Unity horn with 2 x Eminence Alpha 6 drivers - Page 2 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #31 of 75 Old 05-03-2012, 04:45 PM
AVS Special Member
 
Ivan Beaver's Avatar
 
Join Date: Dec 2005
Posts: 1,631
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 38
Quote:
Originally Posted by johnbomb View Post

1) The first shows ground plane measurements of the driver sitting on its side on the ground (green) vs handheld against the back of the horn (red). Things look really good until about 8 khz, where the FR rolls off pretty quickly, faster than it does for the raw compression driver measurement. Also, I had hoped for lower extension from the compression driver, to make my crossover easier... Oh well. Hopefully, bondo contour shaping will help things.

.

The 4550 is a really nice driver. It is used in a number of Danley products (including the SH50).

It has a lower freq response than most all (that I am aware of) compression drivers (that have a good HF response).

As with most all compression drivers on a horn-there is a "haystack" response-ie a hump in the middle.

But there is also a high sensitivity-so you have to "throw away" that excess gain-to match up with the rest of the drivers in the cabinet.

When you do that-the HF and low end response will "appear" to be flatter.

Did you change the gain or mic position in either of the measurements?

It could be that the HF is not actually lower-but the "middle" of the response is simply louder-making the ratio of what you are looking at different.

When doing measurements-it is really important to keep all things the same-or you aren't sure what you have ended up with.

Danley Sound Labs

Physics-not fads
Ivan Beaver is offline  
Sponsored Links
Advertisement
 
post #32 of 75 Old 05-03-2012, 09:39 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Got the mount built- stole that one from Paul Spencer (thanks for the documentation of your builds, Paul!). What I did differently: I used a 1 inch forstner bit for the center throat hole and a bandsaw/belt sander for the outside instead of a lathe. Glue up underway, and filing/filling tomorrow.

Ivan, I've been thinking about those measurements, and I don't think they were controlled enough. I'll redo everything once I have some corner fill in place.

For the corner fill, I have 2 ideas in mind: first, quick fill with Play-doh- it's easy, and my kids have it in the house (always a plus). If I start getting results, then I'll pull it out and redo with drywall mud. It goes on and sands more easily than bondo, and dries pretty quickly. Once I have the shape I want, I can lock it into place by soaking it w/ superglue... at least that's the idea. I'll try it on a test piece first.

John
LL
LL
johnbomb is offline  
post #33 of 75 Old 05-04-2012, 07:24 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Got everything assembled after lots of corner fill with drywall mud (will show pic later). The groundplane corner frequency of the BMS 4550 has dropped from around 1100hz to around 875hz (curves staggered for clarity)- now, we're cookin' with butter! I suspect that adding a 1/2" extension to the horn throat in the form of the compression driver mount is mostly responsible for the lower corner, given that, even after filing and contour shaping, the throat is now nearly cylindrical with very little early expansion.

Strangely, though, the high corner rolloff rate has been increased a bit as well. I'll try measuring the high frequencies differently (and with a real condenser microphone) to see how it looks. I think I'm almost there! By the way, 122db at 2.83v... that's just insanely loud.

John

Edit: went back later and coated the drywall fill with Zinsser Guardz, a drywall sealer/stabilizer. Hopefully, this will keep everything from falling apart later.
LL
LL
johnbomb is offline  
post #34 of 75 Old 05-05-2012, 07:43 AM
AVS Special Member
 
Ivan Beaver's Avatar
 
Join Date: Dec 2005
Posts: 1,631
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 38
Quote:
Originally Posted by johnbomb View Post

By the way, 122db at 2.83v... that's just insanely loud.

John


Where is that measured from? Either you are VERY close to the driver-or your calibration is way off-or your volt meter is wrong-or something.

That driver cannot do anywhere near that- at a standard 1M distance with 2.83V input.

Yes it has a high sensitivity-but not nearly that high.

Danley Sound Labs

Physics-not fads
Ivan Beaver is offline  
post #35 of 75 Old 05-05-2012, 12:16 PM
FOH
AVS Special Member
 
FOH's Avatar
 
Join Date: Apr 2010
Location: Midwest
Posts: 4,747
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 208
Quite interesting so far,...thanks for sharing and best of luck.

------------------------------------
Flat, Deep, Clean, Linear, and Loud
------------------------------------
Active 16.8kw, 7.3 system
(3)Seaton Cat12C up front, (4)QSC K8 sides/rears
(2)Seaton SubM-HP, (4)18" IB
FOH is offline  
post #36 of 75 Old 05-05-2012, 06:17 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Crossover success! Crossed at 950hz, using 2nd order LR slopes and .36ms delay on compression driver w/ minidsp. To get the time delay right, I kept adding delay and watching the reverse null- I stopped when I got the sharpest dip. Acoustic slopes are actually higher order and a bit eliptical- will try to measure and post later. Pic 1 is crossover phase and spl, pic 2 is reverse null at 180 degrees out of phase. Oh yeah- pics 3 and 4 are drywall corner fill.

Check out Zaphaudio.com, especially his discussion of crossover design on his ZD5 project- really great info:

http://www.zaphaudio.com/ZD5.html

Don't have off axis plots of crossover (deleted them accidently), but no phase shift through the crossover as far off axis as I measured. Looks like a pretty damn good reverse null to me. I've spent the better half of this afternoon just listening to tunes in mono with no bass to speak of and loving it! Now, I need to build two more of these things and add some <300 hz signal...

Just a word of encouragement to others who might be intimidated by the prospect of speaker building: YOU CAN DO IT!!

John


Edit: Ivan, I'm sure you're right about the SPL- I'll redo some outdoors measurements and see what turns up.
LL
LL
LL
LL
johnbomb is offline  
post #37 of 75 Old 05-06-2012, 10:09 AM
AVS Special Member
 
Ivan Beaver's Avatar
 
Join Date: Dec 2005
Posts: 1,631
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 38
Quote:
Originally Posted by johnbomb View Post

I've spent the better half of this afternoon just listening to tunes in mono with no bass to speak of and loving it!.

Hence part of the real magic of the Synergy design. When the sound all comes from a single location-it doesn't matter where you move in the coverage pattern-the response (particularly phase) is going to be the same.

When you have drivers that are offset from each other (say a woofer with a tweeter above it or off to the side)-as you move around the room-the arrival times change (yes the little timing changes mean a lot), so there is a notch in the response that changes as you move.

With speakers- it is far better to be "a country mile" off- than "pretty close".

Danley Sound Labs

Physics-not fads
Ivan Beaver is offline  
post #38 of 75 Old 05-06-2012, 11:45 AM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 16,115
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 198 Post(s)
Liked: 912
"When you have drivers that are offset from each other (say a woofer with a tweeter above it or off to the side)-as you move around the room-the arrival times change (yes the little timing changes mean a lot), so there is a notch in the response that changes as you move."

that's true if the horizontal directivity of the two sources is different in the crossover region for vertically aligned sources. good designs don't suffer from that problem.

mr. bomb, the results look fantastic. can you widen the response to something more like 200-20k hz?

Listen. It's All Good.
LTD02 is online now  
post #39 of 75 Old 05-06-2012, 02:47 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by LTD02 View Post

mr. bomb, the results look fantastic. can you widen the response to something more like 200-20k hz?

As requested... All results here taken indoors in an acoustically terrible room loaded with early reflection points, at 83cm from horn apex, 1 volt, SPL calibrated.

mr. bomb- HAH! You're too kind!

I still can't explain the high end rolloff that starts at around 9k, although it is better than the 7k that I was getting outdoors. I'm guessing a combination of normal high end rolloff of the BMS 4550 plus a lack of horn loading occurring at those higher frequencies. Maybe it's related to the fact that I'm measuring with a C weighted SPL meter, and the compensation in REW and the actual C weighting rolloff don't match. Whatever- it's no big deal to me, and my hearing is good to around 14500 hz (guns, drums, powertools, you know, the usual stuff ). I had to add a smooth negative high shelf to make it sound good anyway, just as jmorken did with his horn at diyaudio. I still want to try the horn filled with 30ppi foam. Beyond that, I don't see this speaker getting much better... like I can complain at this point!

John


Edit: What does this horn sound like? I've sat in some high end recording studio control rooms (Belmont University, Nashville TN), and even in my crappy space, the horn very much reminds me of that super dry, super accurate sound that you get in a million dollar room that's set up by people who know what they are doing. I can't wait to get my room treated properly, and apply some LEDE principles via ETC analysis... maybe I can make a set of these horns sound even better!
LL
LL
LL
LL
LL
johnbomb is offline  
post #40 of 75 Old 08-25-2012, 05:15 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Wanted to give an update...

After lots of testing at high volumes, I've come to the conclusion that my current design doesn't meet my HT goals. There's way too much distortion generated by the horn when the Alpha 6s are driven to around 119db/meter, which is the volume I need to reach reference at my listening position. So far, I've tried everything I know to lower distortion, including the following:

1) putting a 1/8" spacer between the drivers and the horn body, thinking that the cones were rubbing against the wood at their max excursion- this might have helped, not sure yet. Xmax shouldn't be more than 1.5mm or so at 300hz at 119db. This lowered the upper f3 by about 75hz or so.

2) adding an entire sheath of wood around the horn, with green glue between the leaves, to deaden internal resonance. This has helped- the cabinet did ring quite a bit when I hit it before the modification, and now it makes more of a "thud."

3) added metal surrounds to the backs of the Alpha 6s to create stiffer sealing, thinking that the cardboard/tape combo was letting air out, which might have caused turbulent noise. This lowered my lower f3 by about 75 hz.

During my testing, I put my hand into the horn and closed off ports while playing at 119 db at 300hz to check the effect, and man, you would not believe how fast that air zooms in and out of those things! I modeled a 4th order bandpass box in WinISD to try to match those created by the Alpha 6s and the horn body, and low and behold, the port velocity at 300 hz, 119db is something like 100 m/s! Also, I think there's some buzzing/humming that's caused by the ports at certain frequencies- the port edges are so thin that they might be resonating.

So, I'm going to stretch the ports out along the radial axis of the horn to get the port air velocity down and also try to reshape them to make air flow more smooth. My fear in so doing is that I will create more diffraction generating interference for the higher frequencies handled by the compression driver. At this point, this project has become one of aerodynamics, and I don't have any formal training in this field. All I can do is guess at proper shapes, test them, and post results. Hopefully, I'll have something positive to report in the near future.

John



johnbomb is offline  
post #41 of 75 Old 08-25-2012, 06:56 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Here are the distortion measurements generated by the 2 Alpha 6s mounted on the horn with respective FR graph. All measurements taken outdoors, groundplane. First, an 800hz tone was played until 119db was reached.











johnbomb is offline  
post #42 of 75 Old 08-25-2012, 09:18 PM
Advanced Member
 
mtg90's Avatar
 
Join Date: Aug 2011
Location: Millbrook, IL
Posts: 782
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 11 Post(s)
Liked: 139
I'm glad to see you still at it, sad to see it giving you trouble.

What are you hearing when the volume is turned up? I cannot interpret those distortion graphs at all. confused.gif
mtg90 is offline  
post #43 of 75 Old 08-25-2012, 10:18 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Hey mtg90,

I don't really like those graphs, either- just put them there for completeness. The sound I'm hearing is that typical overdriven, "buzzy" sound that we all know and hate- whatever sound of crappy distortion you hear in your mind, that's pretty much it. I think my horn design is good, but the port execution isn't. I'm going to bondo the ports and start over... was thinking that drilling lots of small holes might be the way to go- maybe would limit compression driver diffraction while lowering port velocity (if I drill enough of them).

John
johnbomb is offline  
post #44 of 75 Old 08-26-2012, 02:51 PM
Advanced Member
 
mtg90's Avatar
 
Join Date: Aug 2011
Location: Millbrook, IL
Posts: 782
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 11 Post(s)
Liked: 139
I think you have it figured out, it needs a major port overhaul. Your trying to squeeze a roughly 20 sq in radiating surface through what looks to be two 3/4" ports, thats less then 1/20 the area.

The multiple small holes is an interesting idea.
mtg90 is offline  
post #45 of 75 Old 08-26-2012, 08:35 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
So, I drilled a bunch of holes, and the results are encouraging. First, a little background...

After deciding that I had too much air squeezing through ports that were too small, I started trying to figure out how to balance increased port size with increased diffraction/signal loss. My first thought was to elongate the ports in the radial axis of the horn, as I said originally, and that might still work. However, the more I thought about diffraction, the more the multiple small hole idea started to make sense. Without a megabuck fluid dynamics modeling program, I went with what seemed right for the remodel.

First, I bondo-ed the original ports and planed everything down. Next, I modeled 4th order bandpass boxes in WinISD, comparing the best guesstimate of the original 2 port alignment with the new multi ported one until I got equivalent FR curves. I decided to make the holes 1/4 wavelength the limit of my hearing, which is around 14khz, which is approximately 1/4 inch. I wound up drilling 30 holes with a 1/4" bit. As you can see, I changed the spacing between the holes (dowel rod fillers in the first two rows). Also, I made sure that the holes were as evenly spaced from one another as possible. Then, I used a countersink bit and a dremel sanding cone to relieve the edges of the driver sides and horn sides of the holes, respectively, which will hopefully lower air turbulence and maybe even diffraction.

I only got one side done and need to do a bit of refining, and I ran out of time today (kids in bed, no more noise!). I did cheat and ran one quick sweep with only one driver attached- not too shabby, I think. Subjectively, I heard less distortion, but it's too early to know for sure without some controlled measurements. All in all, though, things seem to be going my way, at least for tonight smile.gif

John







Edit: disregard this last graph- data is unreliable, probably showing the effects of power compression at high SPL.
johnbomb is offline  
post #46 of 75 Old 08-26-2012, 09:16 PM
Advanced Member
 
mtg90's Avatar
 
Join Date: Aug 2011
Location: Millbrook, IL
Posts: 782
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 11 Post(s)
Liked: 139
That looks cool, just like I imagined, and the graph sure looks good.
mtg90 is offline  
post #47 of 75 Old 09-09-2012, 07:10 AM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
I got lots of great information yesterday- one of the most useful testing days yet. All measurements taken outside from groundplane with a CAD 179 condenser mic.

First, the players. Pics #1 and 2 show my original horn (the one I built in April) with a modification. I cut out part of a port side wall and replaced it with a 1/4" piece of birch plywood that I had perforated with multiple 1/4" holes, spaced 1/4" apart. I also deeply recessed the holes with a recess bit. I used a single Alpha-6CBMRA driver, attached to the tested side, with the other side taped off. I decided to buy one of these drivers to see if it would outperform the standard alpha-6 drivers that I have been using. This driver has a solid metal sealed back, and looks a heck of a lot better than that homebrew seal job that I showed above. Long story short- the 6CBMRA is a better choice.

new birch hole plate:



other side, original ports:



new alpha driver:




This first set of graphs shows something interesting and unexpected. At first, I had run sweeps through the alpha into the birch hole plate with the BMS 4550 compression driver attached but not hooked up to the amp. I was getting a dip at 800hz and couldn't explain it, until I thought that maybe the compression driver was acting as a selective passive absorber. With this idea in mind, I attached the BMS to another amp channel with no signal applied (with a 50uf capacitor in series to protect against DC thump), thinking that maybe the amp would "hold in place" the membrane of the compression driver. Sure enough, that was the case, and the 800hz dip went away.





The second graph shows the birch hole plate response compared to that of the original two ports. Wow, what a difference! The hole plate evidently improves the 4th order bandpass efficiency by letting more air through the chamber into the horn. The ports are effectively around 1/8" long after recessing. I remember Tom Danley recommending that the ports should be very short.



Next, I'll show the effect of the hole plate vs the ports on the BMS 4550 response.

John
johnbomb is offline  
post #48 of 75 Old 09-09-2012, 08:56 AM
Advanced Member
 
mtg90's Avatar
 
Join Date: Aug 2011
Location: Millbrook, IL
Posts: 782
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 11 Post(s)
Liked: 139
Nice, looking forward to seeing the CD before/after graphs.


For the midrange dip at 800hz just as an experiment try shorting the terminals on the BMS with a jumper it should give the same effect as hooking it up to the amp.
mtg90 is offline  
post #49 of 75 Old 09-09-2012, 09:42 AM
Advanced Member
 
bwaslo's Avatar
 
Join Date: Jul 2006
Posts: 812
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 73
Tom Danley said that he makes his crossovers so that the last element (in parallel with the tweeter) is a small inductor. That gives the CD compensation and also acts as a short to lower frequencies to keep the CD diaphram stiff.

How do the little ports work as far as modeling in Hornresponse? Do you just take the average distance down the horn as the distance from the throat? Or is it more empirically designed? Can't argue with the graphs, it seems to work there.

DIY Synergy horn spreadsheet http://libinst.com/SynergyCalc/
XSim -free crossover designer and simulator http://libinst.com/Xsim/XSimSetup.exe
bwaslo is offline  
post #50 of 75 Old 09-09-2012, 04:58 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Bill, that's a good question. The short answer is, WinISD's modeling of a 4th order bandpass box helped me the most, likely because I didn't set up the horn/driver model properly in hornresp. For example, WinISD shows that the combination of adding front chamber volume (in this case, area under the cone), shortening the ports, and adding number of ports will shift the FR curve up and increase overall volume through the passband, with a slight increase in the low frequency corner, and that's what has happened. I pretty much just guessed, though, at the number of holes in the plate. I tried lots of other stuff, like adding/subtracting holes in my other horn- nothing else gave me the results I wanted. Better lucky than good!

By the way, disregard the graph in post #45- it's crap. It was taken indoors with the mic too close to the mouth and likely shows power compression. That plus probable internal reflections make the curve artificially flat.

John
johnbomb is offline  
post #51 of 75 Old 09-10-2012, 01:45 PM
Member
 
mattlong8's Avatar
 
Join Date: May 2011
Posts: 15
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Bravo Sir.
This is a wonderful contribution to the collective DIY data set.
Thank you,

-Matt

-Matt Long
San Francisco, CA
mattlong8 is offline  
post #52 of 75 Old 09-10-2012, 04:51 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by mattlong8 View Post

Bravo Sir.
This is a wonderful contribution to the collective DIY data set.
Thank you,
-Matt

Thanks, Matt. Hopefully, this story will have a happy ending! More to come...

John
johnbomb is offline  
post #53 of 75 Old 09-12-2012, 08:37 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
More results today... good news and bad news. First, the good news. My second Alpha-6CBMRA driver came in today, so I made another hole plate and mounted it up:



I measured again from groundplane with the CAD 179 condenser mic at 1 meter. No significant FR variations between sides within the passband of 350-1000 hz, and a 6 db increase when both measured together in parallel. Incidentally, mtg90, you were right: I jumpered the compression driver and that 800hz dip went away, just as you said. The 600 hz dips originate from the alpha 6 drivers themselves- the second graph shows the raw response of the Alpha-6CBMRA, handheld and close mic'd.




Now, the bad news. The two hole plates conspire to gobble up low end FR of the BMS 4550 compression driver! Damn!

Graph 1 shows the BMS with both hole plates exposed. Jumpering the alpha drivers made only a slight difference in the low end response. The BMS measurement was again taken with the 50uf capacitor in series.



This next graph shows the comparison of all holes taped to that of one hole plate taped and then both plates totally exposed. The cavities created by the holes evidently globally absorb CD frequency response with the biggest hits at 1540hz and from about 3500-5000hz, not exactly insignificant losses... what a disappointment frown.gif



Just for the heck of it, I taped the centers of the hole plates just to get a quick idea of the effect of the center holes. I didn't measure the alpha driver responses because my computer ran out of battery power.





...Not much better. Oh well, at least I can take those plates out and try some thing new... More to come.

John
johnbomb is offline  
post #54 of 75 Old 09-12-2012, 09:11 PM
Advanced Member
 
mtg90's Avatar
 
Join Date: Aug 2011
Location: Millbrook, IL
Posts: 782
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 11 Post(s)
Liked: 139
frown.gif
mtg90 is offline  
post #55 of 75 Old 09-13-2012, 06:58 AM
AVS Special Member
 
maxmercy's Avatar
 
Join Date: Nov 2008
Posts: 2,335
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 131
John,

Have you tried a line of holes on an equidistant arc from the CD? That may be a decent compromise between the Swiss-cheese plate and the 2-hole plate.

JSS
maxmercy is offline  
post #56 of 75 Old 09-13-2012, 08:04 AM
Advanced Member
 
bwaslo's Avatar
 
Join Date: Jul 2006
Posts: 812
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 73
i'd be curious about what would happen if you covered the holes wih something like a paper towel, something that might affect waves from the cd differently than from the mid?

DIY Synergy horn spreadsheet http://libinst.com/SynergyCalc/
XSim -free crossover designer and simulator http://libinst.com/Xsim/XSimSetup.exe
bwaslo is offline  
post #57 of 75 Old 09-13-2012, 08:58 AM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Those are some good ideas, guys... which leads me to wonder: how about replacing the plates with window screen? That might provide a fairly opaque surface to frequencies above 1k and yet still let out lots of air for the mids.... I'll try it out. Thanks!

John
johnbomb is offline  
post #58 of 75 Old 09-13-2012, 10:06 AM
Senior Member
 
dlouw's Avatar
 
Join Date: May 2006
Location: Oregon
Posts: 248
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Perhaps try covering 1/2 to 2/3rds on opposite sides?
dlouw is offline  
post #59 of 75 Old 09-16-2012, 09:43 PM - Thread Starter
Senior Member
 
johnbomb's Avatar
 
Join Date: Aug 2005
Location: Nashville
Posts: 331
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
I may be onto something with the hole covering idea. I taped up all but the throat side 2 rows and got some good results.


This graph shows the overall transfer functions of the sealed back alpha mids vs the BMS 4550... not too bad! Kind of looks like a second order-ish built in crossover. Of course, I'll have to get the phase flat with the proper delay, but that shouldn't be too hard.


The second graph shows the BMS 4550 response with the front 2 rows of holes exposed and then all holes taped


Next, the BMS through just the front 2 rows of holes vs the entire hole plate (as shown in post #53)


Finally, the alpha mids firing through just the front 2 rows of holes vs the entire hole plate


I'm not holding my breath yet, but this appears to be shaping up. Analysis: the holes in the plate past the first two rows don't add much to the mid response but kill the high response. I still have a 5db or so suckout in the BMS response until about 1400hz, but it's smooth. Also notice the lack of FR deviation beyond 1400hz-. That surprised me, as I figured that the closer the holes are to the throat, the more FR disturbance there would be- evidently, not so.

Now, it's time to test the 119 db barrier with distortion and FR plots as well as port velocity checks...

Keep watching!
John
johnbomb is offline  
post #60 of 75 Old 09-17-2012, 06:03 AM
AVS Special Member
 
Ivan Beaver's Avatar
 
Join Date: Dec 2005
Posts: 1,631
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 38
Quote:
Originally Posted by johnbomb View Post

I'm not holding my breath yet, but this appears to be shaping up. Analysis: the holes in the plate past the first two rows don't add much to the mid response but kill the high response. I still have a 5db or so suckout in the BMS response until about 1400hz, but it's smooth. Also notice the lack of FR deviation beyond 1400hz-. That surprised me, as I figured that the closer the holes are to the throat, the more FR disturbance there would be- evidently, not so.
Now, it's time to test the 119 db barrier with distortion and FR plots as well as port velocity checks...
Keep watching!
John
It is not just the holes-or the size of hte holes-but mostly the POSITION o fhte holes in relation to the HF driver. The reflected mid sound off of the HF driver is what is causing the notches.

It is all about path length and size fo the particualr wavelength involved. And a lot of other little things as well.

When you start thinking in terms of wavelength-THEN all sorts of "revolations" will come.

Danley Sound Labs

Physics-not fads
Ivan Beaver is offline  
Reply DIY Speakers and Subs

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off