Magnetron Issue.

[farwest1]

The Magnetron works well and sounds amazing. But it seems like something's not quite right. Maybe I got a component or resistor wrong?

I have to max out the Level pot for the wet repeats to be loud enough, and even then they're slightly quieter than the dry guitar signal. If I set the level around 12 o'clock, the wet repeats are really quiet, almost just a texture. It sounds good, ambient, but not quite what I expected. I guess I expected a stronger "wet" signal.

Any thoughts? Did I get something wrong in the circuit?

I madse two mods to the circuit: #1 I put the gain trimmer on the face, which was a good move. It's super warm and decaying—very Boards of Canada like. #2 I changed the Feedback pot from logarithmic to linear, also good, but self-oscillation happens now at around 3 o'clock rather than at max.

(If you're interested, the name comes from a weird short story called "Academy Leader" by cyberpunk writer William Gibson. I was thinking about some sort of glitchy animal that keeps reappearing in lower and lower fidelity—and a line from that story kept coming back to me:
"in the hour of the halogen wolves, in radio silence.")

 

[zgrav]

very nice enclosure!

 

[farwest1]

Here's what a user on Reddit recommended:
"Play with the value of the resistor in the feedback loop, and use a linear pot instead of a Logarithmic pot. To boost the volume of the wet path, try an 18k or 20k resistor instead of a 15k in the pt2399 circuit."

 

[farwest1]

Here's what a user on Reddit recommended:
"Play with the value of the resistor in the feedback loop, and use a linear pot instead of a Logarithmic pot. To boost the volume of the wet path, try an 18k or 20k resistor instead of a 15k in the pt2399 circuit."

I replaced the Feedback logarithmic pot with a linear pot (A100K to B100K). I like the sweep better now, but I have to be really careful not to go into self-oscillation.

I guess I'm wondering how this circuit could me modified to allow a "full wet" version of the repeats. Any ideas?

 

[phi1]

Did you try increasing R12 (described in the reddit post)? Another idea may be to reduce R13 to say 1k (I’m not sure if this would have the desired effect). Both of these resistance changes would impact high cut filters, so you’d want to scale the capacitors conversely to keep the exact same filter.

I haven’t built this, but you could try checking videos or reviews of the Keeley to see if yours is behaving like the original, otherwise maybe there’s some small fault in your build. Or check who else has built one on the build reports section, and message them directly.

 

[farwest1]

One other note: I hear a pulse-clicking noise when I turn the level up. Can't decide if it's the LFO rate that I'm hearing, or the delay time clock.

I did use a TL074 instead of a OPA2134 (OPA wasn't available through Tayda.) Some people say the OPA is far clearer than the TL074, so maybe it was a mistake to go with the cheaper op-amp. It's socketed, so I'll get the OPA and test.

 

[phi1]

Sounds to me like LFO ticking. Unfortunately I don’t know of a fix on that. Might have to sos pedalpcb and chuck d bones. Although first I’d double/triple check your build since the behavior seems weird as it is.

The TL072 substitution wouldn’t cause any of the issues you’re describing (it’s unclear to me if it would make any difference at all).