Simple Tweaks
Janos
easy to do.
Janos
This is a very important issue, and for low-powered SET amps seems to
be critical - at least what I have experienced with my Darling & Voigt
pipes.
The mod:
Adding a non-inductive 8ohms resistor in series to the drivers.
(Purchased at Radio Shack, then cryogenically treated).
Theory:
Changes the impedance that the amp sees. The OPT of my Sweetheart (the
Darling amp) does not exactly match the parallel 1626s output impedance
to the 8 ohms (around 7.5ohms to be precise for around 40Hz-20kHz) of
the FE204 driver of the Voigt pipes; it matches to around 9.6 ohms. The
fact, that the impedance of the 8ohms resistor is stable in the entire
frequency range, and that the amp sees an easier load, especially at
the frequency extremes, improves the performance. The drawback is that
half of the power is "wasted" on the 8ohms resistor, and the driver
sees roughly 0.5W.
Result:
I expected a sharp drop in volume level handling, as instead of the
already low 1W the drivers are driven by 0.5W per channel. But the
sound became only a bit quieter; the change in volume level is less
than removing a Z-sleeve from the IC between amp and preamp. So, watts
are not the only thing.... actually, during loud passages on a chamber
orchestra I get only 0.2V as the highest reading on the multimeter on
the speaker binding post. R (8 ohms) = U / I ; P = U*I, so that's 5mW
only. But I have efficient speakers, and no crossovers.
There is a stunning improvement in the mid-base. The Sweetheart -
Voigt pipe combination has always been a bit midbase shy. (Stereo 70
- Voigt pipe: very strong midbase in comparison.) I did not mind this
"lack" of midbase, as the existing, though weaker midbase was very
clean and precise, and closer to natural sound than the ubiquitous
bloated midbase. However, this combo did not "rock" on most
material, as that genre depends on inflated mid-base.
The surprise was that the whole sonic spectrum became cleaner, more
alive, and more defined and detached in space. The soundstage became
much deeper, I'd say twice as deep.
Overall:
Best change in my system I ever heard. If you have a SET, try a small,
non-inductive resistor in series with the speakers. It also should work
on multi-driver setups to some (or even greater) extent, as the
crossover can cause very big dips in impedance. Even a relatively
benign, 6ohms speaker, like Linn Olson's Ariels, has a big dip to 2ohms
somewhere in the frequency range, if I remember correctly.
You'll have to toy with the optimal value of the resistor, between 1
ohm to 20ohms, depending on the amp and the impedance of the driver,
the only thing is to use non-inductive resistors.
Janos
I've made a listening test comparing the following:
1. untreated CD
2. demagnetized CD player loading tray
3. demagnetized CD p. tray and demagnetized CD
4. demagnetized RCA conncetions, and speaker binding posts.
Used a small tape deck head demagnetizer, and listening material was
Sinead O'Connors "jackie" from the Lion and the Cobra album, Chrysalis
Records. It is rock music, and very thoroughly composed. I never
imagined her records were so good until I heard them on a good setup.
1. Nice sound, but her voice has a clear plastic taste. I had the
impression that the recording engineer did something wrong with it.
2. Somewhat better sound, and her voice has definitively lost its
plastic-quality. The recording engineer did a fine job, after all.
3. Wow! Instruments have their bite, shine, life, and her voice is to
die for. Very sweet, emotional, yet powerful, dynamic, and articulate.
4. Disapointment. The system sounds as if all my wire was not broken
in! Base lost, weak sound, thin highs, no power, no dynamics, no
harmonic richess. Quite stunning change, a plunge from the skies to the
abyss.
Seems that demagnetizing disrupts the fine atomic alignment of the
contact surfaces... I hope this effect passes soon!
Janos
somewhat greater than simply disconnecting and then re-connecting
cables.
Kyle
full-range drivers at:
http://www.firstwatt.com/images/current_amps/cs-amps-speakers.pdf
He thinks that high efficiency full-range speakers like Lowther and
Fostex are overdamped and need higher output impedance / lower damping
factor to allow the bass range to bloom and provide better overall
balance. This could be what you are hearing. I think increasing the
load impedance to the opt secondaries by adding series resistance will
increase the primary impedance, similar to going from the white
secondaries (2.5k:8) to the yellow (5k:8). Usually results in a little
less distortion, but you also lose a little power, so in addition to
the power lost as heat through the resistor, there will be an
additional reduction in power output by raising the primary impedance.
Damping factor will increase due to the higher primary impedance, but
will decrease by the series resistance to the speaker. I tried the
5k:8 setup and noticed increased depth of the soundstage, but
everything sounded pushed back. The 2.5k:8 setup places vocalists and
lead instruments clearly in front of the other musicians and I guess I
prefer it that way. Sounds punchier, too.
I'm not sure this would work with multi-way speaker systems. Higher
output impedance can create problems with crossovers, as it did with
Ted's 6EM7 amp used with Greg's HE-10.1's. A little feedback to lower
the output impedance made a world of difference in that case. I guess
you would have to try it and see.
BTW, most multimeters are pretty bad at measuring AC unless you stay
pretty close to line frequency. Usual frequency response is aorund
50hz to 1khz.
Kyle