This observation, accrued over a period of months, may be of help to others looking for a 1" compression driver. The post is meant for those who have the ability to measure and correct the response themselves: no curves or designs are given.
I bought the ND1060 with the XT1086 horn in the hope they would match well, but DE250s that I already had turn out to be a much better solution (may have claimed they are good, but there are few comparisons).
Various RTA measurements were made on an XT1086 horn. The source was from a 0.5 ohm, purely resistive.
The horn was stuffed with absorbing material to reduce cavity resonances.**
The ND1060 was a good fit to the horn with minimal throat discontinuity. The DE250 required more filler to avoid a small but worrying gap.
RTA resolution ranged from 1/6h to 1/48th octave at various locations. (I'm not going to post graphs as they would be non-standard and easily misleading.)
The contrast in the smoothness of response between these drivers was surprisingly large (and repeatable). The ND1060 was tried first and showed multiple significant resonant features: a mixture of relatively high-Q resonances and comb/notch structures. The big surprize I had later on was that DE250 was very much smoother with only a few low-Q features, one not very pronounced notch feature and no sign of combing. (The drivers were exchanged a couple of times as a check.)
Equalising to about +/- 1 dB at 1/12th octave resolution from 1.5 kHz to 15KHz is relatively simple with the DE250 but far beyond my ability with the ND1060.
The measurements fit with listening experience - the ND1060 sounded unnatural on vocal sibilants no matter what EQ I tried (over a substantial time, i.e. months, while the DE250s sounded quite good on the first attempt, with just 3 biquadratic stages plus overall high pass and compensation for HF droop (CD horn and stuffing).
So I add another voice favouring the DE250.
Ken
** Thanks to Dr Geddes for the inspiration to damp the cavity modes, with a bit of thought it seems possible to do this quite well even with a (gentle) diffraction horn like the XT1086.
I bought the ND1060 with the XT1086 horn in the hope they would match well, but DE250s that I already had turn out to be a much better solution (may have claimed they are good, but there are few comparisons).
Various RTA measurements were made on an XT1086 horn. The source was from a 0.5 ohm, purely resistive.
The horn was stuffed with absorbing material to reduce cavity resonances.**
The ND1060 was a good fit to the horn with minimal throat discontinuity. The DE250 required more filler to avoid a small but worrying gap.
RTA resolution ranged from 1/6h to 1/48th octave at various locations. (I'm not going to post graphs as they would be non-standard and easily misleading.)
The contrast in the smoothness of response between these drivers was surprisingly large (and repeatable). The ND1060 was tried first and showed multiple significant resonant features: a mixture of relatively high-Q resonances and comb/notch structures. The big surprize I had later on was that DE250 was very much smoother with only a few low-Q features, one not very pronounced notch feature and no sign of combing. (The drivers were exchanged a couple of times as a check.)
Equalising to about +/- 1 dB at 1/12th octave resolution from 1.5 kHz to 15KHz is relatively simple with the DE250 but far beyond my ability with the ND1060.
The measurements fit with listening experience - the ND1060 sounded unnatural on vocal sibilants no matter what EQ I tried (over a substantial time, i.e. months, while the DE250s sounded quite good on the first attempt, with just 3 biquadratic stages plus overall high pass and compensation for HF droop (CD horn and stuffing).
So I add another voice favouring the DE250.
Ken
** Thanks to Dr Geddes for the inspiration to damp the cavity modes, with a bit of thought it seems possible to do this quite well even with a (gentle) diffraction horn like the XT1086.
"I am going to use the XT1086 in the sides and rears of my HT"
Interesting. Are they not good enough for mains use?
CD is nice in having even power response, but what about the directivity?
I have contemplated CD, or however close an 8" B&C coaxial approaches it, and aiming them upward for diffusion.
What freq is your XO?
Interesting. Are they not good enough for mains use?
CD is nice in having even power response, but what about the directivity?
I have contemplated CD, or however close an 8" B&C coaxial approaches it, and aiming them upward for diffusion.
What freq is your XO?
Stuffing
Brett,
after concluding that it would be hard to get the right foam in the correct shape I tried some old loudspeaker wadding. It was unused because I did not think it very good for its intended purpose, due to it not absorbing as much as other wadding. I've had it for years and am not sure what it is (dark colour, less dense than normal wadding) .
I estimated that a few dB loss would be helpful but still easy to EQ, and kept in mind where the reflections would be most strong (at the driver, at the mouth and at the abrupt change in slope) when positioning the wadding.
I started by putting some wadding into the throat section (in the first ~5cm) until I could see just a little loss at high frequency, then used the density of that as a guide to loosely filling the rest of the volume with wadding, finally with a sheet of it right over the mouth, extending just beyond the edge.
When I started I thought I'd have too much loss, but this does not seem to be the case. Also it was easy to match a pair, the pieces used were paired and the horns are almost filled so it all stays in place.
Actually measuring the benefit is not easy and I'm mainly just convincing myself that a few dB of loss must help damp any modes there are, without significant reduction of maximum output (after digital EQ).
Might be worth a go with any wadding you have around, particularly when you can measure the loss - you can't really go far wrong there.
Oh, and of course you have some special foam to compare with to know how much loss to add ...
Ken
Brett said:
Brett,
after concluding that it would be hard to get the right foam in the correct shape I tried some old loudspeaker wadding. It was unused because I did not think it very good for its intended purpose, due to it not absorbing as much as other wadding. I've had it for years and am not sure what it is (dark colour, less dense than normal wadding) .
I estimated that a few dB loss would be helpful but still easy to EQ, and kept in mind where the reflections would be most strong (at the driver, at the mouth and at the abrupt change in slope) when positioning the wadding.
I started by putting some wadding into the throat section (in the first ~5cm) until I could see just a little loss at high frequency, then used the density of that as a guide to loosely filling the rest of the volume with wadding, finally with a sheet of it right over the mouth, extending just beyond the edge.
When I started I thought I'd have too much loss, but this does not seem to be the case. Also it was easy to match a pair, the pieces used were paired and the horns are almost filled so it all stays in place.
Actually measuring the benefit is not easy and I'm mainly just convincing myself that a few dB of loss must help damp any modes there are, without significant reduction of maximum output (after digital EQ).
Might be worth a go with any wadding you have around, particularly when you can measure the loss - you can't really go far wrong there.
Oh, and of course you have some special foam to compare with to know how much loss to add ...
Ken
I haven't fitted my XT1086s yet (with DE250), but I was concerned about how much they rang when tapped.
I have covered the ribbed surface with children's non-setting modelling clay (called plasticine or "fun clay" in Aus).
There's over a kilo per waveguide there now, and they ain't ringing no more.
David
I have covered the ribbed surface with children's non-setting modelling clay (called plasticine or "fun clay" in Aus).
There's over a kilo per waveguide there now, and they ain't ringing no more.
David
Hi Kris,
Unfortunately I won't have them running for a while (quite a complicated speaker I'm building), but I'll publish specs when they're done.
I noted too that the discontinuities at the throat will need to be smoothed.
Then there is the challenge of getting the combination of open-cell foam, and whatever else, into the flare!!
I'll be using a DEQX HDP3, so I know from experience that I will get a good frequency response, but better still if "mechanical" aspects can be optimised.
David
Unfortunately I won't have them running for a while (quite a complicated speaker I'm building), but I'll publish specs when they're done.
I noted too that the discontinuities at the throat will need to be smoothed.
Then there is the challenge of getting the combination of open-cell foam, and whatever else, into the flare!!
I'll be using a DEQX HDP3, so I know from experience that I will get a good frequency response, but better still if "mechanical" aspects can be optimised.
David
I think you'll find that mounting the driver to the horn and mounting the horn to a baffle already damps the ringing of the horn a great deal. Of course a little extra external damping does no harm: good idea.
The DE250 gives a reasonably smooth curve, as I said above, but EQ is necessary - so you need a means to measure on your own.
I won't provide curves measured at a single point because moving the microphone makes a significant difference (and my measurements are made with stuffing in the horn that is hard to match over a pair, never mind specify for someone else). Look around for DE250 curves and you'll get a qualitative impression the main features. The horn sets a lower limit around 1.2 to 1.8kHz depending on your taste (needs measurements at many angles to see how to integrate with your next driver down).
I can easily get +/-1 dB (mostly much better), on-axis at 0.6m, from 1.5kHz to 10 kHz (or to 15 kHz with more effort). You can easily imagine what that plot looks like.
The main point I made above was that with the DE250 there was no sign of high-Q notches that were seen with the 18sound driver (would not show up clearly on 1/6th octave, but were really bad looking on 1/24th). I gave them to someone who might be able to use them for low quality PA. The lesson I learned was not to be swayed by "features" and very unrevealing data sheets. I'm guessing the differences come from an over-sized compression chamber and the radial phase plug on the 18sound (but that is mostly conjecture, perhaps best to disregard).
It would be great if someone would be able to check the other drivers of good reputation (Beyma CP380, BMS...) for notches or other high-Q irregularities that only show up on high resolution spectra. Meantime I'm content with the B&C.
Bigger horns would allow it to work down a bit lower, I should do that sometime, but there are so few available.
Ken
The DE250 gives a reasonably smooth curve, as I said above, but EQ is necessary - so you need a means to measure on your own.
I won't provide curves measured at a single point because moving the microphone makes a significant difference (and my measurements are made with stuffing in the horn that is hard to match over a pair, never mind specify for someone else). Look around for DE250 curves and you'll get a qualitative impression the main features. The horn sets a lower limit around 1.2 to 1.8kHz depending on your taste (needs measurements at many angles to see how to integrate with your next driver down).
I can easily get +/-1 dB (mostly much better), on-axis at 0.6m, from 1.5kHz to 10 kHz (or to 15 kHz with more effort). You can easily imagine what that plot looks like.
The main point I made above was that with the DE250 there was no sign of high-Q notches that were seen with the 18sound driver (would not show up clearly on 1/6th octave, but were really bad looking on 1/24th). I gave them to someone who might be able to use them for low quality PA. The lesson I learned was not to be swayed by "features" and very unrevealing data sheets. I'm guessing the differences come from an over-sized compression chamber and the radial phase plug on the 18sound (but that is mostly conjecture, perhaps best to disregard).
It would be great if someone would be able to check the other drivers of good reputation (Beyma CP380, BMS...) for notches or other high-Q irregularities that only show up on high resolution spectra. Meantime I'm content with the B&C.
Bigger horns would allow it to work down a bit lower, I should do that sometime, but there are so few available.
Ken
D OB G & kstrain,
Thank you for your input.
I was hoping to cross (passive x-over) at around 2500 or even 3000 Hz well away from resonance of the B&C. I am looking for a driver + waveguide requiring no narrow notch filtering.
I have my own measuring gear so that part is taken care of, but finding a driver for the xt1086 is almost like firing in the dark.
Perhaps there is a better driver for the 2500-14000 range than the BC250...
Kris
Thank you for your input.
I was hoping to cross (passive x-over) at around 2500 or even 3000 Hz well away from resonance of the B&C. I am looking for a driver + waveguide requiring no narrow notch filtering.
I have my own measuring gear so that part is taken care of, but finding a driver for the xt1086 is almost like firing in the dark.
Perhaps there is a better driver for the 2500-14000 range than the BC250...
Kris
I don't understand the desire to keep "well away from resonance of the B&C". A compression driver on a horn usually has one system resonance right in the middle of the passband. CD resonances are very well damped as they have to be to work on a waveguide. Remember that a CD on a waveguide basically operates in the resistance controlled region of the diaphragm, not the mass controlled region as a direct radiator does.
hasselbaink said:
A lot been mention in this forum for DE250. DDS ENG90 is another reasonable cheap horn that worth trying with DE250. Have a read at post #41 by AJ.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=112641&perpage=25&pagenumber=2
454Casull said:
From John_E_Janowitz (Acoustic Elegance) in a thread at HTGuide called "This worked out well ( BMS tweeters / DDS Waveguides )", as well as around here and other forums. I've mentioned it here several times too while referring to that thread because I've been interested in that waveguide forever now, but was very disappointed when I came across these problems. So, I've never tried them myself, but I still really want to..
On the other hand, from that thread alone, I'd always assumed it was only a temporary problem with bolt-on versions of the waveguide coming from US Speaker when DDS had first been sold and/or changed location (or whatever happened, I don't know.. I just know that they're here in Texas now).
However, John has made replies to posts of mine on more than one occasion, as well as more than one forum (here and maybe his own forum, don't remember exactly).. that the DDS waveguides do have severe problems, and that he has witnessed other people receiving the same quality issues more recently, and basically to avoid them..
But at the same time, I've seen so many other people very happy with them, with nothing but good things to say, which certainly wouldn't be the case if they all looked like the pics posted by John Janowitz in that thread at HTGuide. On top of that, doesn't that AudioKinesis guy at www.audiokinesis.com use them in his speakers? (sorry, forgot his name.. I think it might be Duke?) Where's he at? I'd love to hear from him about this.
There's obviously quite a few people out there who have been extremely pleased with these waveguides, to say the least. I'm just starting to find it odd that the only negative experiences I've seen regarding these waveguides are from John E. Janowitz from AE Speakers, and those other 1 or 2 people (or however many there were) in that same thread at HTGuide.. yet John claims they still have quality issues. I'm not suggesting that he's lying or anything like that, just that it doesn't make any sense to me.
I'd really like to get to the bottom of this somehow. If there was some way to gather everyone together who has purchased this waveguide to comment on this matter, including those with negative experiences of any kind, the problem would be solved and I'd be on the phone with Assistance Audio first thing tomorrow. I know that'll probably never happen though, so even a word from another person or two that's purchased this waveguide recently would be great.
So.. where's this AudioKinesis guy?
I've searched and searched on these DDS waveguides over past months and found anything and everything there is to find about them, including photos. Still, I can't find any negative comments except that one thread at HTGuide and from John, and people spreading that information (mainly me).
I'm determined to get to the bottom of this, lol.. hopefully they're fine now and I can end up as happy as some other people have.
EDIT: forgot these..
http://www.htguide.com/forum/showpost.php4?p=403123&postcount=62
http://www.htguide.com/forum/showthread.php4?t=19368&page=2&pp=35
I am using the DDS WG with the DE250. No worries. This combo even sounds pretty flat with no EQ.
However, I did have to do a little grinding to get the driver to fit. Nothing major though. The quality issues on the DDS waveguides are in the bolt-on mouting plate area only. The threaded versions are fine. I just couldn't find a threaded driver that I liked and there is NO WAY I would use those screwy adapters.
I may upgrade down the road to a larger WG (12 or 15) from Earl. I don't think the DDS 10 inch can really support a sub 1K crossover point.
However, I did have to do a little grinding to get the driver to fit. Nothing major though. The quality issues on the DDS waveguides are in the bolt-on mouting plate area only. The threaded versions are fine. I just couldn't find a threaded driver that I liked and there is NO WAY I would use those screwy adapters.
I may upgrade down the road to a larger WG (12 or 15) from Earl. I don't think the DDS 10 inch can really support a sub 1K crossover point.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.