Mini-Synergy Horn Experiment

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Here is the data for the new configuration. To keep it consistent I eq'd flat to 20° as I did before. I can see some improvement. The curves are a bit smoother and the dip at larger off axis angles at 5khz has been taken care of - however that "excess" energy around 7khz is still present.

1-2 port synergy polars.png

I decided to tape up the ports and see what I get. I should have done this sooner! I've made hundreds of measurements of these waveguides so I thought I knew their response well. Apparently not! Here are some comparisons. Hopefully it's labeled clearly. The only eq is a 4khz 1st order hp filter. There has always been a bit of "bunching" around 7khz but I've never seen the 30° curve rise as dramatically as it is. You can see polars of this wg without mid ports on post #423. Definitely something going on around 7khz but not to this extent. The ports are making whatever is going on there a bit worse but I wouldn't say it's the cause.

0deg.png

20deg.png

30deg.png

Here's the polar response with the holes taped:

taped ports.png

Mid ports open:

open mid ports.png
 
Hi Nate:
Good pictures. I can see that you've pretty much got the holes as close in as possible. I wonder how much further out you could move them without lowering XO frequency (too much) and how much that would improve the polars.
Jack

Wondering the same myself, however the data I'm showing (after your post) with the ports taped suggest it wouldn't get much better. This is really mostly a wg issue. IIRC I estimated the internal pathlength of these 4552 cds to be about 1.4". The center of the ports is 1.35" from the throat which would put the cancellation notch around 1.2khz. In practice there is no definite notch and the mids high corner is 1.3khz. I'd like to get as much bandwidth out of the mids as I can to keep the xo to the cd higher.
 
I agree. I think the wavequide is as good as it needs to be. The first 30 degrees are all highly correlated so after EQ that entire window should be nicely flat. The behavior further off axis is curious but is going to have at most a 2nd order effect on the sound quality at a listening position within the first 30 degrees.

You could place and shape some bondo under the mid cones if you need to squeeze a little more bandwidth out of them.
 
I agree. I think the wavequide is as good as it needs to be. The first 30 degrees are all highly correlated so after EQ that entire window should be nicely flat. The behavior further off axis is curious but is going to have at most a 2nd order effect on the sound quality at a listening position within the first 30 degrees.

I should add that I looked at the vertical response a bit and the 7khz "anomaly" is quite evident there as well. I think in terms of the the overall power response it would be wise to pull that down a bit.
 
I've been having issues with my BMS cds. Things are fine at lower levels but fall apart as I turn up the volume. They've always had a bit of an "edge" at higher spl but I attributed that mostly to my narrow, untreated, and sparsely furnished living room. Lately that "edge" has gotten much worse. When I was doing the synergy conversion to my waveguides I had the 4552's sitting on my workbench in the basement and I think that there is gunk in the gap - I tend to be hard on equipment at times! That, and I really don't think these do well below 1.5khz. A neo brother to the 4550 they are not!

All I've ever used are ring radiator cds so I've been wanting to try a conventional dome anyways. I picked up a couple B&C DE500 hoping that they would have a similar throat angle being a neo motor as well. The are actually wider - about 35° total exit angle compared to the 4552 which is about 24° iirc - so I need to do a bit of work to the throat on my wg. Here's some data with just a 4khz 1st order hp for eq. Here is the DE500:

10-20-17 eoswg de500 raw.png

A bit smoother below 10khz and I think above that will improve when I match my wg to the driver, and can see greater output below 1khz. The "edge" is gone with these new drivers and things are working quite well again. I spent quite a bit of time yesterday listening to my favorite tracks.

Here's the BMS for comparison measured at the same time:

10-20-17 eoswg 4552nd raw polars.png
 
I've been having issues with my BMS cds. Things are fine at lower levels but fall apart as I turn up the volume. A neo brother to the 4550 they are not!
Here's some data with just a 4khz 1st order hp for eq. Here is the DE500:

Here's the BMS for comparison measured at the same time:
Nate,

Looking at the differences between the frequency responses of the two drivers, I don't see anything that would indicate "smoothness" as being an issue. In testing the 4552 and 4550, what was apparent to me is the LF distortion increase at the bottom end of their response came on very strong even with a 24 dB per octave slope. A 4kHz 1st order slope provides almost no excursion protection at all, odd order harmonics from the bottom end will certainly sound "harsh".

I'd be interested to see the difference in distortion between the DE500 and the 4552, I think that will clarify the difference in sound signature between the two- there is "no replacement for displacement", the larger diaphragm area of the DE500 should have a "big" advantage down low.

Cheers,
Art
 
Art, I wasn't meaning that the different freq response was the issue I was hearing. The increasing measured distortion of the BMS down low and as spl increased was what clued me in to a possible mechanical issue. These things always had higher distortion down low as you've pointed out over the years, but lately it has been much worse. I'm not one to look at the distortion as an indicator of quality per se, but it can be a useful tool for finding issues.

Here's some data taken on different days but about the same level. Both have an acoustical lr6 xo at 1.3khz. First is the BMS. What you see for distortion over 1khz is all in the cd. 1.3% @ 1khz. I also did a sweep at 100dB and the harmonics were notably higher. Didn't save that one.

10-22-17 bms distortion.png

This is the B&C. As you say the displacement has a clear advantage and these haven't been abused like my BMS :rolleyes:

10-22-17 de500 distortion.png
 
I'm feeling defensive :) about my BMS4550's. I wonder if you considered them and if not why not. They are about the same price as the DE500 and rated for 800 Hz XO. I'm using them at 1050 Hz and they sound fine and measure THD quite low at normal listening levels. At really high levels I hear some harshness but those levels make me grab for the volume control ASAP.
 
I'm feeling defensive :) about my BMS4550's.

I wouldn't! Doesn't Danley use them in the SH50? BMS claims an 800hz low xo.........but they also claim 1khz for the 4552. Let's not overlook the fact that I woulnd't be surpised if there was metal dust and god knows what else in the gap of my 4552s, so at this point I'd call them damaged. I haven't seen any complaints about the 4550, but I have seen comments over the years about the 4552 not being as capable.

I originally got the 4552 to use in a Paraline synergy project about 5 years ago. I needed a 1" cd capable of a 1khz xo that was also quite small to sit between 4x 4" mids. Tbh I think these BMS drivers are kind of goofy after studying their construction and how the sound moves through the device - coming off the back of the diaphragm and being guided by a fairly crude phase plug along the back chamber before changing direction and entering the throat. The throat itself has a short, straight duct section before a sharp transition to the final exit angle. Reminds of said failed Paraline project :rolleyes:. That said the freq response isn't really any better/worse than a comparable driver, however imo a conventional dome driver just seems like a better design.

I chose the DE500 this time for it's low xo capability and I like the fact that it has a short, rapid flare in the throat due to the smaller neo magnet structure. I designed my waveguides for a 24° entrance angle and wanted to stick with that. While I don't have any way to prove it, I think having less of a transition in angle in the wg is good from a diffraction perspective.
 
I should add that the short throat on these neo drivers can be an advantage in a Synergy horn as far as the mid/cd reflection notch is concerned. I gained maybe 50hz in high end extension on the mids which is nothing really, however there is now a pronounced reflection notch around 2khz. With the 4552 there was no real discernable notch.
 
I think there's three things going on here:

1) the BMS drivers simply have less output below 1500hz. If you compare a BMS to a B&C, it's like comparing a woofer with low QTS to a woofer with high QTS. The F3 may be similar, but the roll-off is shallower in the former. The cause of this is low MMS; the BMS has low MMS because it's a ring not a dome.

This might be the reason Danley uses them; the slower roll-off makes it easier to get the mid-range and tweeter in phase.

2) the problem may have become worse over time because the drivers broke in. You see the same thing with woofers, the qts and fs goes down when they're broken in.

3) there are a few solutions you can try. First, use a waveguide with a narrower angle. If you cut the coverage angle in half, the tweeter works half as hard to maintain the same SPL. Another option is (gasp) a diffraction slot. A diffraction slot loads the compression driver by reducing it's radiation to a smaller volume. (A Paraline is a diffraction slot, albeit an oddly shaped one.) Or all of the above. If you look at the Danley catalog, he uses all of these options in various models.
 
2) the problem may have become worse over time because the drivers broke in. You see the same thing with woofers, the qts and fs goes down when they're broken in.

I think "broken" rather than "broken in". I just looked in the gap on the 4552 under magnification and there is definitely some fine metal shavings sticking to the walls. Also, there appears to be some sort of corrosion on the edge of the vc with the former.....due to the metal dust I guess. I would take a pic but my phone can't focus that small.

Moral of the story - keep the damn compression drivers out of the shop!
 
Bah! I'm getting my "Erics" mixed up. The Holdaway family owns Speakerworks in Orange CA, and they largely introduced the use of prosound gear into the car audio world back in the early 90s. A lot of Emilar gear (Emilar was a few blocks from their shop, and I believe a lot of the people from Emilar wound up at Radian, which is also in the same vicinity.) Eric Holdaway was one of the people who encouraged me to DIY horns back in the 90s, and sent me off on this crazy DIY journey.

Eric Stevens is also a big proponent of horns. He's one of the founders of Image Dynamics. Bruce Edgar's warehouse used to be next door to Image Dynamics.

Eric Stevens is the one who encouraged car audio guys to clean those compression drivers:

Cleaning Voice Coil Gap on Compression Drivers - Car Audio | DiyMobileAudio.com | Car Stereo Forum
 
FWIW: In post #239 and #255 of this thread Dr. Geddes refers to the papers of Dr. David Griesinger on perception of phase coherence, or offically: "Phase Coherence as a Measure of Acoustic Quality".

A quote:
"The three papers in this series focus on engagement, which results when sounds demand, and hold, our attention. Engagement is related to the perception of distance, and is distinct from intelligibility.
Sounds perceived as close demand attention and convey drama. Sounds perceived as further away can be perfectly intelligible, but can be easily ignored. The properties of sound that lead to engagement also covey musical clarity – one is able, albeit with some practice, to hear all the notes in a piece, and not just the harmonies."

All three papers are available here.
 
These poor things will never get a rest. Mocking up woofer mounts to go full 3-way synergy. I moved in with my (now) fiancee a little over a year ago and the 2way Synergy over 15" woofer stack speaker setup was stuck in the basement and the are far too large for that space. 3d printed the mount to fit a B&C 8bg51 8" woofer (2 per horn) and nailed it on the first go. Nice to see that my hand built fiberglass horns conform to my 3d model pretty well. Looking now I see that the PRV 8mb500v2 looks to do the same job at half the price. Anybody have woofer suggestions with high eff, power handling and can do 60ish vented in a small cab?

Horns will get boxed in and then the final unit will be ceiling mounted.
 

Attachments

  • IMG_20191231_151522.jpg
    IMG_20191231_151522.jpg
    567 KB · Views: 301
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.