6418 tube DI

Blue Jinn

Rider of the ARPocalypse
So, I had these microphone input transformers Mercury Magnetics L115M, 1:10, but they don't have shielding. I thought, OK use 10:1 and for a DI. I have about a dozen 6418 tubes. I found a couple of Dell 19v laptop PSUs in the ecycling bin at my office building. So, why not.

So far, I've optested with a Strat knockoff Yamaha, that has two single coil and one humbucker. I get distortion with the Yamaha at 10 on all pickups. (It's the "good" overdrive sound distortion, but it would be nice to have the option of not) But going below 10, I get the dreaded 2-8 hum! and so there's no volume control either. (I built an active FET buffer into a box to plug into right close to the guitar jack, and that cures the hum with just a small amount of "color")

The plate and screen resistor values I got from another website guitar boost pedal. I didn't want to have to use the buffer all the time, so I have some MXL 2:1 mic transformers leftover, so I tried sticking one of those on the front end that helped tame the distortion, but doesn't sound quite right and I lose the "good" distortion. The laptop PSUs are dirty as hell, so needed some filtering. Lose about 2 volts or so. Now note, my optest is into a vintage Shure M68 mixer to my old Boombox. But I don't think that's the distortion. I also stuck a simple transistor buffer to get a line level output and used the line in on the 'box. Anyway, here it is schematic and breadboard for your comments and criticisms and any suggestions for improvement.


  • IMG_4591.jpg
    503.3 KB · Views: 13
  • IMG_4590.jpg
    876 KB · Views: 13
Last edited:


Well-known member
G'day Bue! Geez, I have not worked with directly heated battery valve for around 55 years! I have often wondered what could be done with them in guitar circuits but not bothered to buy any to play with.

First of all, I am ot surprised you have uncontrolled distortion, as a pentode that valve will give massively too much gain, strapped as triode (link anode and G2 with ~100R) it will I am sure behave better. Also have a much lower output resistance to drive the transformer load which at best cannot be higher than 100k. Might even be more useful as a cathode follower?

"Biasing" is a bother, I can't remeber how it was done in battery valve radios, will look it up. I suspect 1k5 in the cathode is too high plus the filament volts need to be "the right way round" ref ground or the grid is positive WRT the filament/cathode!

Speaking of filament volts, that need to be UBER clean and I would use a Zener regulator.

I have other reservations but lack proper data as yet but will come back with a circuit and PMMWMMI!