I just finished my Thermionic four Pi's today, after waiting quite a while for the horn flares to arrive. They sound pretty good on instrumental music, but I find that vocals exhibit a lot of the stereotypical 'horn' sound. My previous speakers were Magnepan MMG's, so is it possible that I'm just expecting too much in terms of neutrality, or could something else be wrong with the speakers? They also seem to be missing a bit of transparency and detail compared to the MMG's, but again, I might just be expecting too much. On a positive note, they do get quite loud and can be heard throughout my dorm, even when powered by my 55 watt Jolida 302A. The main reason I built them, though, was to have a neutral sounding high-efficiency speaker so that I could start experimenting with low powered SET's. From all the positive things I've read about Pi speakers, I think I might have done something wrong when I built the speakers. Any comments or suggestions?

Thanks.
Mike Borzcik

I used the standard Peavey CH-3 horn flare and built my own crossovers using the schematic you sent me. I just took some pictures of the crossovers, and I'll post them here as soon as the film is developed and I can scan them in. In the meantime, however, I'll try to give a detailed description of how I wired the crossovers, and maybe you can figure out what's wrong with it:

For starters, I have a 10 uf 250V capacitor connected directly across the terminals of the woofer. The negative lead from the woofer goes directly into the negative binding post. The positive lead connects to a 1.0 mH (or uH or whatever the schematic called for; I don't have a copy of the schematic right now) inductor, which in turn is connected to the positive binding post.

Also connected to the positive binding post is an 8.2 uF capacitor (the only deviation from the schematic that I know of--my sources didn't have 8.0 uF capacitors), which then connects to a 0.6 mH (or whatever...) inductor, the other end of which is connected to the negative binding post. The junction between the 8.2 uF capacitor and the 0.6 inductor is where the compensation network is connected. The compensation network consists of a 4.7 uF 450V Hovland Auricap capacitor connected in parallel with two 47 ohm Mills non-inductive resistors. The resistor in parallel give 23.5 ohms of resistance, which should be close enough to the 25 ohms called for in the schematic. In case it matters, the red lead of the capacitor is on the side nearest the positive binding post. The other end of the compensation network is connected to the positive terminal of the compression driver. The negative terminal of the compression driver connects to a 22 uF capacitor, which then connects to the negative binding post.

As far as I can tell, I've followed the schematic exactly. I'm hoping that someone can wade through the description I've given and maybe find something wrong with my assembly. The speakers are listenable right now, but at the moment they don't match the transparency, neutrality, or detail of my Magnepans.

Thanks for any help you can give!

Mike Borzcik

I'm starting to wonder if my amplifier is partly to blame. Now that I think about it, I had a little bit of the same problem with my Magnepans. They sounded awesome at low levels, but as soon as I pushed them louder everything kinda jumbled together. I was hoping that these new speakers would solve that problem by not making my amplifier work as hard to play loud, but maybe they're actually making the effect more pronounced for some reason. If we find that there's nothing wrong with my speakers, I might try to redo a triode modification that I tried on my amp when I was using the Magnepans. I didn't like the modification then because it made the sound seem weak and thin, but it might work better with these speakers.

From what I've read, I definitely believe that these speakers should sound better than they do now, so hopefully we're able to find something that I did wrong. I unfortunately can't send the crossovers themselves to you because the components are firmly glued to the inside of the cabinets. I should have the pictures of the crossovers in a few days, so I'll e-mail them to you as soon as I get them. Until then, I'll continue to enjoy these speakers at low levels, or at very high levels, forcing everyone in my dorm to listen to whatever I damn well want them to listen to. ;->

Thanks!
Mike Borzcik

Thank you, Wayne, for all the help and for designing such good speakers.

Mike