Patrick Bateman said:
John
That data is not taken to show true HF response simply because I don't give above 10 kHz much importance. SO what you are seeing is more the result of the anti-aliasing filter than the drivers response.
In practice I have seen differences in drivers above 10 kHz but I haven't seen anything that stands out a "clearly better" They all seem to loose control way up there, but in different ways. I would simply say that I have not seen a significant difference in drivers in this regard. The specific horn design is also a big factor in how this HF response looks. I've tested several drivers all on my same waveguide and there is nothing that would lead me to belive that any one driver is that different from another.
gedlee said:
I agree, the difference between most compression drivers is mostly the size of the throat, the design of the phase plug, and the design of the motor. Which means that if we're using a 1" compression driver, there isn't a tremendous amount of difference between B&C, Radian, Selenium, etc.
But the BMS is unconventional; it doesn't use a dome. Here's a pic of what it looks like if you take it apart:
http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=11051&stc=1&d=1132259749
A few years back Vifa created a driver that became an instant classic, the XT25.
An externally hosted image should be here but it was not working when we last tested it.
The BMS and the Vifa sound a lot alike; and both use a ring radiator to extend the respond out past 20khz. Here's the response of the Vifa:
An externally hosted image should be here but it was not working when we last tested it.
As far as I know, the BMS 4540nd and the JBL 2407H are the only compression drivers with significant output beyond 18khz. The only alternative is to use a ribbon, and as you've discussed, that won't work on a waveguide.
Its not commonly used anymore because of its uncommon price tag, but the TAD2001 has a very high and smooth response for a 1". I haven't personally measured better on my horn but I also haven't compared it to that many others (Radian, B&C, etc) It also has a 3" or so long smooth tapered throat between the phase plug and the exit. I always wondered how much that extra throat length helped with its response.
nickmckinney said:
Yesterday I was checking out your old Lambda site, since it has a lot of lost information on the Unity.
Using the "wayback machine", I discovered that you sold a set to someone who lives about 30 minutes from me.
I've emailed the owner, and may get to finally hear a pair!
Very exciting - the only "real" unities I've ever heard were in a Las Vegas night club.
nickmckinney said:
The Summas originally used the TAD driver, but after some "shootouts" against the B&Cs, I switched. After several other trys with different drivers, I have consluded that the driver is a minimal part of the sound. Its the waveguide and the foam that makes the difference. There might be a subtle audible difference in drivers, but it pales compaired to the difference that the waveguide and foam makes. And these differences tend to be personal. Everyone will prefer any driver on a good waveguide with foam, but the differences between drivers on this same setup will be almost random with no strong preference.
gedlee said:
My Summas have a plug of foam that's about ten inches deep. If I build a Unity with a 30" mouth, it will require a foam plug that's about TWENTY SIX inches deep.
One thing I noticed when putting the foam into a tractrix horn was that the improvement was more audible than it was with a oblate spheroidal waveguide. With the OS waveguide the improvement was noticeable; with the tractrix it was substantial.
One thing I'm curious to find out is if the 30" waveguide will see a more noticeable improvement with the foam plug. In other words, is there a point of diminishing returns, where too much foam is making no further improvements? The 30" waveguide will require a foam plug about four times as big, and the efficiency will suffer as a result.
It should be interesting to find out, to say the least.
I just don't have the confidence in the importance of signals above 10 kHz that you guys have. Our hearing IS NOT naturally stimulated by sounds in nature that high (air absorption above 10 kHz is very high) and as such it is logical to expect that over time our hearing would also not become very acute. All data says that this IS the case. SUre we can measure this high AND we will see tones this high on near field mics, but does any of this have anything to do with sound quality. I just don't see it.
We did a test at my shop one time with single tone sweeps to see who can hear how high on a good flat response system. At the time 15Khz was my limit, some of the older folks were losing out at 10Khz. Nearly 10 years later and I wonder how I would fair today on this same test. I have been known to run quite a few air tools............
How high is to high?
I worked with a gentleman that could consitently tell when an ultrasonic cleaner was working. It vibrated at 40 khz! His lab was across the hallway about 40 ft from the source. It drove him nuts.
This fellow worked as an instrumentation calibration technician. He measured the frequency of the cleaner and was shocked at how high it was.
Mark
I worked with a gentleman that could consitently tell when an ultrasonic cleaner was working. It vibrated at 40 khz! His lab was across the hallway about 40 ft from the source. It drove him nuts.
This fellow worked as an instrumentation calibration technician. He measured the frequency of the cleaner and was shocked at how high it was.
Mark
Re: How high is to high?
My experience with ultrasonic cleaners is by no means exhaustive, but all the ones I've encountered produce clearly audible noise and are very irritating. If this model was really quiet (to everyone but this guy) please tell me the manufacturer if you remember, I would love to get one.
mwmkravchenko said:
My experience with ultrasonic cleaners is by no means exhaustive, but all the ones I've encountered produce clearly audible noise and are very irritating. If this model was really quiet (to everyone but this guy) please tell me the manufacturer if you remember, I would love to get one.
gedlee said:
Cmon now Earl, I could hear 15Khz just fine at the time at about the same level as the frequencies just below. I was actually a bit bummed because my hearing dropped like a brick right after whereas my gf could hear out to 18Khz cleanly. Its not like we were ABx testing cables.................
.
Here's another DIY unity.
An externally hosted image should be here but it was not working when we last tested it.
gedlee said:
I put together a little system for my bedroom the other day using a small unity horn I made about 5 years ago. It uses two Eminence Beta 8's and a B&C DE25 in a 40 degree horn. Here are some pictures of the horn:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
The original idea was to get the 8's to transition from the CD smoothly and then play down to a lower frequency than normal for a unity - 100hz or so (or lower - I forget exactly what I designed for after 5 years) so it would only need a sub to go with it, if that (for non-critical listening). Anyway, I never got it to sound good with passive components. This week I dusted it off and set it on top of a 2-way bass cabinet I built 6 years ago with a JBL 15" in a small sealed box and two 10" MCM speakers in a ported box. I used a DSP crossover which greatly simplified things (I've never been that great at passive crossovers - not much practice). The 8" to CD crossover ended up at 850Hz. I got it setup using on-axis measurements in room and then took measurements at roughly every 5 degrees moving horizontally going out to about 30 degrees. That was as far as I could go before I ran out room. My data is kind of nasty due to being in-room, but I don't see anything that's obviously a bad interaction at the crossover (no huge dip). The horn is ~ constant directivity above 2kHz but has a dip outside the wall angle at 10kHz.
Earl, if you are reading this, is a big dip off-axis at the crossover frequency what you'd expect to see if the interaction was bad? Would you expect this immediately off-axis or a farther angle away from on-axis?
John Sheerin said:
The effect at the crossover point can be almost anything from an axial hole to an axial peak to the same off-axis. There are too many variables to determine what will happen in the arbitrary case. The lower the frequency the lower the expected effect. In the BMS dual diaphragm driver the response aberations were significant both on and off axis and these diaphragms were very close together, but the frequency was high - about 6 kHz. At 850 Hz you are not going to see a huge effect, but that is very low for the CD. I read that these crossover point are usually much higher.
PS - no pictures
Yes, I had that for sale on my website (still do actually), but now I've decided to use it. I only have one of my them. It's just running in mono for background music.
Earl, thanks for the reply. The pictures are linked off my website. If they didn't show up for you, try again. They are working for me right now. I think most of the unity-style horns crossover from the CD to the mids in the 1-1.5kHz range. It's hard to get cone drivers mounted closer to the compression driver than the minimum distance required for those crossover points. In any case, thanks for the feedback on interference. At some point I may try taking some measurements in a larger space to get a clearer idea of what's going on, but first I need to get myself a laptop that doesn't feedback internally (the stupid IBM I have right now is worthless for audio measurements). For now it seems okay to me.
Earl, thanks for the reply. The pictures are linked off my website. If they didn't show up for you, try again. They are working for me right now. I think most of the unity-style horns crossover from the CD to the mids in the 1-1.5kHz range. It's hard to get cone drivers mounted closer to the compression driver than the minimum distance required for those crossover points. In any case, thanks for the feedback on interference. At some point I may try taking some measurements in a larger space to get a clearer idea of what's going on, but first I need to get myself a laptop that doesn't feedback internally (the stupid IBM I have right now is worthless for audio measurements). For now it seems okay to me.
"first I need to get myself a laptop that doesn't feedback internally "
Always a potential problem, but I have found that it is usually fixable. One time I spent days trying to fix a problem only to learn that there was this internal feedback - with a delay - which was causing all the problems. Now I always check for that by disconnenting inputs etc. to make sure that things are what I think they are.
Always a potential problem, but I have found that it is usually fixable. One time I spent days trying to fix a problem only to learn that there was this internal feedback - with a delay - which was causing all the problems. Now I always check for that by disconnenting inputs etc. to make sure that things are what I think they are.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.