• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

C3M 300B SET build, input appreciated.

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
After a lot of research I've arrived at the following complete schematic for my 300B SET build. Any thoughts would be greatly appreciated:


Most of the iron has already been purchased: Thomas Mayer custom PT, Lundahl 1623 OPTs, and Lundahl 1673 10H/200mA chokes. PSU and signal chassis will be separated. All filaments will be DC (except the rectifier bridge). I plan on using ASC X386S motor run caps throughout with Obliggato and Elna Silmic IIs on the C3M cathodes, and an Angela/Jensen for the coupling caps. Resistors will be Mills and Vishay non-inductives with Kiwame/KOA SPR for the 2-watters.

My questions are these:
  1. Will channel separation be sufficient in this arrangement?
  2. The Lundahl 1673 chokes are rated at 200 mA with saturation at 290 mA. How far into this range is it safe to operate? I'd like to use OA2 regulators on the C3Ms but I'm already drawing 200-210 mA from the PSU.
  3. Are the ASC X386S caps from Allied Electronics acceptable? I've heard some question whether the current US made units are as good as the older Japanese units.
Joined 2004
Paid Member
Stereo separation should be fine, construction technique will determine how good separation is at all but the lowest frequencies where some cross-talk through the supply could be significant. Typically not much separation in most program material below a few hundred Hz.

You are at the limit of your chokes. I would not operate them beyond the Lundahl recommended operating conditions.

I have used U.S. ASC X386S series caps in an amp I built a decade ago, they seemed completely fine to me at the time.
Before L1 You can connect cap to ground about 20-30uF. Ripples decrease about 2 times on it.

Axazello: the choke input was deliberately chosen to reduce inrush current, provide a more stable current supply, and to put less stress on the mains transformer. A lot of people believe they simply sound better, too. Modeled in PSUDII I get 8.6mA of ripple at the second LC, and 0.3mA to the driver.

Mach1: Absolutely, I would love to use gas regulators; however, I'm already drawing 200-210mA from the PSU as is, and the Lundahl 1673s are "recommended" for 200mA only. This is where I wasn't sure how far you can run the Lundahls into their 290mA saturation rating and what consequences you experience in between. Any input on this would very helpful.
Regulate the C3m screen with gas regs connected between g2 and the top of the C3m cathode resistor Rk. You can then reduce Rk substantially and dispense with the bypass caps. Win-win.

My main concern is the noise associated with gas regulators. Have you experienced this or is it rare? The remaining option is to use zeners, such as one 1N5383B (150V, 5W) or smaller zeners in series (e.g., 1N5377B + 1N5371B); I'm just not well read on how substantial the current draw or noise factor will be in comparison to the VR tubes.
I doubt if the circuit would work properly as drawn. The 0A2 needs an 8-10k voltage dropper resistor or a ccs dialled in to around 30-35mA.

That puts current draw around 24-31mA for each OA2 with those current limiting resistors—does it need to be that high? If that's the case then I may have to forego regulation all together, I'd be creeping uncomfortably high into the saturation rating of my chokes. What about the use of solid state zeners?
Last edited:
I thought the 0A2 was an octal gas reg which generally work best at around 30mA - just checked and it's actually a noval with an optimum current draw of 17.5 mA. You could probably drop that to around to around 12mA, and allowing an additional 4mA for the g2 current, means around 16mA (provided by an 18k voltage dropper) will be required in addition to the plate current.

You could probably save around 6-8 mA/channel by using zeners instead.
Last edited:
Your schematic kinda looks like something from Thorsten Loesch. :)

Well... I think your 220k input shunt will be a bit high. If you think your HF is rolling off too much due to input Cmiller then change this to 100k or maybe even a bit lower.

I hate gas regs for this duty.... they are noisy, subject to variation.. I always found them to be a bit of a pain. Maybe you will make a similar discovery. I hope not. I found Zeners much easier to implement and (with a little bit of selection) significantly quieter.

Anyway, the C3M is a very special valve. I suspect you will be pleasantly surprised.
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.