[andyr]

I bought some Maggie IIIas in about 1990 from Mike Kontor (Leading Edge Audio, in Melbourne), when he switched from selling Maggies to selling Apogees - I bought his personal pair.

After a couple of years I got stuck into replacing the stock external & then internal XOs (after attending a 6-week night course at Box Hill TAFE on speaker/XO building) with ribbon coils and good quality caps ... although 18 years ago there wasn't the range of film caps which we have now!    I also did significant structural mods to stiffen up the MDF frame, and had some braced stands made up to lock the panels in place.  Then I went 3-way active.

Each mod improved their sound and this lasted a few years but the final mod I made in 2007 was to replace the stock MDF frame with a hardwood frame (recycled ironbark), as there'd been a lot of posts on the Planar Asylum as to how this greatly improved the sound they produced.  It certainly did - and, as proof, I can say that Hugh Dean (designer of my AKSA amps, who doesn't like Maggies!  ) said "Andy, at last they're interesting to listen to"! 

So, having done all I could to my IIIas (basically, throwing everything away except the drivers themselves), I started to think how I could make them even better.  The obvious thing to do was build an equivalent to my IIIas (ie. a 3-way Maggie with the same sized drivers) but which had the bass panel on a separate sheet of mylar to the mid panel (thus eliminating - or at least reducing - IM distortion).

After researching the various superseded Maggie models, I ended up combining the bass panels from a pair of MG-2.5Rs (a 2-way which has a bass panel the same size as my IIIa bass panels plus a 2/3rds size true-ribbon) with the mid panels and ribbons from some Tympani-IVas.  After having identified these were the components I wanted to use ... then I had the problem of how to source them! 

Eventually, I did - and the result is my "Frankenpans".  These are unique Maggies - they have drivers the same size as my IIIas but with the mid panel entirely separate to the bass panel (in effect, Tympani-IVas with only 1 bass panel per side!  ).  After 2 years of planning and doing, I was delighted a few months ago to be able, finally, to listen to them and realise that they were indeed performing according to theory!    IE. they are a significant step up from my IIIas, being much "cleaner", as:
a) there's no IM distortion happening between bass panels and mid panels, and
b) because the bass panels are separate, I was able to implement 2 hardwood frames each side - one for the bass panel and a second one for the mid panel & ribbon.  So no vibrations from the bass panel get to the ribbon. 

Even Hugh was impressed, when he came round to hear them a few weeks ago! 

Pics below:










Regards,

Andy

PS:  Yes, they are ugly (Maggie naked is not a good sight!    )!!  So the next task I have promised my wife is to have some black socks made up, so they'll come out looking like my IIIas:

[SteveFord]

Those are far and away the stoutest stands I've ever seen for the Maggies.
What sort of amps are you using for those and could you tell us more about what you did with the crossovers boxes?
I'll never go as far as you did but you do hear an awful lot about modifying the stock crossover boxes.

[andyr]

Those are far and away the stoutest stands I've ever seen for the Maggies.

Hehe - thank you, Steve.   

I put a fair bit of thought into the stands - and these are my second go at them.  Of course, it's always possible something else would be better!   

The U-shaped bases are made of 75 x 50 x 6mm steel U-channel, for weight; the A-struts are made from 50 x 25mm aluminium angle (aluminium rings less than steel).  I didn't want the struts to be made of square tube because then you have to decide what to fill the tubing with!    So I chose angle.

The A-struts are attached to the hardwood frames by 4 screws each and are bolted to the U-base by M8 bolts.  In fact (not shown in the second picture - as it was taken before I'd done this) the single strut which supports the mid/ribbon frame has isolation mountings - ie. a cylinder of rubber with a bolt sticking out each end - rather than just bolts.  I did this because with the mid/ribbon strut just bolted to the U-base, I was getting a transfer of vibrational energy into the mid/ribbon frame.  (When the bass panel vibrates, it shakes its frame, these vibrations shake the A-struts ... which causes the U-base to vibrate.  Without the isolation mounts, these vibrations in the U-base were simply coming up into the A-strut holding up the mid/ribbon frame and causing that to vibrate!    Installing the isolation mounts, and resting the base of the mid/ribbon frame on a strip of 10mm Sorbothane, prevents any vibrations from the bass frame from reaching the mid/ribbon frame.)

What sort of amps are you using for those and could you tell us more about what you did with the crossovers boxes?

Inside the black case shown in the second pic are 3 AKSA Lifeforce modules and a 3-way active XO (using Rod Elliott's PCBs).  And their associated PSes.  So I have (measured into 8 ohms):
* 100w on the bass panels
* 55w on the mids, and
* 25w on the ribbons.

What did I do with the original external XO boxes - I junked the components but kept the boxes.  I used them in my replacement external XOs which I made over 15 years ago:




My long-suffering wife put up with these ugly things for several years but the current "black boxes" are the culmination of a many-year project to improve the WAF of the Maggies! 

I'll never go as far as you did but you do hear an awful lot about modifying the stock crossover boxes.

Maggies are wonderful speakers but they are built to a price.  So they can be made to sound better in a whole number of ways - one of which is to rebuild the XOs using high quality parts.  I also believe that while the slopes should be retained, the -3dB points of each filter should be tweaked slightly, for a better overall FR.  (To this end, I purchased lspCAD about 10 years ago.  )

Regards,

Andy

[gregmacknass]

Hey Andy,
What did you end up doing about that 4a mid panel that had the broken magnet? Did you just leave well enough alone  or did you get the repairman to fix em.
-Greg

[andyr]

Hey Andy,
What did you end up doing about that 4a mid panel that had the broken magnet? Did you just leave well enough alone  or did you get the repairman to fix em.
-Greg

Ah, that was a bloody annoyance which took me a few days diagnosis before I could figure out how to make a workaround (as "fixing it" would IMO have meant sending that mid panel back to the factory for a re-lay of its mylar.  All at my own expense, of course - for a bloody manufacturing fault!  ).

The workaround was to use a small piece of 1/16th" sheet steel on the mylar - so the magnets suck it in, trapping the mylar around that position.  See here:





Regards,

Andy

[peck555]

Thanks for posting this.

[rollo]

As a 3A owner way cool. They must sound great.


charles

[*Scotty*]

Great piece of work, I always loved the sound of the Tympanys and the rest of the series was a sonic compromise compared to them. Your replacement for the mdf is heroic. I have wondered what the result would be if the mdf was replaced by Corian in a properly built frame,but I could never afford the material to work with. I stopped modding my MGIIIs after re-engineering the cross-over and implementing a 12dB/oct. Linkwitz-Riley network.
Scotty

[pelliott321]

I have followed Andy's trip for the pass year on AA MUG and he has been a big help to me with my rebuild. I am presently using his passive xover circuit and I am liking the sound, but I would like to try active down the road. 
Scotty is your Linkwitz-Riley active or passive?

[pelliott321]

a couple of pics of the external xover:




the xover design is from AndyR at
http://www.integracoustics.com/MUG/MUG/tweaks/AndyR/
Most of the info for this project came from either AndyR or Peter Gunn at AA Planer
Andy is most generous and patient.....PG well is PG   

[*Scotty*]

It was a passive network.
Scotty

[pelliott321]

Scotty
I would be interested in the circuit diagram of your passive

[*Scotty*]

A L-R2 network is a 12dB/oct. parallel network. The topology is the same as a 12dB/oct. Butterworth
network. Only the parts values are different the layout remains the same.
Scotty

[pelliott321]

thanks Scotty
I will do the research and see what it looks like.
Of course the other option is to go active like Andy. The active xover is the big thing.  Alot over at AA use the behringer and there are mods all over the net to make it audiophile.  There is a L-R active board from the guy in Australia that looks interesting.  I have also looked at miniDSP, but worried about the quality of its audio output.
lots to think about