Another generous contribution by Sebastian, this time a Sovtek MIG 100H amp emulator!
The Acorn MIGFET post inspired me to whip up this layout for a
version of the MiG 100H, which is more like a hot-rodded JCM800 than the
pseudo-Plexi MiG 50. An interesting variation in its own right.
I've
not built this myself yet, since I'm out of 100k trimmers and A1M pots,
so due diligence is encouraged. I've attached the amp schematic for
scrutiny. The layout follows the circuit pretty much 1-to-1 until the
phase inverter. One half of the phase inverter is used, then the output
stage is pilfered from the Menatone Foxy Brown (I've found it great in
various AIAB builds, so it should do the trick here too). The master
volume is moved to the end of the circuit.
Like
the JCM800, the MiG 100H has two channels with separate inputs, the
lower-gain of which skips the first stage of the preamp. This layout has
a footswitch to toggle between the two channels for (hopefully) added
utility.
.png)
For reference, the schematic I was working from is posted on Schematic Heaven. The biggest difference between this and the JCM800 is that the tone stack is plate-fed, as opposed to cathode follower, which does have a significant effect on the curve of the tone stack.
ReplyDeleteI'm curious about the channel switch. I read once that jumping the first stage you lose a lot of volume.
DeleteThe channel switch is a bit of a speculative jump, since it's a direct lift from the amp schematic and I've not actually played a MiG 100. But the MiG 100 is, more or less, a JCM800 clone. (Getting into amp history here) Old 4-jack Marshall amps had separate treble & bass tube input stages (where the tone-shaping occurred to distinguish treble from bass), each with their own volume control. These channels then intersected and passed through 2 more tube gain stages before the tone stack. So the signal would pass through a total of 3 tubes before the tone stack. JCM800s were designed to be higher gain, and put those first 2 input tubes *in series*, so the signal would pass through 4 tubes before the tone stack. The "low gain" channel just skips the first tube (like the signal path in older Marshalls). So the difference in volume is really about the difference in gain.
Deletebuilt it but having some issues...
ReplyDeleteworth noting: i was using the old layout. but luckily, i caught the problem with the Q3 trim pot and was able to fix it on the fly using a similar solution to what's on the revised layout. also noticed noticed is the source resistor on Q5 needs to extend down one more row to reach ground (at L19). easy enough fix.
i'm using j201s since i don't have any of 245As or 4393s, and i saw some people saying they'd had luck subbing those in other circuits. and so far they seem good to me.
right now, i'm getting a sputtery, almost gated fuzz sound from the output. it's also pretty quiet, just barely above a bypassed level with the master dimed. presence knob isn't working (much needed in this circuit). and i can't get Q6 to bias properly, stuck at 7.2v at the lowest setting.
i poked around a bit with my audio probe, and i'm getting a really great sound (with decent volume) all the way to the gate of Q6. but it's falling apart after that.
tbh, my knowledge of circuits is still fairly rudimentary. i'm able to follow along well enough on the original schematic, but i'm not sure how to troubleshoot much beyond what i've done so far. i'm guessing the problem with both the presence not working and Q6 not biasing might be related? but like i said i'm out of my depth now.
would appreciate anyone's thoughts on what to try next. thanks!
Found the problem: the 220k resistor from D8-I8. It should be F8-I8. That's killing the drain voltage on Q6. Glad the probe shows strong signal until the gate of Q6, because that means it should be working fine once that resistor is shifted. And thanks for spotting Q5 source resistor! I'll revise the layout and get it to Anders ASAP.
DeleteFailing that, try moving the 2n2 cap over to D8-G8 and run the 220k resistor from B10-I10. That's what I've done in the revised layout that I just shipped to the good sir Fuzzhead.
Deletethanks for the quick reply, sebastian!
Deleteand thank you for the excellent layout!!
can't believe i didn't notice that 220k. that's exactly the kind of stuff i'm usually good at spotting 🤦🏻♂️
made all the changes you suggested. i was able to bias Q6, but it was still sputtering so i ended up tuning Q6 and Q5 by ear (on the clean/crunch channel). sounds pretty great now.
still not getting anything from the presence control tho. probing around some more i did notice that it's having an effect on the other side of the 82k (at D13) although not what i'd describe as typical presence function. so it's doing something somewhere, just not propagating to the output.
Okay, try moving the 470R resistor *back* to D19-H19. Looking again, I realized that it's part of the presence circuit. That would also explain why you were having trouble biasing those last two JFETs.
DeleteThe lesson here is I really ought to draw up a proper schematic before starting a layout, instead of just winging it from the amp schematic, LOL.
lol, i feel you. i was in the process of reverse-engineering a schematic. i have a tough time looking at layouts and wrapping my head around what exactly is going on, but schematics i can grok.
Deletei'll give that a shot and report back tomorrow. thanks again!
Also! There's a cut at G19 hiding under the 470R resistor.
Deletemoved the 470r back but still no presence control :(
DeleteDamn, Brain, I'm sorry you get to be the guinea pig for a really slapdash effort! Two last things I can suggest:
Delete1) my dumb ass mislabeled the presence as 25k when it should be 5k. Maybe the control IS working but it's all whizzing by at the beginning of the rotation.
2) failing that, yank the 2n2 cap (D8-G8) and 82k resistor (D13-G13) and see if that brings any life to the control.
Beyond that, I've drawn up a schematic and sent it to Anders so y'all can peruse it for any further flubs en route to the vero layout.
don't mind being the guinea pig at all :) for the most part, i enjoy the troubleshooting and almost always learn something in the process. plus, i think this circuit will be worth it in the end.
Deleteafter trying your suggestions above and not getting any results, i took a little break and decided to breadboard those last two stages based on the schematic you uploaded. the presence control is kinda subtle, but definitely working. tried a 20k trim which made it more noticeable (and is preferable, imo).
i'm about 99.99999% sure there aren't any errors in this version of the layout, so i must've screwed something up in my build. i've gone point by point with a DMM, checked things out with a magnifying glass, but for the life of me i can't find it. i'm gonna take another little break from it and come back with fresh eyes again. might even start over with a fresh board.
hopefully we can get this sucker verified soon :D
just as i suspected... build error. i re-soldered everything from row H and up. now it's working and officially verified :)
Deletethanks again for the layout and all your help, sebastian!
this thing sound sweet.
Fantastic! I salute your perseverance, and thank YOU for having seen it through to the end. Very glad to know it was worth all the trouble, ha ha ha!
DeleteAhoy, Fuzzhead, just a reminder that, after much patience and back-&-forth, brain verified this layout!
ReplyDelete