Planning ahead for Trem

[Rich_S]

I'm planning a conversion of a Bogen PA amp into a "13-Watter" with ECL86's doing double duty as the output stage and the PI.

My question concerns the trem channel of the 18-Watter. Eventually, I want to build in the trem function, but since this is my first build, I figure I'll get it working without trem for now, then fuss with the trem later.

For the time being, I plan to leave out the phase-shift network, but how should I wire the trem side of the 12AX7? I figure I can just ground the grid, but what about the plate? What will give me operation closest to having the full trem circuit installed? In particular, what's the 220K plate resistor do to the preamp tube's operating point? Should I include the 220K in the initial build, or just use the 100K plate resistor, similar to the non-trem channel?

Can anyone explain the theory of the trem channel with regard to biasing, plate resistors, load lines, etc., or is the theory simply, "fool around with it 'til it works"?

[markh]

You can simply ground the grid of the trem side for now.

Leave in both the 100k and 220k plate resistors. The combination of the 220K and the varying plate resistance of the right triode causes the trem effect. When the trem is off, the right triode acts like a constant Z. Keep the 820 ohm cathode R to bias both triodes, but you can leave out the intensity pot for now.

The right side triode is a simple linear amplifier, nothing fancy in terms of load lines or biasing.

--mark

[zaphod_phil]

Hey Rich, as you probably know from the AX84 site I'm slowly building a similar '13W 18watter', using ECL86/6GW8s. Now I happen to love trem, as I'm into a lot of alternative/indy style music, and trem'ed distortion sounds great for that musical style. However, I don't particularly like the way the trem channel in the original 18W has been designed, so I'm planning to do it a little differently. I'm building the Lite II variant, and when that's working add a trem oscillator to inject the low frequency modulation signal somewhere in the vicinity of the PI or power tubes. There's a number of ways you can do it, either at the grids or the cathodes. I just haven't decided which I prefer. The amp will remain with one channel, with the dual-triode input, but will have a tremolo effect than can be switched on or off by means of a footswitch. The 18W pursists may shudder at this approach, but the result will be the gutsy sounding normal channel with the option of having switchable tremolo on it.

I'm planning a conversion of a Bogen PA amp into a "13-Watter" with ECL86's doing double duty as the output stage and the PI.

My question concerns the trem channel of the 18-Watter. Eventually, I want to build in the trem function, but since this is my first build, I figure I'll get it working without trem for now, then fuss with the trem later.

For the time being, I plan to leave out the phase-shift network, but how should I wire the trem side of the 12AX7? I figure I can just ground the grid, but what about the plate? What will give me operation closest to having the full trem circuit installed? In particular, what's the 220K plate resistor do to the preamp tube's operating point? Should I include the 220K in the initial build, or just use the 100K plate resistor, similar to the non-trem channel?

Can anyone explain the theory of the trem channel with regard to biasing, plate resistors, load lines, etc., or is the theory simply, "fool around with it 'til it works"?

[Rich_S]

Phil's handle is Zaphod, a reference to -> Zaphod Beeblebrox of Hitchiker's Guide to the Galaxy fame, by author -> Douglas Adams, who also wrote liner notes for the album Unsung, by his good friend -> Robbie McIntosh, who is an avid user of a -> Marshall 20-Watt guitar amplifier, which is the younger, bastard son of the Marshall 18-Watt guitar amplifier, to which this web site is dedicated.

Weird, huh?

I, too, love tremolo. I'm putting on the Bogen "13-Watter" conversion because the Bogen has a big chassis with plenty of room to experiment. I'm starting with the stock Marshall trem circuit because I have no experience on which to base another circuit. Who knows where it will end up, though.

I have another PA amp that's going to be a more "pure" 18-Watter. Just a simple Lite II version, since its chassis is too small for anything else. THAT amp will be my benchmark for evaluating my experiments on the Bogen.

[zaphod_phil]

Yup, you've got that Zaphod thing well figured, and even the tenuous link back to the Marshall 18W. ) And I live in Jim Marshall's part of London. One of my buddies used to be the Marshalls' next-door neighbor - they're garage was always full of guitar amps!

Anyway, there are a lot of different trem circuits out there, and it's worth looking at a few schematics to see all the different ways it can be done. For example you can just place the LFO signal on the input of the PI, at the same point as the guitar signal comes in. Or you can apply it across the cathode bias resistor of the output tube, provided you don't have an *excessively* large cap there. In fixed bias amps, it's neat to apply it the LFO signal to the bias voltage. Or you could apply it at the common-cathode(s) of the preamp stage. The stock 18W circuit, as originally designed Watkins, attempts to cut costs by doing the preamp and trem oscillator in the same tube and mixes them together in a wierd kind of way at the anodes and cathodes. Some like the trem effect you get that way - it's not bad. I prefer the gutsy parallel-triode preamp of the Lite II (which also uses the trem channel 'tone wafer'), with one of the more conventional trem schemes.

[zaphod_phil]

I think I've found a tremolo schematic I like, which could be added to the Lite II, from the AX84 Rail Rocket. It injects the tremolo into the grids of the power output tubes. So you get a kind of Normal Channel sound, but with the option of the tremolo's amplitude modulation. http://mhuss.com/ax84/media/ax84_m79.gif

[Rich_S]

Thanks for the heads up, Phil. I'll check it out, and maybe make it "Rev. 2" after the stock 18-Watter version.

On a related subject, check my new post at AX84 titled "EL84 (ECL86) at High Voltage". Any experience/insight?

[zaphod_phil]

Yeah, I saw it and responded. I was kind of surprised at Bogen running ECL86s with a 400V B+, but with everything plugged in and warmed up, the actual plate voltage may be somewhat less. Just make sure to bias within the 9W max plate dissipation. To protect the screens, I would recommend using Jason's 18W variation for the screen supply, discussed in the 'the PI Mods' thread on this board (http://www.18watt.com/modules.php?name= ... opic&t=203), where the single 100 ohm screen resistor is replaced by a 1k and then fed to each screen via individual 100 ohm resistors. The 1.5k/2W dropper resistor also gets increased to 2.2k/2W.

Thanks for the heads up, Phil. I'll check it out, and maybe make it "Rev. 2" after the stock 18-Watter version.

On a related subject, check my new post at AX84 titled "EL84 (ECL86) at High Voltage". Any experience/insight?