I thought I would start a new tread on this one.
I liked reading about your project. I found it interesting because I recently just completed a design very similar.
The effort was to create an unbeatable mini-monitor.
The design was for Creative Sound Solutions located in Abbotsford, Canada.
It was also to use the G2 ribbon. This is the same tweeter as the ESg2. Made by the same company. The ES is Andre's (E-Speakers) designation.
CSS (Creative Sound Solutions) is the Canadian distributor for those tweeters.
One request I made when I started on the design was to have the manufacturer redesign a smaller round face plate. Hobbyist generally do not like countersinking odd rectangular shapes.
So all the ribbons he is bringing in now have the smaller round face plate and the ARG2 designation. http://www.creativesound.ca/details.php?model=ARG2
Interestingly enough CSS is also a Jordan distributor too.
The new mini-monitor I designed will also be made available here in the states through GR Research, but I do not have the rights to sell the tweeters separately. I can only sell them in a finished product, or in these case a loudspeaker kit.
Since the ARG2 was designed by the same guy that designed the Ravens I was not surprised to find them very similar. But, I must say I do like the ARG2 much better.
The shielding for one doesn't hurt.
Secondly, the higher 8 ohm impedance is much preferred.
I also believe the ARG2 to handle a little more power than the Raven R-1.
Still neither of them will handle any low frequency information.
I get the feeling that the large L-pad you are using on it and the small cap value is the only thing saving it from sudden death.
Any other first order ideas with that tweeter I would strongly advise against.
Another thing I found was that the higher you cross them the closer the driver need to be together. This is of coarse true with any driver. So I was surprised at the distance you have them apart. Then again the 6db per octave slope really helps.
I tend to think though, that a first order network crossed that high would give you a good sized dip in the response about one octave about the crossover point.
It could be that the physical driver alignment adjustment you have made has avoided much of that, but it may be there pretty heavy in the vertical off axis. Have you tried measuring in incremental vertical heights?
Then again if the on axis is smooth enough and the room interactions are not showing the signs of problems in the vertical off axis then it may be worth not adding more components in the signal path.
I like the minimalist approach myself whenever possible.
For this recent design that I did I was able to get a higher order series network to provide the design objectives I had of minimal shift in phase acoustically and electrically and a flat impedance while still maintaining a near +/-1db response across the board.
It was also a good idea for you to put the tweeter in its own enclosure. This type of tweeter is very dependant on the baffle size. Added surface reflections will cause problems in the response.
My design maintained only a 7.5" wide baffle with a 1/2" radius on the edges, and I managed to avoid any real problems.
Putting it into a baffle as wide as the TL type typically used with the Jordan to minimize its baffle step loss would have cause a lot of problems it the tweeter response.
I really like the sound of that ribbon tweeter.
I can imagine your design to sound really good as well.
Nice job.
Danny Richie
I liked reading about your project. I found it interesting because I recently just completed a design very similar.
The effort was to create an unbeatable mini-monitor.
The design was for Creative Sound Solutions located in Abbotsford, Canada.
It was also to use the G2 ribbon. This is the same tweeter as the ESg2. Made by the same company. The ES is Andre's (E-Speakers) designation.
CSS (Creative Sound Solutions) is the Canadian distributor for those tweeters.
One request I made when I started on the design was to have the manufacturer redesign a smaller round face plate. Hobbyist generally do not like countersinking odd rectangular shapes.
So all the ribbons he is bringing in now have the smaller round face plate and the ARG2 designation. http://www.creativesound.ca/details.php?model=ARG2
Interestingly enough CSS is also a Jordan distributor too.
The new mini-monitor I designed will also be made available here in the states through GR Research, but I do not have the rights to sell the tweeters separately. I can only sell them in a finished product, or in these case a loudspeaker kit.
Since the ARG2 was designed by the same guy that designed the Ravens I was not surprised to find them very similar. But, I must say I do like the ARG2 much better.
The shielding for one doesn't hurt.
Secondly, the higher 8 ohm impedance is much preferred.
I also believe the ARG2 to handle a little more power than the Raven R-1.
Still neither of them will handle any low frequency information.
I get the feeling that the large L-pad you are using on it and the small cap value is the only thing saving it from sudden death.
Any other first order ideas with that tweeter I would strongly advise against.
Another thing I found was that the higher you cross them the closer the driver need to be together. This is of coarse true with any driver. So I was surprised at the distance you have them apart. Then again the 6db per octave slope really helps.
I tend to think though, that a first order network crossed that high would give you a good sized dip in the response about one octave about the crossover point.
It could be that the physical driver alignment adjustment you have made has avoided much of that, but it may be there pretty heavy in the vertical off axis. Have you tried measuring in incremental vertical heights?
Then again if the on axis is smooth enough and the room interactions are not showing the signs of problems in the vertical off axis then it may be worth not adding more components in the signal path.
I like the minimalist approach myself whenever possible.
For this recent design that I did I was able to get a higher order series network to provide the design objectives I had of minimal shift in phase acoustically and electrically and a flat impedance while still maintaining a near +/-1db response across the board.
It was also a good idea for you to put the tweeter in its own enclosure. This type of tweeter is very dependant on the baffle size. Added surface reflections will cause problems in the response.
My design maintained only a 7.5" wide baffle with a 1/2" radius on the edges, and I managed to avoid any real problems.
Putting it into a baffle as wide as the TL type typically used with the Jordan to minimize its baffle step loss would have cause a lot of problems it the tweeter response.
I really like the sound of that ribbon tweeter.
I can imagine your design to sound really good as well.
Nice job.
Danny Richie
Danny,
Nice to hear from yet another person, especially from one with such experience and insight as you have, working with the same not so much explored driver. Can I ask what are you using for mid bass - PHL perhaps, or your own GR drivers?
I also concur that G2 is a wonderfully sounding tweeter. It has all the detail one could possibly have, yet it newer sounds harsh or biting. The only downside is it's very narrow vertical response window.
How low you consider is to low for crossing G2? I have tried it as low as 1700Hz with 4th order acoustical slope, padded down about 8dB and I must say it sounded sweetest. Of course without padding the tweeter level I wouldn’t tried this, also I didn't crank the volume all way up with this crossover though. But couple of times when I did some measurements with MLS and accidentally didn’t turn the level down, the driver got really big unpadded and unfiltered signal through it. For a second I thought I blow the ribbon element but it just wobbled heavily and nothing disastrous happened. In a base line I would say it could handle much more than it is expected from ribbons usually. Also see my post on this thread http://www.diyaudio.com/forums/showthread.php?threadid=7264 where I point on Aurum Cantus kit design where I believe they are using just one 6,6uF (actually two 3,3UF caps parallel) cap and L-pad for their G2 tweeter.
I agree with you, that this type of tweeter is very dependent on baffle size. When it is installed into wide baffle, because of the added surface reflections the response changes but I wouldn’t call this strictly as a problem. Of course, minimal approach with just one series cap for crossover is not working out so nicely anymore but by adding a shunt coil will shape the response into nice 3rd or 4th order curve, plus you will have the possibility to make the cutoff much lower.
What comes to Jordan JX92S, as I said in original thread I have found out, that it doesn’t like any crossover very much at all, except baffle step correction (large coil and resistor in parallel). I have tried all kinds of crossover orders and cutoff points for JX92S and I must say that crossing it is the waste of the otherwise excellent driver - you'll loose its spectacular midrange magic it has become so well renowned. In case one needs to cross it, he may replace it with any decent mid bass as well.
At the moment I have come up with crossover topology I can live with. It sounds as single Jordan but with extended high frequency attributes. I wont disclose the actual crossover yet because there is still some more tuning to do and it wouldn’t help much anyhow if the baffle size is not kept to original. But the preferences I have described above should give some hints of the topology.
Still, as usual, there is not a single perfect crossover only - there are always other, possible even better solutions yet to find so I am all ears about yours and Roberts findings on the designs with these drivers.
cheers,
Argo
Nice to hear from yet another person, especially from one with such experience and insight as you have, working with the same not so much explored driver. Can I ask what are you using for mid bass - PHL perhaps, or your own GR drivers?
I also concur that G2 is a wonderfully sounding tweeter. It has all the detail one could possibly have, yet it newer sounds harsh or biting. The only downside is it's very narrow vertical response window.
How low you consider is to low for crossing G2? I have tried it as low as 1700Hz with 4th order acoustical slope, padded down about 8dB and I must say it sounded sweetest. Of course without padding the tweeter level I wouldn’t tried this, also I didn't crank the volume all way up with this crossover though. But couple of times when I did some measurements with MLS and accidentally didn’t turn the level down, the driver got really big unpadded and unfiltered signal through it. For a second I thought I blow the ribbon element but it just wobbled heavily and nothing disastrous happened. In a base line I would say it could handle much more than it is expected from ribbons usually. Also see my post on this thread http://www.diyaudio.com/forums/showthread.php?threadid=7264 where I point on Aurum Cantus kit design where I believe they are using just one 6,6uF (actually two 3,3UF caps parallel) cap and L-pad for their G2 tweeter.
I agree with you, that this type of tweeter is very dependent on baffle size. When it is installed into wide baffle, because of the added surface reflections the response changes but I wouldn’t call this strictly as a problem. Of course, minimal approach with just one series cap for crossover is not working out so nicely anymore but by adding a shunt coil will shape the response into nice 3rd or 4th order curve, plus you will have the possibility to make the cutoff much lower.
What comes to Jordan JX92S, as I said in original thread I have found out, that it doesn’t like any crossover very much at all, except baffle step correction (large coil and resistor in parallel). I have tried all kinds of crossover orders and cutoff points for JX92S and I must say that crossing it is the waste of the otherwise excellent driver - you'll loose its spectacular midrange magic it has become so well renowned. In case one needs to cross it, he may replace it with any decent mid bass as well.
At the moment I have come up with crossover topology I can live with. It sounds as single Jordan but with extended high frequency attributes. I wont disclose the actual crossover yet because there is still some more tuning to do and it wouldn’t help much anyhow if the baffle size is not kept to original. But the preferences I have described above should give some hints of the topology.
Still, as usual, there is not a single perfect crossover only - there are always other, possible even better solutions yet to find so I am all ears about yours and Roberts findings on the designs with these drivers.
cheers,
Argo
ribbons and stuff
I had a 5" driver made by Eton. It sounds very good!
The lowest I can see crossing it without forcing the response curve into submission with a shunt coil that will lower impedance too much, and allow it to handle good power without fear of a popped ribbon, I would say no lower than 3kHz.
I agree. An independent test from one of the worlds foremost ribbon designs indicated it handled more power then the Raven R-1. While simultaneously sending them both the same signal the Raven popped its ribbon.
Any low frequency information that even starts these to moving is a bad idea.
In regards to what you mentioned about crossing these with the Jordan: I would likely try a higher order series network with it and the driver you are using. This would give you the driver controlled needed and still keep the component count low, plus keep phase shift to a minimum.
The Jordan's rising response and slight break up in the higher ranges needs more attenuation (steaper network) to get the response down low enough not to cause problems with the response in the range the tweeter needs to be covering.
Let's face it the Jordan does not have much natural roll off.
The higher order series network can also compensate for the baffle step loss by using the larger value inductor as the second inductor. I would guess 1.5 to 2.mH
Also adding an impedance compensation or even just an additional shunt cap at the end of the circuit would likely be needed as well. This will actually make the network a third order series network on the woofer. Guessing 3.9 to 4.7uF on the last cap.
This configuration might get you where your going with it.
I had a 5" driver made by Eton. It sounds very good!
The lowest I can see crossing it without forcing the response curve into submission with a shunt coil that will lower impedance too much, and allow it to handle good power without fear of a popped ribbon, I would say no lower than 3kHz.
I agree. An independent test from one of the worlds foremost ribbon designs indicated it handled more power then the Raven R-1. While simultaneously sending them both the same signal the Raven popped its ribbon.
Any low frequency information that even starts these to moving is a bad idea.
In regards to what you mentioned about crossing these with the Jordan: I would likely try a higher order series network with it and the driver you are using. This would give you the driver controlled needed and still keep the component count low, plus keep phase shift to a minimum.
The Jordan's rising response and slight break up in the higher ranges needs more attenuation (steaper network) to get the response down low enough not to cause problems with the response in the range the tweeter needs to be covering.
Let's face it the Jordan does not have much natural roll off.
The higher order series network can also compensate for the baffle step loss by using the larger value inductor as the second inductor. I would guess 1.5 to 2.mH
Also adding an impedance compensation or even just an additional shunt cap at the end of the circuit would likely be needed as well. This will actually make the network a third order series network on the woofer. Guessing 3.9 to 4.7uF on the last cap.
This configuration might get you where your going with it.
Been there done that.
I used them in a design about three years ago.
I have built some nicer combo's since then, but at the time I thought it to be very good.
Here is a link to some info on it: http://www.gr-research.com/custom/r2phl.htm
I built several different test boxes. The one that works the best with these tweeters allows no surface reflections to either side of the tweeter.
I assisted a client with one deign using the R-2's that had a polymer and concrete front baffle that was a pored mold system that was swept back away from the tweeter. It had a nice response free of the surface reflections. It used Focal Audiom 7NV2 woofers in an MTM design with a 12" Scaaning lower woofer.
The only thing I didn't like about both of those models was the limited vertical dispersion of the R-2 tweeters.
I used them in a design about three years ago.
I have built some nicer combo's since then, but at the time I thought it to be very good.
Here is a link to some info on it: http://www.gr-research.com/custom/r2phl.htm
I built several different test boxes. The one that works the best with these tweeters allows no surface reflections to either side of the tweeter.
I assisted a client with one deign using the R-2's that had a polymer and concrete front baffle that was a pored mold system that was swept back away from the tweeter. It had a nice response free of the surface reflections. It used Focal Audiom 7NV2 woofers in an MTM design with a 12" Scaaning lower woofer.
The only thing I didn't like about both of those models was the limited vertical dispersion of the R-2 tweeters.
Danny
Hi, I guess I wasn't clear. I had seen this on the GR website and was wondering if you thought using the G2 tweeter instead of the Raven would be better? I like the PHL drivers because the have a good reputation, are available in 16 ohm so the finished speak is 8 ohm, and they are quite efficient, BUT I have never heard them. Do you prefer the sound of the Focal mid/woofs?
Hi, I guess I wasn't clear. I had seen this on the GR website and was wondering if you thought using the G2 tweeter instead of the Raven would be better? I like the PHL drivers because the have a good reputation, are available in 16 ohm so the finished speak is 8 ohm, and they are quite efficient, BUT I have never heard them. Do you prefer the sound of the Focal mid/woofs?
Okay
I get what you are saying now.
The G2 certainly has some advantages over the Ravens, but I am not eager to try it with any of the PHL's at present.
I do like the sound of the PHL's over the Focal units I have tested, but the one thing I hate about the PHL's is the odd shaped frame that does not allow easy countersinking.
If they expect any hobbyist to use these they really need to think that through a little more.
Currently I wouldn't swap any of them for the new 5" Eton.
I get what you are saying now.
The G2 certainly has some advantages over the Ravens, but I am not eager to try it with any of the PHL's at present.
I do like the sound of the PHL's over the Focal units I have tested, but the one thing I hate about the PHL's is the odd shaped frame that does not allow easy countersinking.
If they expect any hobbyist to use these they really need to think that through a little more.
Currently I wouldn't swap any of them for the new 5" Eton.
Have we overlooked the Eton?
Those Etons are cool looking (in the photo and the graph)
additionally, The 7" model 372/32LH seems flat flat flat (OK, one bump)
It has a cylider sticking out the middle which they call a heat pipe, but I'll bet it also works as a phase plug-they are flat out to about 8k (except for that one bump)
Also 90 dB I'm sure with the usual narrowing dispersion, but maybe that works with the narrow ribbon dispersion. We sure like that efficiency around here!
Those Etons are cool looking (in the photo and the graph)
additionally, The 7" model 372/32LH seems flat flat flat (OK, one bump)
It has a cylider sticking out the middle which they call a heat pipe, but I'll bet it also works as a phase plug-they are flat out to about 8k (except for that one bump)
Also 90 dB I'm sure with the usual narrowing dispersion, but maybe that works with the narrow ribbon dispersion. We sure like that efficiency around here!
I like those round G2 plates- they would be a lot easier to work with. You know your clients!
One think I reallly would like for the G2's is a frequency response curve, both on and off axis. They aren't shown on any site that I can see, but would you be willing to share any that you have made? I am very curious as to the dispersion of 'em
One think I reallly would like for the G2's is a frequency response curve, both on and off axis. They aren't shown on any site that I can see, but would you be willing to share any that you have made? I am very curious as to the dispersion of 'em
G2 tweeter
The actual responses off axis will vary some with baffle size.
I do not have any measurements on hand of just the tweeter itself. I just have the measurements made with the Criterion.
I will be working this coming week to get all the Criterion info up on our site. It will have complete measurements posted as well.
That might give you a pretty good idea of the tweeters response.
CSS already has some info on the new model up on their site: http://www.creativesound.ca/details.php?model=CRITERION
But no measrements posted yet. They will have some up soon too.
The actual responses off axis will vary some with baffle size.
I do not have any measurements on hand of just the tweeter itself. I just have the measurements made with the Criterion.
I will be working this coming week to get all the Criterion info up on our site. It will have complete measurements posted as well.
That might give you a pretty good idea of the tweeters response.
CSS already has some info on the new model up on their site: http://www.creativesound.ca/details.php?model=CRITERION
But no measrements posted yet. They will have some up soon too.
measurements posted...
I finally got the info up on our web site on the new Criterion.
http://www.gr-research.com/lucidity/criterion.htm
Off axis measurements are listed at the bottom of the page.
Variac, I hope this tells you what you needed to know about the off axis responses of that tweeter.
I finally got the info up on our web site on the new Criterion.
http://www.gr-research.com/lucidity/criterion.htm
Off axis measurements are listed at the bottom of the page.
Variac, I hope this tells you what you needed to know about the off axis responses of that tweeter.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.