Beyond the Ariel


2003-02-04 12:23 am
-speaking of drivers for this design, here is the Test Bench of the Radian 745PB:

Test Bench: Radian Audio Engineering 745PB 1.4” Compression Driver | audioXpress

Other than the typical "hash" at the top of the pass-band, pretty good on a sub-par horn.

Still, I'd recommend the 18 Sound ND 1460 A or ND 1480 A .. and a good efficient ribbon around 6 kHz.

For the midbass section I'd pick the Faital Pro 12PR300 16 ohm x2 wired in parallel with a large volume bass-reflex and a much lower tuning freq. than typical.. (..sort of "sealed +") and a good filter for integration + baffle-step compensation.


Paid Member
2007-08-17 5:48 pm
Oakland, CA
If I’m reading you correctly you already have a pair of Ariels. Is it too complex a matter to just measure the DCR of the inductor currently in place?

Lynn talks about the Ariel here: Nutshell Hifi

He recommends using 10 or 12AWG air core inductors from North Creek. North Creek is no longer in business and their inductor data sheet/price list is gone as well. From my own experience, North Creek’s listed DCR was lower than the actual DCR of the product I received. For that reason I started buying the slightly less expensive 10AWG air core inductors from Solen. Parts Express & Madisound also carry 12AWG air core inductors.
Yes, I bought Ariels second hand, but crossover is mess with budget components and i need to rebuilt it. I cant afford 10 or 12 AWG coils with low DCR at the moment cause budget is tight, shipping to my country is really expensive and we dont have anyone making something like that here. Thx for answer, Brinkman
The DCR of the Ariel inductors are fairly non-critical. Anything less than 0.7 ohm is fine. My first choice would be the Solen Litz inductors, but can be expensive in the larger values. A good alternative are iron-core inductors with transformer-style laminations; I would avoid ferrite-cores because they are prone to sharp-onset saturation problems. Why not keep the inductors you already have? The worst that can happen if the DCR is too high is the bass will get kind of boomy. I do recommend good-quality caps, though, at least for the tweeter circuit.

As for the Beyond the Ariel project (the genesis of this thread), I regard Gary Dahl's build as the definitive version. ScottG's variant, described above, is an excellent alternative. I've heard nothing but good things about the 18Sound ND1480A (aluminum) compression driver, and if ScottG endorses the Faital Pro 12PR300 16 ohm x2 wired in parallel, that sounds promising. The basic concept is a 2-way system with quick decay characteristics (like the Ariel), smooth response in the HF section, high efficiency, and a fundamentally simple design.

If controlled directivity is more important to you, then other horn profiles are a (much) better choice. The HF crossover then gets more complex (in-band EQ is necessary, and possibly a notch filter), and some care has to be exercised to make sure that the time-decay characteristics remain favorable.
Last edited:
If cost is really a driver of the project, you can do a lot worse than track down a pair of Klipsch Cornwalls (cabinet in good condition, with all drivers good and a working crossover) and hang out on the Klipsch forums for info on upgraded crossovers and modernized drivers. My take is: this model was Paul W. Klipsch's flattest-response speaker (much more so than Heresies), and not only that, is very responsive to tweaks and modernizing.

The Heresy has a nasty response step ... the MF and HF drivers are nearly 5 dB hotter than the bass driver, while the Cornwall is close to flat (the 15" bass driver is a little bit more efficient, while the MF and HF drivers use a lower tap on the autoformers in the crossover). Upgrading the LaScala, Belle Klipsch, or a full-scale Klipschorn is a bigger job, since the folded bass horn has wacko response at the top of its range, and the MF horn is pushed to its limits to meet the upper part of the bass horn.

Taking on anything Altec is much more ambitious, involving serious crossover work and tracking down better horns than the usual 511 and 811's. The Altec 604 Duplex, although beautiful and a remarkable history in the US recording studios, is really really hard to equalize ... DSP is about the only method I can think of. By contrast, the phenolic-diaphragm Klipsch drivers are more tolerant of simple crossovers and out-of-band energy, and, by lucky coincidence, the MF horn and direct-radiator bass driver of the Cornwall are nearly perfectly time-aligned.

P.S. This mostly applies to enthusiasts in the USA; Klipsch isn't as common outside this country, and they've been made in Hope, Arkansas since the late Forties, so there are lots and lots of them around.
Last edited:
Updated schematics for the Karna Kay and Symmetric Reichert amplifiers. Voltages and currents are based on measurements of the Thom Mackris Reichert SE 300B and the existing Karna amplifiers.

Apologies for resurrecting an old thread. I don't know if anyone else ever actually built the Symmetric Reichert circuit? However, I have just completed the build on one of a pair of monoblocks based on it. I made some minor changes - the driver valves are triode connected 6V6's rather than half 6SN7's, both power supplies use 5AR4 rectifiers, and I upped the values of the coupling caps to 0.33uF and 0.47uF for the input -> driver and driver -> output interfaces, and also tweaked the grid to ground resistor values based on the results from a Spice model. CCS's on the 6V6's were from K&K.

The results, even based on listening to just one channel, are pretty sublime so I can definitely recommend trying it out. The clarity and transparency are extraordinary, seemingly even better than the OTL design I was using before.

Anyway, just putting this here in case anyone is interested.
Last edited:
I ended up with no input transformer, 6SN7 differential input LL1692A step down + another 6sn7 LL1692A step down. Both 6sn7 stages shunt VR on the centre tap of the IT. Tried direct coupled 6sn7 first stage and exotic Teflon capacitors but the IT step down sounded better (IMO). If using power tubes as drivers I found triode connected 6550 (Electro Harmonix) sounds better than the 6V6 triode connection.
Thanks for sharing your findings, I too am a fan of the 6V6 as a driver for the 300B. Could you please reference the drawing that Lynn is referring to?

Sure thing, here's the original schematic that Lynn shared in 2015:


This is a link to a PDF of the circuit as I built it:


2018-09-06 4:09 pm
Hmm, using the Vifa's in a tiny closed-box enclosure is a possibility. What surprised me is you have some spare drivers, since they went out of production about 10 years ago. I think I might have bought the very last ones that Madisound had.

The forum members probably have good suggestions for other drivers that have very smooth response, extended bandwidth to 5 kHz, and no peaks above that. Most small drivers get really rough between 5 and 10 kHz, and require an aggressive crossover to suppress the roughness and peaking.
Hey Lynn, thanks a lot for your suggestion. At time of writing I didn't knew that the Vifa's went out of production, so I was thinking I can easily get some and build enclosures around them. Unfortunately I couldn't get any, was even looking for used ones, but no luck at all.