AMT build, suggestions/experience

[Squidspeak]

I have owned a AMT speaker system for over thirty yrs. I bought a pair of
ESS AMT1C speakers back in the early eighties and when the woofers and
cabinets were past there lifespan around early 2000 I used the AMT and did
my second DIY build turning them into a 3-way with active analog X-overs.
AMT on top 1000hz. and up, Morel 5" for the mid at 100hz. up to the heil AMT,and a Titanic Mk1 as a built in sub in each cabinet from 100hz.down. The
AMT is still mounted in a Di-pole config.,the mid in a too small sealed chamber,
and the sub aprox 1.25 cu. sealed. Recently went digital with the X-over(DCX-2496). Anyway it still sounds better than almost any systems that I've
heard. I started getting back into the DIY about two yrs. ago and plunged into trying to learn as much as possible and lurking on several forums and what I found out was I did'nt know jacksh#t and would have done many things differently. I was lucky I had the option of the active X-overs or it pro-
bably would have been a nightmare. I have humbled myself and started doing
proven designs and trying to learn but at the age of 55 and not great math
skills I have to leave the X-over designs to someone else or go active. Back
to the point here is I want to build another 3-way modern AMT system with
Di-pole tweeter and mid so what drivers would people recommend for the AMT
Eton, ESS, Mundorf,Arum Cantus, Beyma, Hawthorne? Obviously this will be
top end build so also need top end mid that will work for Di-pole and a musical
sub that can handle HT as well as music. I'd say 80% music 20% HT. This will
be my final build for myself so I want to do it right.

[Pete Schumacher]

Re: AMT build, suggestions/experience

The Beyma AMT comes with a rear chamber. But I don't think it's a requirement. No doubt it could be adapted to dipole use. And if this is a "last" build, then go for the gusto. It's a great tweeter.

As for a mid to match it, I'd go large. SEAS Excel W22 comes to mind.

[GranteedEV]

Re: AMT build, suggestions/experience

not great math
skills I have to leave the X-over designs to someone else or go active

If you want true, consistent dipole operation, you'll probably want to go at least hybrid active-passive for dipole equalization.

I think a TPL-150 + Neo8 + 2X Scanspeak 18WU, in minimal baffles, would make a pretty sick speaker. :D :D :D :D

Throw in a pair of RSS315 around 140hz for good measure.

[Squidspeak]

Re: AMT build, suggestions/experience

If you want true, consistent dipole operation, you'll probably want to go at least hybrid active-passive for dipole equalization.

I think a TPL-150 + Neo8 + 2X Scanspeak 18WU, in minimal baffles, would make a pretty sick speaker. :D :D :D :D

Throw in a pair of RSS315 around 140hz for good measure.

At first, I was thinking of just putting a AMT into the Linkwitz/Phoenix design
using the ESS unit because of the low X-over capability and use his active
X-over. I think that would probably work but I think I can do better with a
newer AMT design. I never see a DIY design using the ESS driver, my old one
sounds great but cost is most likely the reason. I have heard the Scanspeak
as well as the Seas mids would be great for Di-pole mid duty, do you think the
Neo8 would be needed if the X-over on the AMT was low enough? And this
will only be my personal "final build" as I plan finishing the two I'm already into
and many more!

[Pete Schumacher]

Re: AMT build, suggestions/experience

At first, I was thinking of just putting a AMT into the Linkwitz/Phoenix design
using the ESS unit because of the low X-over capability and use his active
X-over. I think that would probably work but I think I can do better with a
newer AMT design. I never see a DIY design using the ESS driver, my old one
sounds great but cost is most likely the reason. I have heard the Scanspeak
as well as the Seas mids would be great for Di-pole mid duty, do you think the
Neo8 would be needed if the X-over on the AMT was low enough? And this
will only be my personal "final build" as I plan finishing the two I'm already into
and many more!

The Neo8 is probably superfluous if you're using a larger AMT, like the Beyma, or one of the big Aurum Cantus Aerostriction tweeters which can cross at 1000Hz.

Another alternative for a mid would be to use one of the Bohlender Graebner line source panels. They can cross down around 200Hz to a nice woofer. http://www.parts-express.com/pe/show...number=264-706

[Face]

Re: AMT build, suggestions/experience

Or you can use an AMT to 1-2K, a Neo 10 or two beneath it, with a nice woofer such as a L26 Roy up to 200-300hz.

[GranteedEV]

Re: AMT build, suggestions/experience

The Neo8 is probably superfluous if you're using a larger AMT, like the Beyma, or one of the big Aurum Cantus Aerostriction tweeters which can cross at 1000Hz.

What I had in mind was to get as wideband a dipole response as possible, similar to this:



So I don't know if the TPL-150 would play down that low give the 6db/oct gradient rolloff. It's faceplate is 3.5 inches. A larger mid would stop being dipole much earlier than where you`d probably cross the free air TPL, leading to a discontinuity.

The aerostriction tweeter has an even smaller faceplate, which means it should become dipole slightly higher in frequency.

You can use a baffle to dodge the rolloff in acoustic response, but then you`re not really getting the biggest benefit from the dipole, which IMO is a waste with such a wicked tweeter.

[Pete Schumacher]

Re: AMT build, suggestions/experience

What I had in mind was to get as wideband a dipole response as possible, similar to this:



So I don't know if the TPL-150 would play down that low give the 6db/oct gradient rolloff. It's faceplate is 3.5 inches. A larger mid would stop being dipole much earlier than where you`d probably cross the free air TPL, leading to a discontinuity.

The aerostriction tweeter has an even smaller faceplate, which means it should become dipole slightly higher in frequency.

You can use a baffle to dodge the rolloff in acoustic response, but then you`re not really getting the biggest benefit from the dipole, which IMO is a waste with such a wicked tweeter.

Please clarify how the above picture jives with your characterization "wide band dipole" response. The baffle in the picture is not the "ideal" dipole baffle. The NaO note is closer to that "ideal" than the simple, large, rectangular baffle you have pictured.

[GranteedEV]

Re: AMT build, suggestions/experience

Please clarify how the above picture jives with your characterization "wide band dipole" response. The baffle in the picture is not the "ideal" dipole baffle. The NaO note is closer to that "ideal" than the simple, large, rectangular baffle you have pictured.

That is not a baffle. The drivers are suspended in free air on a frame with very tight wire.

[Squidspeak]

Re: AMT build, suggestions/experience

At first it looked like that was a glass/plexi baffle, but very interesting to
know it is suspended by wire,anyway great ideas so far. Out to dinner at
a freinds so can't wait to get into the fray when I get home.

[jonasz]

Re: AMT build, suggestions/experience

Would have loved to have had the Beymas when I was experimenting with my Phoenix dipoles. Go for the W22's, awesome mid!

The panels, here with the RS28A front and rear.

[GranteedEV]

Re: AMT build, suggestions/experience

At first it looked like that was a glass/plexi baffle, but very interesting to
know it is suspended by wire,anyway great ideas so far. Out to dinner at
a freinds so can't wait to get into the fray when I get home.

As long as there is a baffle, at higher frequencies it will reflect the "omni" portion of the rear and forward waves, preventing them from cancelling each other out and creating the dipole null.

This also applies to the faceplate, and eventually the driver itself.

What John K did with the Nao Note, pictured in Pete's post, is similar, but probably better. It uses the tiny SS discovery 10F mid above 1khz in what you can barely call a baffle at all. But instead of crossing to the tweeter while the mid is a dipole, it uses the mid's natural narrowing of dispersion to maintain the narrowed off-axis response up to 4khz - tough to tell apart from a dipole.

At 4khz, the Neo3 tweeter rolls in. But rather than using the off-axis nulls to get the dipole response, it uses a waveguide on both the front and back to match the narrowing dispersion of the 3.5" midrange. The result is as smooth a polar as you could want in the front hemisphere (the back hemisphere is slightly worse due to the magnet of the 10F). it's truly dipole up to around ~maybe 2khz, where the transition to a pseudo-dipole is almost impossible to recognize:



I don't know how easily a two-sided 120deg waveguide could be built around an AMT. Certainly would be cool though. You could match it to an ~8 inch midwoofer.

[Squidspeak]

Re: AMT build, suggestions/experience

I know this is going to be a tech intensive project and am trying to gather as
much info as possible before chosing drivers and design. So as much as possible, I definetly want theory and explanations for peoples recomendations.
I have tried to absorb every thing on the Linkwitz site but unless that's your
field it is alot to take in. Anyway probably looking at fall to make driver and
cab design choices. I do like L26 Roy suggestion, possibly in as used in the
Orion or John M's Di-pole project pending.