My ugly room treatments had to come down when we had the listening room painted recently. This left the system sounding too bright, so I experimented with different resistors in series with my MG-20 tweeters (actually, in series with the tweeters and their crossovers so as not to disturb the crossover points).
First step was to determine the amount of resistance needed. I made up some composite resistors from salvage store 2-watt units to get close to 2.0, 2.5, and 3.0 ohms after I found that 1.5 ohm power resistors did not reduce the tweeter level enough.
Six 15-ohm resistors in parallel gave 2.5 ohms, and this gave the best overall balance of cymbals and voices.
I then ordered a selection of 2.5 ohm resistors from Michael Percy:
Mills 12-watt MRB-12Ni
“NI” 50-watt (not branded);
Caddock 20-watt MP-821;
and I found some huge Dale 100-watt resistors at a local salvage store.
My home-made crossovers have provisions for alternative connections, so the smaller resistors could be inserted into the wire clamps. I made leads for the larger ones from Silver Audio Quad Litz wire left over from the speaker rewiring project.
The Caddock resistors come in transistor packages and have to be built up into something usable. I had a couple of large old heat sinks that I modified to take the Caddock resistors, and soldered leads onto them. The heat sinks were damped on about half the aluminum surface with Dynamat X-treme. The heat load is not an issue, but ringing in the heat sink will definitely fog the sound-stage and may add colorations.
The Caddocks are clearly far superior in timbre accuracy to any of the other resistors I tried. Their only drawback is the need for care in damping the heat sinks and placing them so they do not chatter on hard surfaces. This affects the sound-stage accuracy, and is obvious when it is wrong.
The Mills were the worst of the lot, and unlistenable compared to the composite junk resistors. The “NI” from Percy and the much larger Dale resistors were better than the composites, but not by much. They all added a glare to the treble that was obvious by its absence when I installed the Caddocks. This glare enhances some aspects of sound-stage, so care is needed in listening comparisons to avoid rejecting the smoother and more accurate Caddocks. The Caddocks are actually more revealing, but do not force the detail into your face the way the other resistors do.
I did not try 2.5-ohm cement power resistors similar to those supplied by Magnepan, so I cannot comment on them. If money permits, I may try some 100-watt Caddock resistors from Mouser: they come in 5-ohm size, and I could parallel two of them to get my needed resistance.
The system comprises a Wadia 861 CD player driving a pair of home-made Atma-Sphere MA-1 monoblocks for the midrange and treble, and an InnerSound ESL-300 for the bass panels. The stock bass/everything else outboard crossovers are used as low-pass filters for the bass panels only. The high pass for the midrange is built in to the inputs of the MA-1s. The midrange and tweeter crossovers are built to mimic the stock contours but with higher quality units. Silver Audio Silver Sympony speaker cables are used throughout. These are damped with Power Wraps and loaded with R-C networks to damp internal RF ringing. The wire to the midrange and tweeter drivers is Silver Audio Quad Litz, and allowed to dangle outside the speakers.
This post is made possible by the generous support of people like you and our sponsors:
Topic - Magnepan tweeter resistor comparison results (long). - Al Sekela 18:50:50 08/23/04 (1)
- made the same experience, cooling fins - Arbelos 08:35:05 08/24/04 (0)
You can not post to an archived thread.