer cabinet to house speakers in even if i can't afford the xda unit now.
okay i just checked out that xda unit on the site. wow, i understand now what you are referring to. for a 100.00 ea. i couldn't build a nicer cabinet to house speakers in even if i can't afford the xda unit now.
I'm looking to match a budget sub with the Madisound Recession Buster bookshelves.
BR-Kit
I was wondering if the Foster WF-100k plate amp will be sufficient to push one 11-075-2 in a sealed enclosure. I'm trying to do this on the cheap for my garage/workshop audio. I picked up some free 8, 10, 12 inch Sonotubes and have some scrap 3/4 MDF for endcaps. I figured a 1 cf sonotube with the 075 and Foster plate amp would be sufficient for what I'm trying to do.
I've also considered adding the 11-025-1s to the RB kit for a 3 way but that would be a bit more difficult and my mounting options for the bookshelves is limited.
Thanks,
Frank
BR-Kit
I was wondering if the Foster WF-100k plate amp will be sufficient to push one 11-075-2 in a sealed enclosure. I'm trying to do this on the cheap for my garage/workshop audio. I picked up some free 8, 10, 12 inch Sonotubes and have some scrap 3/4 MDF for endcaps. I figured a 1 cf sonotube with the 075 and Foster plate amp would be sufficient for what I'm trying to do.
I've also considered adding the 11-025-1s to the RB kit for a 3 way but that would be a bit more difficult and my mounting options for the bookshelves is limited.
Thanks,
Frank
Due to the high impedance of the npt-11-075-2 woofers, the WF-100k amplifier will only be able to deliver about 60W or so. It will probably be loud enough for normal listening in the garage, but you will be wasting the large Xmax of the 075 driver.
I have some other 10" and 12" subs that are 4ohms that will work well with the WK-100k amplifier. Xmax is around 10mm. The price is the same, $20/$30.
I have some other 10" and 12" subs that are 4ohms that will work well with the WK-100k amplifier. Xmax is around 10mm. The price is the same, $20/$30.
Thanks Jack. Looking at your spreadsheet I didn't see any subs with the 10mm Xmax. Are other 10s (or 12s) you have that are more appropriate for the small amp not listed? Will they be put up on your web page?
Frank
Frank
Frank,
The 10" and 12" I was referring to in my previous post are not on the website. I'm not going to put them on. I have hundreds of drivers that I'm not going to put up because there are only a couple of them available. It's not worth the 2 hours it takes to add a driver to the site.
I only post drivers that I have at least 50 or so of.
The 10" and 12" I was referring to in my previous post are not on the website. I'm not going to put them on. I have hundreds of drivers that I'm not going to put up because there are only a couple of them available. It's not worth the 2 hours it takes to add a driver to the site.
I only post drivers that I have at least 50 or so of.
Jack Hidley said:
Grab bag anybody?
"5 pairs mixed mids" or "5 pairs mixed woofers" might be a popular item, if the price was low (naturally shipping would bump it up), and allow you to clean out inventory with low piece counts.
I know I'm fighting the urge to snag a pair of XDs, that molded baffle is sexy, and I have a solid active XO.
Another sub question: what about a pair of the 8" Peerless with the Foster plate amp? I was thinking in a sealed box (Q ~0.5) and letting the boost at 35Hz bring up the bottom octave. Good idea? Bad?
BTW ... as shown in the schematic, the R/C combination that forms the Sallen-Key high pass filter has its poles at 30Hz with a Q of 0.5. This is a flat response.
Does the schematic accurately represent what is in the amp as shipped? If so, where does the 6dB bump at 35Hz come from? I guess I assumed (wrongly?) that the 6dB hump was owing to the Q of the 18dB high-pass.
Thanks,
Barry
BTW ... as shown in the schematic, the R/C combination that forms the Sallen-Key high pass filter has its poles at 30Hz with a Q of 0.5. This is a flat response.
Does the schematic accurately represent what is in the amp as shipped? If so, where does the 6dB bump at 35Hz come from? I guess I assumed (wrongly?) that the 6dB hump was owing to the Q of the 18dB high-pass.
Thanks,
Barry
Barry,
I'm not 100% sure if the part values in the schematic are correct or not. They should be.
Did you analyze the entire circuit around IC602 (all three caps and resistors)? Since the second and third RC networks are affected by the OP amp, the behavior of the first RC network isn't simple to analyze by hand. You will need to plug the whole thing into a circuit analysis program.
A long time ago, I used one of these amps to build some power floor monitors. I'm pretty sure that once I retuned this circuit the reponse of the amp became flat.
If anyone modifies the HP circuit, I would just jumper C607, and remove R615. That way you can use very simple math to calculate values for the other two caps and resistors.
I think this amp would work very well with two of the Peerless 8" woofers in a sealed box.
I'm not 100% sure if the part values in the schematic are correct or not. They should be.
Did you analyze the entire circuit around IC602 (all three caps and resistors)? Since the second and third RC networks are affected by the OP amp, the behavior of the first RC network isn't simple to analyze by hand. You will need to plug the whole thing into a circuit analysis program.
A long time ago, I used one of these amps to build some power floor monitors. I'm pretty sure that once I retuned this circuit the reponse of the amp became flat.
If anyone modifies the HP circuit, I would just jumper C607, and remove R615. That way you can use very simple math to calculate values for the other two caps and resistors.
I think this amp would work very well with two of the Peerless 8" woofers in a sealed box.
Thanks for the reply Jack.
I simply looked at the 3 R's and the 3 C's and plugged into a Sallen-Key calculator.
You are correct that I did not include the output impedance of the prior op-amp, nor the impedance seen at the output of the Sallen-Key in my calcs.
Its been a while since circuit analysis class, but my gut says they shouldn't matter much. My gut has been wrong before ... many times!
Look here for a "calculator":
http://sim.okawa-denshi.jp/en/Sallen3tool.php
Barry
I simply looked at the 3 R's and the 3 C's and plugged into a Sallen-Key calculator.
You are correct that I did not include the output impedance of the prior op-amp, nor the impedance seen at the output of the Sallen-Key in my calcs.
Its been a while since circuit analysis class, but my gut says they shouldn't matter much. My gut has been wrong before ... many times!
Look here for a "calculator":
http://sim.okawa-denshi.jp/en/Sallen3tool.php
Barry
One more question (until my next):
The 2x 8" peerless are about the same price ($28) as a single AR 12" ($30).
From a price/performance standpoint would I be a fool not to get the AR 12"?
My calcs are:
2x 8" Peerless 11-047-1 in 74 liter box has Q ~ 0.5
1x 12" AR in 54 liter box has Q ~ 0.5
Why sealed with a low Q? Because of the published 6dB boost at 35Hz on the Foster plate amp.
The 2x 8" Peerless are down ~11 dB at 35Hz, and the 1x AR 12" is down about ~9 dB.
I'm sure there are better design ideas out there than mine, and I'm certainly not married to either of these choices. I do like the price though!
Barry
The 2x 8" peerless are about the same price ($28) as a single AR 12" ($30).
From a price/performance standpoint would I be a fool not to get the AR 12"?
My calcs are:
2x 8" Peerless 11-047-1 in 74 liter box has Q ~ 0.5
1x 12" AR in 54 liter box has Q ~ 0.5
Why sealed with a low Q? Because of the published 6dB boost at 35Hz on the Foster plate amp.
The 2x 8" Peerless are down ~11 dB at 35Hz, and the 1x AR 12" is down about ~9 dB.
I'm sure there are better design ideas out there than mine, and I'm certainly not married to either of these choices. I do like the price though!
Barry
Barry,
I think the single AR 12" is a better design. You get slightly more cone area than 2x Peerless, about the same Xmax, but about double Xmech. So you will be able to get more bass output from the AR 12". You could really get a Qcb of 0.5 in a 45l box when you include some polyester fill.
Any even better idea is this. The guy that owns the WF-100 amps and AR 12" woofers, actually purchased them to start selling subwoofers as a side business a long time ago. It went no where, but he still has a number of the finished products available. One 12" AR driver in a 16" cube made of 3/4" MDF with black HPL (high pressure laminate) on the outside. It has a large 3" or so flared port and the WF-100k has already had the HP filter tuned to work with the cabinet volume and port tuning. Grille is included. $150 each.
I think the single AR 12" is a better design. You get slightly more cone area than 2x Peerless, about the same Xmax, but about double Xmech. So you will be able to get more bass output from the AR 12". You could really get a Qcb of 0.5 in a 45l box when you include some polyester fill.
Any even better idea is this. The guy that owns the WF-100 amps and AR 12" woofers, actually purchased them to start selling subwoofers as a side business a long time ago. It went no where, but he still has a number of the finished products available. One 12" AR driver in a 16" cube made of 3/4" MDF with black HPL (high pressure laminate) on the outside. It has a large 3" or so flared port and the WF-100k has already had the HP filter tuned to work with the cabinet volume and port tuning. Grille is included. $150 each.
Hi Jack,
In the spreadsheet, i see that xdw is listed, but is not available yet. Any idea when it'll be available? I'm interested in 5 xds and 2 xdw 😀
How i wish i could lay my hands on an xda or two! 🙁
Thanks!
In the spreadsheet, i see that xdw is listed, but is not available yet. Any idea when it'll be available? I'm interested in 5 xds and 2 xdw 😀
How i wish i could lay my hands on an xda or two! 🙁
Thanks!
Member
Joined 2003
Crossover work
Hello,
Allow me to share with you the work I have done today. I have used the measurements provided at Jack Hidley's website to simulate a crossover for the XdS speakers. Compared to the crossover provided at Jack's site, I have reduced the BSC a small bit, and added a notch filter for the woofer resonance.
IMAGE: XdS Crossover Schematic
IMAGE: Frequency Response, Transfer Function, and Impedance
For comparison, here's the simulation of the crossover that's included in the measurement package at Jack's site:
IMAGE: Stock Crossover
If anyone is interested in designing a crossover using the FRD tools, I have converted Jack's MLS measurements to FRD and ZMA files. Download here.
The crossover shown above uses values available at Solen. For the inductors, I have used the standard solid wire inductors, not the litz. The 1.5mH and the 0.1mH are 16AWG, the 0.3mH is 20AWG. Cost should be about CAD$30 for each speaker.
Note: The design shown above has not been tested. I cannot tell you "how it sounds".
[Edit: Also note that I am making the following assumptions: 1.5" between acoustic centers (z-axis), center of tweeter is 3.87" above woofer (y-axis)]
Hello,
Allow me to share with you the work I have done today. I have used the measurements provided at Jack Hidley's website to simulate a crossover for the XdS speakers. Compared to the crossover provided at Jack's site, I have reduced the BSC a small bit, and added a notch filter for the woofer resonance.
IMAGE: XdS Crossover Schematic
IMAGE: Frequency Response, Transfer Function, and Impedance
For comparison, here's the simulation of the crossover that's included in the measurement package at Jack's site:
IMAGE: Stock Crossover
If anyone is interested in designing a crossover using the FRD tools, I have converted Jack's MLS measurements to FRD and ZMA files. Download here.
The crossover shown above uses values available at Solen. For the inductors, I have used the standard solid wire inductors, not the litz. The 1.5mH and the 0.1mH are 16AWG, the 0.3mH is 20AWG. Cost should be about CAD$30 for each speaker.
Note: The design shown above has not been tested. I cannot tell you "how it sounds".
[Edit: Also note that I am making the following assumptions: 1.5" between acoustic centers (z-axis), center of tweeter is 3.87" above woofer (y-axis)]
Reid,
Thanks for the alternate crossover design. A couple of things I noticed.
If you compare your simulation of the R1 (my) crossover system response to my actual measurements of that crossover, there are some significant differences. Your simulation shows the phase at the crossover frequency to be quite a bit out. From my measurements, it looks like the actual phase is just right since they sum to 0dB. I do not have the individual measurements here to see if they sum at their -6dB levels. I assume they do. There is about 4dB difference in level around the crossover range between your simulation and my measurements. I think something isn't quite correct with your model. I might be your softwares interpolation routine that is operating below 300Hz causing some phase errors at higher frequencies. Is it possible to tell your software what the response actually does at low frequencies? You could also try manually editing the ASCII file to make the low end do what you know it really does.
Something I do when modeling any speaker system is to load one of the drivers responses into the software with the phase reveresed. Look at the frequency response simulation with both drivers active and no crossover. Compare this to the actual measurement of the same thing. Woofer and tweeter with tweeter phase reveresed. The measurement should have a deep notch. You can compare the shape and depth of this notch to the simulation results. It is an easy way to see the accuracy of the phase modeling. In the data package, this file is XdS woofer-tweeter 1M MLS. In this case, put the ASCII file in the package, but forgot to include the pdf version. I'll put that in tomorrow.
I suspect that the average level in your system design is going to be too high above 2kHz. If you look in the nearfield measurement file for the woofer, you can see that the raised area from 800Hz-2kHz in the response is real.
Edit: I just saw your assumptions about the driver acoustic center locations. Since the mic was never moved during my measurements, there is no reason to have a Z parameter at all. Just set both Z values to 0. That is where your phase error is coming from.
The correct Y value is 4.07". I hid it in the XdS ts.txt file.
Thanks for the alternate crossover design. A couple of things I noticed.
If you compare your simulation of the R1 (my) crossover system response to my actual measurements of that crossover, there are some significant differences. Your simulation shows the phase at the crossover frequency to be quite a bit out. From my measurements, it looks like the actual phase is just right since they sum to 0dB. I do not have the individual measurements here to see if they sum at their -6dB levels. I assume they do. There is about 4dB difference in level around the crossover range between your simulation and my measurements. I think something isn't quite correct with your model. I might be your softwares interpolation routine that is operating below 300Hz causing some phase errors at higher frequencies. Is it possible to tell your software what the response actually does at low frequencies? You could also try manually editing the ASCII file to make the low end do what you know it really does.
Something I do when modeling any speaker system is to load one of the drivers responses into the software with the phase reveresed. Look at the frequency response simulation with both drivers active and no crossover. Compare this to the actual measurement of the same thing. Woofer and tweeter with tweeter phase reveresed. The measurement should have a deep notch. You can compare the shape and depth of this notch to the simulation results. It is an easy way to see the accuracy of the phase modeling. In the data package, this file is XdS woofer-tweeter 1M MLS. In this case, put the ASCII file in the package, but forgot to include the pdf version. I'll put that in tomorrow.
I suspect that the average level in your system design is going to be too high above 2kHz. If you look in the nearfield measurement file for the woofer, you can see that the raised area from 800Hz-2kHz in the response is real.
Edit: I just saw your assumptions about the driver acoustic center locations. Since the mic was never moved during my measurements, there is no reason to have a Z parameter at all. Just set both Z values to 0. That is where your phase error is coming from.
The correct Y value is 4.07". I hid it in the XdS ts.txt file.
Member
Joined 2003
Thanks a lot for this info and for the tips. I'll have to play around with the crossover some more I guess. I did notice that my simulated response was different from you're measured response, but hadn't looked into it a whole lot as of yet.Jack Hidley said:
I'm just getting into crossover design, and your input will help a lot.
Member
Joined 2003
Let's try this again. Here's what I have come up with, using proper driver offsets. Phase alignment isn't quite as good, but still acceptable for me. As far as off-axis response is concerned, using my crossover the speaker will sound better if it it above your head, than below.
IMAGE: Crossover Schematic
IMAGE: Frequency Response, Transfer Function, and Impedance
For comparison, here's the simulation of the crossover that's included in the measurement package at Jack's site (using corrected driver offsets, this is the real response):
IMAGE: Stock Crossover
Note that my crossover uses standard Solen values. Inductors are solid wire, not litz. The 1.5mH and 0.1mH are 16AWG, the 0.3mH is 20AWG. The 0.1mH is most important, as it's DCR of 0.08 Ohms forms part of the notch filter.
IMAGE: Crossover Schematic
IMAGE: Frequency Response, Transfer Function, and Impedance
For comparison, here's the simulation of the crossover that's included in the measurement package at Jack's site (using corrected driver offsets, this is the real response):
IMAGE: Stock Crossover
Note that my crossover uses standard Solen values. Inductors are solid wire, not litz. The 1.5mH and 0.1mH are 16AWG, the 0.3mH is 20AWG. The 0.1mH is most important, as it's DCR of 0.08 Ohms forms part of the notch filter.
Very happy to see some thought going into these XdS crossovers. I'm getting ready to pull the trigger on a pair of them ; not that it's a difficult decision to make given the value here. I simply do not want to under utilize the performance available in them. My c/o design skills are basic and the more I read , the more I realize how little I know. Don't want this to be an unfinished project. I have plenty of those🙁 . This appears to be a great opportunity for me to assemble a really good speaker system ; one I can live with. I would be most interested if anyone else has something to add to this design.
One question I do have. I have several reasonable quality active crossovers(analog). Some 24db L/R , one 48db L/R. If I were to incorporate them with this speaker , how would I impliment the Baffle step that is necessary? I will be using the XdS with and active sub below 125hz
One question I do have. I have several reasonable quality active crossovers(analog). Some 24db L/R , one 48db L/R. If I were to incorporate them with this speaker , how would I impliment the Baffle step that is necessary? I will be using the XdS with and active sub below 125hz