2. best-of
  3. 1066162131

Horns and Triodes...?

Buddy of mine is interested in (anguishing over, really) building a system based on a pair of super-efficient horns and a low-powered triode amp. As this stuff tends to run even more towards the esoteric than the regular esoteric fare around these parts ("guy in X building out of his basement/garage...") upping the comfort level prior to taking a leap of faith (let alone actually listening to something first) is turning out to be darn near impossible.

Personally, being that I run Thiels and a huge sucking monster of an amplifier (relatively speaking), I know bubkis about this stuff -- so I throw myself at the mercy of the collective wisdom.

What's worked for folks? Is there anything or anyplace to listen in the NYC area? Tastes are running towards the exceptionally spartan, relatively cheap, diamond in the ruff among the recently-post-home-brew-eque market (if that makes any sense) with a focus on piano more than anything, if it were necessary to choose. Likely to be using a single source, too, so we're really talking no frills. Pointers?
mezmo

Showing 3 responses by gregm

gregm
3,411 posts
 
Paul -- thanks for yr kind words. OTOH, youngwon?? gidarilso?? iqkeso?? Ya got me there!
gregm
3,411 posts
 
I tried the Norwegian (the originals, as per Tom) pipes and wasn't thrilled -- with EX-4's.
HOWEVER, what is spectacular: front loaded (tactrix horn) PM-4's driven by "good-bandwidth" 300B (i.e. using big Tango OPT) + an open baffle (sub)woof system.
The tactrix cuts mechanically @120Hz; the woofs x-over 4th order, passive, preceding woofer amp.

And, Tom (Twl) I also experienced PM4's on flat, open baffle with back damping and the same tactrix @ 120Hz. Special construction to support the PM4's.
Sound? Oh well... let me not waste server space with superlatives.

Unsurprisingly, measured response @ 1 m on axis was +3/-2 db to +19kHz. In room response was ~+/-8db which is outstanding in my book.

Now all that's left is money for two PM4's. Better still 4 or 6 to cover mutlichannel needs...

I'm rambling -- Mezmo, sorry.
gregm
3,411 posts
 
Hi Tom, Paul- Geez, I don't have the exact measurements handy and no pics yet (I'm a bit of a wet rag -- sorry).

The whole construction is on an open baffle:
The Lowther is wearing the wooden phase-plug.
The tractrix's flare frequency was calculated for 120Hz; mouth diametre just under 1m; throat is just larger than the Lowther's cone+suspension; horn's length under ~1m; presently (and cheapo-ly) the horn is made out of cardboard & I don't remember how many-sided it is (12 probably).

We plan on using suitably treated (i.e. "good-looking") plywood strips for the final, WAF approved, version...

The back of the cone is damped -- good point Tom. In fact, we tried it open baffle (hmmm), covered the lowther with some material (admittedly better), finally made a small rectangular box, damped the spider, no box damping (best - amazingly!). The point is we were getting interfering reflections from the Lowther's back on the open baffle scheme (we *were* close to the back wall, after all -- about 1m away only)

Now to the subwoof. Quite simply, there were 15" Beyma 400's lying around. With a total q in the upper 0,2's, these were unlikely candidates... but then, they were immediately and freely available for the experiment. Xmax looked OK at ~7mm.

We attached these to the baffle (~65cm width) -- one per side. We attached two 50cm sides to the baffle -- more for structural reasons than anything else and to simulate an H woof (open baffle).

Amazingly, this Beyma/Lowther combo works! The beyma is perhaps not the most versatile transducer around but "speed"/(phase) discrepancies are not subjectively perceptible. The B's BL factor of ~22 and a reasonable mms (just over 100gr, if I remember) probably help here.

However, marrying the dba's (horn w/ the woof) was not that easy.

Putting this together:
We matched ("equalised") the Beyma/P-L combo's db level at ~60 Hz to the horn's ~130Hz (the beymas were losing ~7db @60Hz vs 120Hz) & "flattened" it with the low-pass.

As you expect, the 300B drives the lowthers x-overless. Subwoof: we are taking output from the 300B, feeding that into a passive low-pass, then into an old 400W/Ch Phase Linear (P-L; for the young, that's an old manufacturer:)) which drives the Beyma coils directly. Strangely, the passive filter sounds better than an active one we tried.

Final task: align acoustic centres (i.e., bring the woofs forward). This will have to wait until we finalise the woofer units. If finances allow, the supravox 400-EXC with the electromagnet field coil. Makes for a very versatile driver (basically, one can adjust vas & qes->qts). Also, at $7-800 apop, it's an expensive driver.
Is dreaming allowed? TWO Supravox 400 per side in a W config...

A final point-- my personal contribution to this construction is practically nill; I just insisted on trying open-baffle and supported (enthousiastically) front-loading the Lowthers. This friend calculated the exponential (tractrix) horn by himself! That's when he was oblivious to the fact that there are spreadsheets around and that the horn he was working on answers to the name "tractrix"... the fact that he is a PhD in maths & has 25yrs diy experience helped, I suppose...

Cheers!
More to discover