• Dual channel amp
DescriptionThis amp has 3 parts: a preamp designed by Kevin O'Connor and published in his book "Tonnes of Tone," a Fender-style power amp, and a run of the mill power supply. I think a large part of the success of this amp for my ears is due to O'Connor's design.
Preamp
Two 12AX7 tubes are used to provide a clean and distorted channel. The clean channel has a tube stage -> tone controls -> tube stage configuration (sometimes noted as a 1+1 configuration), much like classic Fender amps . For the distortion channel, the signal is diverted after the first tube stage to run through 2 other tube stages and a different tone stack before returning to the first channel right after its tone stack. The signal then passes through the 2nd tube stage of the clean channel, giving a 3+1 configuration for the distorted channel. Where Kevin's design skills show up are in the choice of values for the resistors that limit the signal from one tube stage to another, and the choice of the coupling capacitors to keep the signal from getting muddy. Much of the low end is rolled off in the early stages of the distortion channel, and it gradually becomes more full-range.
Controls for the clean channel are: volume, treble, middle, and bass. For the distorted channel, there is drive, volume, treble, middle, and bass. A three position switch allows either channel to be selected, or both at the same time.
Power Amp
The output transformer is a 30w Hammond #1645. Primary impedance is 5000 ohms, with outputs of 16, 8, and 4 ohms available. Ultralinear taps are available on this transformer. I chose this transformer based on a formula I picked up in a thread on alt.guitar.amps:
impedance=(Va -Vsat) ^2 / Pa
where
Va = rectified voltage
Vsat = saturation voltage for tube (50v for 6L6)
Pa = power dissipation for tube
This comes out to 7056 ohms in my situation, but it seems like guitar amps typically use about 80% of the calculated value, which is close to 5K ohms.
Hammond transformer output wires are not simple taps for the speaker impedance, but are combined to provide the different impedances. 4 and 8 ohm impedances are easy to wire if you use separate output jacks, but switches are available to give all 3 impedances. I have this wired so there is a 4 ohm and 8 ohm output jack.
The schematics shows the screens using a tap off the power supply. Since building this amp, I've added a switch that allows for this kind of operation, or ultralinear operation. I was prejudiced that UL operation was just poor sounding silverface design, but then saw a schematic for a Dr. Z Route 66 that uses UL operation. I tried it, and I prefer it in this amp. The bass response sounds more balanced and tighter. (6/9/03 BTW, I don't have this schematic and it's been removed from the web site where I saw it.)
I originally had a 5K pot wired in for a negative feedback loop. I found that I only used very small amounts of NFB, mainly to get rid of some hiss. Since I really wasn't using this control, I removed it and used its chassis space for the UL/pentode switch for the screens.
Power Supply
This is modeled after a blackface Fender power amp with a long-tailed phase inverter and 2-6L6GC power tubes with 420v on the plates. The phase inverter uses 2 stages of a 12AX7 tube. This setup is similar to a Vibrolux, Bandmaster, Pro Reverb etc. amp. The main difference from these amps is that the Fender amps use fixed bias and have a negative feedback loop.
The amp started with cathode bias, because the power transformer had no bias tap, but I've since developed a fixed bias source and made the bias method switchable.
The power transformer is a 330-0-330v job out of an Epiphone amp. The choke is out of the same amp, and I don't know its inductance. Since the secondary only has two windings--6.3v and the high voltage, I chose to go with a solid state rectifier rather fight to get a tube rectifier in there.
Each tube stage has its own filter capacitor. I've read that the under-filtering of tweed Champs is what contributes noticeably to their sound and that later Marshalls are over-filtered to some, but don't have an opinion yet as to whether I agree. Maybe this amp is over-filtered.
The actual plate voltages on the preamp tubes range from 195v to 228v, and the phase inverter has 250v on its plates. I played with dropping these voltages by about 15%, but didn't hear any difference.