The 083 will work fine in terms of frequency response, when used to 250Hz. It just won't sound as clean as the 116 woofer. If you do this actively, it is not difficult to modify your active filter to add the eq for the 116 woofer. If you are sticking with passive crossovers, then I would use the 083, since the eq option isn't easy to do.
Jack Hidley said:
Well that messes up how I was going to use them. I was wanting them for MBM's covering from about 50Hz to 150Hz. My AVR has a fixed crossover of 80Hz if I select the small setting. I had planned on setting my mains to large and connect to the sub (MBM) using the high level input and then connect the small main to the high level output. I assumed that the output had a fixed high pass.
Jack Hidley said:
Well I can confirm 100% that the high level output does not have any high pass filter on the Foster plate amp. I connected a sub to the high level output on the Foster plate amp and the second sub played the same bass as the first sub.
The passive high level crossovers on plate amplifiers are very simple. Usually they are only a 150-300uF nonpolarized electrolytic capacitor. They might also put a 5-10W 25ohm power resistor across the leads to the satellite speaker also.
They are very ineffective due to the nonflat impedance that almost all loudspeakers present, especially in the bass. Look at the attached M-20 woofer impedance measurement in box. This is typical of a loudspeaker. It isn't a constant 8ohms anywhere. The huge peak in the impedance curve around 70Hz makes the HP filter from the single capacitor basically not function at all. You get filtering below 60Hz, and some filtering above 80Hz. The response ends up very messed up. Maybe someone else can post the transfer function of what I'm describing.
The best way to help the problem is this. If the speaker you are using above 150Hz is vented, plug the port completely. Pull the woofer out and stuff the cabinet with as much Polyester fiber as you can fit in there. It should be very difficult to put the woofer back in. This will drop the peak value of the impedance magnitude in the bass a lot. It will also cut the speakers low bass output quite a bit, which is what you are trying to do with the crossover capacitor in the first place.
They are very ineffective due to the nonflat impedance that almost all loudspeakers present, especially in the bass. Look at the attached M-20 woofer impedance measurement in box. This is typical of a loudspeaker. It isn't a constant 8ohms anywhere. The huge peak in the impedance curve around 70Hz makes the HP filter from the single capacitor basically not function at all. You get filtering below 60Hz, and some filtering above 80Hz. The response ends up very messed up. Maybe someone else can post the transfer function of what I'm describing.
The best way to help the problem is this. If the speaker you are using above 150Hz is vented, plug the port completely. Pull the woofer out and stuff the cabinet with as much Polyester fiber as you can fit in there. It should be very difficult to put the woofer back in. This will drop the peak value of the impedance magnitude in the bass a lot. It will also cut the speakers low bass output quite a bit, which is what you are trying to do with the crossover capacitor in the first place.
Jensen woofer closeout
Has anyone purchased any of the 15' Jensen woofers Jack is closing out? Jack had sent me an earlier email that he was not certain if these foam surround woofers were built using rot resistant foam. They are about 9 years old, if I remember his comments correctly.
They look like a really nice woofer, but I cannot afford messing with them if they would need to be refoamed.
Also, what do you guys think of the AR 12" woofer? I wonder how that would work in a push pull arrangement?
Thanks Jack for all of your time and work, by the way.
Homebuilder
Has anyone purchased any of the 15' Jensen woofers Jack is closing out? Jack had sent me an earlier email that he was not certain if these foam surround woofers were built using rot resistant foam. They are about 9 years old, if I remember his comments correctly.
They look like a really nice woofer, but I cannot afford messing with them if they would need to be refoamed.
Also, what do you guys think of the AR 12" woofer? I wonder how that would work in a push pull arrangement?
Thanks Jack for all of your time and work, by the way.
Homebuilder
I haven't sold a single Jensen 15" woofer yet. I honestly can't believe it.
The drivers are fairly old, but in perfect shape. They were made by Foster about 10 years ago. By this time, pretty much all companies in Taiwan were using foam that has antifungal agents in it and was much more UV and ozone resistant.
I can take my finger and smash it into the surrounds on these woofers no problem. The foam rebounds perfectly afterwards.
The drivers are fairly old, but in perfect shape. They were made by Foster about 10 years ago. By this time, pretty much all companies in Taiwan were using foam that has antifungal agents in it and was much more UV and ozone resistant.
I can take my finger and smash it into the surrounds on these woofers no problem. The foam rebounds perfectly afterwards.
Jack Hidley said:
The mains are very small on wall speakers 30" long, 4" deep and 7" wide. They are RBH WM30's. They have two 4" drivers, two 4" passive radiators and a soft dome tweeter in a curved cast aluminum cabinet. The speakers are a sealed design.
www.rbhsound.com/wm30.shtml
Do you think what you are suggesting will work for such small speakers? Since I use my Denon 3300 as a pre/pro I was going to do the following:
Split the right and left main signals.
High pass the lines going to the mains.
Run the lines (line level) that I branched off of the mains to each sub.
I will be able to adjust the sub crossover where I want, but I will have to use a fixed high pass filter (passive) for the mains. I was planning on going this way because I was concerned with heat build-up in the coil by sending signals that the mains can't play. Please advise?
I'm thinking of buying two of the Advent 12" woofers and Seas 4" midrange drivers (and maybe some of those cheap aluminum dome tweeters) to build a very low budget 3-way. Would this be a good idea? Does anyone know how high the Advent drivers can cross over? It would seem they were used in an MTMW with dome midranges, but I'm just working off of old pictures.
Hey Jack...ideally i'd like to refinish the XDS to fit my project, but i'm a bit concerned with the mounting screw warning on the FAQs page..any tips on removing the tweeter?
OK, so what I do is:Jack Hidley said:
1) Remove R615
2) Jump C607
3) Replace R618 and R617 with 1:2 resistances
Are there preferred values for the ratio, or will any resistances do?
mjg100,
Your RBH speakers are a passive radiator design. The passive radiator behaves exactly the same as a ported speaker, which results in a very complex impedance curve that won't work well with the single capacitor crossover. I can't think of any simple way to keep the radiators from moving, to make it a sealed speaker.
Basically your plan was to have the L and R speaker level output of the Denon drive the RBH speakers, and then take the subwoofer output from the Denon, run it to the Foster plate amp which powers your subwoofers with NHT drivers? Correct?
The AR 12" woofers were used up to 300Hz in the AR Limited 9 loudspeaker.
Removing the screw on the back to take the tweeter out. I've attached an exploded view of another NHT speaker that uses the exact same tweeter mounting design. The only difference is that the screw in the exploded view is inside the main cabinet, since this speaker has a midrange chamber inside the main cabinet. Since the XdS doesn't have the midrange chamber, the screw is on the back of the cabinet.
In the next post, I'll attach directions on how to take apart and reassemble the tweeter from the cabinet.
Regarding the filter component values in the Foster plate amplifier. I just double checked them, and they match the schematic 100%. C1=C2=C3=0.068 uF, R1=56k, R2=22k, R3=390k. So either the filter calculator that Barry posted is incorrect or the 6dB boost is coming from elsewhere in the circuit. In the morning, I'll have a chance to hook up one of the amps to a scope to figure out for sure where the boost is coming from.
For R617 and R618, I would try to keep their values between 10k and 50k, just to keep noise low. But for now, hold off until I have another chance to look at the amp on the bench.
Your RBH speakers are a passive radiator design. The passive radiator behaves exactly the same as a ported speaker, which results in a very complex impedance curve that won't work well with the single capacitor crossover. I can't think of any simple way to keep the radiators from moving, to make it a sealed speaker.
Basically your plan was to have the L and R speaker level output of the Denon drive the RBH speakers, and then take the subwoofer output from the Denon, run it to the Foster plate amp which powers your subwoofers with NHT drivers? Correct?
The AR 12" woofers were used up to 300Hz in the AR Limited 9 loudspeaker.
Removing the screw on the back to take the tweeter out. I've attached an exploded view of another NHT speaker that uses the exact same tweeter mounting design. The only difference is that the screw in the exploded view is inside the main cabinet, since this speaker has a midrange chamber inside the main cabinet. Since the XdS doesn't have the midrange chamber, the screw is on the back of the cabinet.
In the next post, I'll attach directions on how to take apart and reassemble the tweeter from the cabinet.
Regarding the filter component values in the Foster plate amplifier. I just double checked them, and they match the schematic 100%. C1=C2=C3=0.068 uF, R1=56k, R2=22k, R3=390k. So either the filter calculator that Barry posted is incorrect or the 6dB boost is coming from elsewhere in the circuit. In the morning, I'll have a chance to hook up one of the amps to a scope to figure out for sure where the boost is coming from.
For R617 and R618, I would try to keep their values between 10k and 50k, just to keep noise low. But for now, hold off until I have another chance to look at the amp on the bench.
Jack Hidley said:
Basically your plan was to have the L and R speaker level output of the Denon drive the RBH speakers, and then take the subwoofer output from the Denon, run it to the Foster plate amp which powers your subwoofers with NHT drivers? Correct?Basically your plan was to have the L and R speaker level output of the Denon drive the RBH speakers, and then take the subwoofer output from the Denon, run it to the Foster plate amp which powers your subwoofers with NHT drivers? Correct?
Actually I am going to split the front right main channel. Then high pass one leg (100Hz) and connect to the right main. Take the other leg and connect to the sub. This way I am not sending anything below 100Hz to the main. I will do the same with the left main. I am doing it this way so that I can use the 10" XDS driver used in MBM higher than the 80Hz fixed crossover in my AVR. I will do the same with the left main. I have the sub out connected to a larger (6 CF) sub in the back of the room.
Jack Hidley said:
Thanks for checking Jack.
The two cascaded adjustable Sallen-Key LPFs that follow the 3rd order high pass *seem* to be showing a cutoff frequency range of 50Hz-150Hz, which is different from what the knob on the side of the amp shows (50Hz-200Hz).
This can be confirmed using simple calculations using the values shown for R and C.
LPF #1: R1=R2=10k, C1=0.115u, C2=0.097u
LPF #2: R1=R2=10k, C1=0.277u, C2=0.018u
http://en.wikipedia.org/wiki/Sallen_Key_filter
BTW, I am absolutely THRILLED to be able to buy an 80W amp for only $35. I would like to know the specs, however, because my (eventual...long term) intent with this amp is to convert one of the three op-amp filters into a "shelving LPF" a la Linkwitz:
http://www.linkwitzlab.com/images/graphics/shlv-lpf.gif
Thanks for going the extra mile Jack ...
Barry
ps. The schematic is posted here: http://home.comcast.net/~jhidley/WF-100K_schematic.zip
Ok, this has me really irritated now.
I ran some frequency response curves from the input of the Foster amp out to various points in the circuit. Up until the input of the power amplifier section, the curves look like they should.
As Barry calculated, the 3rd order high pass is does indeed have a Q of 0.7. In other words, it isn't the source of the bass boost. I've attached the measurements. The 5th curve which is at the speaker output, is way off the top of the graph since their is 40dB more gain in the power amp. The speaker output curve looks exactly like the frequency response curve posted on my website. It has a 6dB/octave slope across the entire bandwidth. That should have been a clue to me that it wasn't from a high Q high pass filter.
Now I'm searching for the shelving network that must be in the power amplifier section. According to the values in the schematic, there is no shelving. It must be some part substitution that isn't on the schematic. More to follow later......
I ran some frequency response curves from the input of the Foster amp out to various points in the circuit. Up until the input of the power amplifier section, the curves look like they should.
As Barry calculated, the 3rd order high pass is does indeed have a Q of 0.7. In other words, it isn't the source of the bass boost. I've attached the measurements. The 5th curve which is at the speaker output, is way off the top of the graph since their is 40dB more gain in the power amp. The speaker output curve looks exactly like the frequency response curve posted on my website. It has a 6dB/octave slope across the entire bandwidth. That should have been a clue to me that it wasn't from a high Q high pass filter.
Now I'm searching for the shelving network that must be in the power amplifier section. According to the values in the schematic, there is no shelving. It must be some part substitution that isn't on the schematic. More to follow later......
Ok, mystery solved, in a good way. This version of the amplifier has a feature that I wasn't aware of. Foster included a spectral tilt control, but they put it in the power amp section.
There is a variable shelving circuit on the power amp PCB. This is a shelf circuit where the slope is adjustable, sort of. In reality, the corner frequency of the shelf is being moved up and down as the adjustment is made. This moves the the transition area of the shelf into different places in the passband. See the first attachment. This is an unmodified amplifier with VR603 turned full CCW, centered and full CW. As shipped the adjustment is near centered I believe.
At full CCW, the slope is about 1.2dB/octave. Very low. At anything beyond centered, the slope is about 6dB/octave. Note that as the pot is turned CW, the gain of the amp drops a lot because you are sliding the corner frequency of the shelving circuit down lower.
If you want to completely disable this circuit, put a 10k resistor across R662 on the power amplifier PCB. The best place to do this is on the solder side of the PCB. Then you get the response curves shown in the next attachment.
For most applications, I think people can just turn it full CCW and forget about it. If you have a low Q alignment, you can use it to tilt the low end up.
The 3rd order high pass filter is still independently adjustable.
I'll try to update the notes on the website tonight regarding this.
There is a variable shelving circuit on the power amp PCB. This is a shelf circuit where the slope is adjustable, sort of. In reality, the corner frequency of the shelf is being moved up and down as the adjustment is made. This moves the the transition area of the shelf into different places in the passband. See the first attachment. This is an unmodified amplifier with VR603 turned full CCW, centered and full CW. As shipped the adjustment is near centered I believe.
At full CCW, the slope is about 1.2dB/octave. Very low. At anything beyond centered, the slope is about 6dB/octave. Note that as the pot is turned CW, the gain of the amp drops a lot because you are sliding the corner frequency of the shelving circuit down lower.
If you want to completely disable this circuit, put a 10k resistor across R662 on the power amplifier PCB. The best place to do this is on the solder side of the PCB. Then you get the response curves shown in the next attachment.
For most applications, I think people can just turn it full CCW and forget about it. If you have a low Q alignment, you can use it to tilt the low end up.
The 3rd order high pass filter is still independently adjustable.
I'll try to update the notes on the website tonight regarding this.
Jack Hidley said:
I agree Jack, that's shocking. If it wouldn't cost me 10 times the price advertised to get them to NZ I'd so be in there, now my dollar if worth half yours, and postage is generally huge. I'm re-entering a speaker building phase and once I get built the speakers with those seas mids finished (only crossover pending) I'll start on my own which use 4 npt-11-078-1 per channel. I'll have pictures if you're interested, though the mids and tweets are danish but from another vendor.