5E3 RobRob Deluxe gain cascading problem

[St4rP0wd3r]

Hi Guys, I buit a 5E3 RobRob deluxe following Rob's istruction and everithyng worked just fine. However, I found the lead channel kind of useless to me as it is very similar to the normal channel, and decided to implement the gain cascading mod, keeping all other spects the same as the RobRob Deluxe.

Something however is not working fine, and I can only guess it is related to the previous master volume pot, or the V1A different cathode bypass specs. My gain chanel is messed up, and both gain pot (previous lead volume) and master volume are behaving strangely whreas the nornmal channel is still just fine.

I also thought it could be relates to the 680k resistor between stages. I didn't have 680k, so I used 470K which resulted not good, and 1M which got wors, with plenty of screetching noise and no sound. I then tried bridging the stages directly, with no resistor, and it got better but still the gain pot behaved very strangely.

Does anyone have any experience on implementing a cascade gain on a 5E3 with master volume or any insight that could be helpful?

Thanks in advance!

[sluckey]

Is this the schematic you followed? Did you separate V1s cathodes as instructed?


Layout and wire routing become more critical once you cascade the preamp triodes. Show us a hi-rez pic of your actual amp.

[St4rP0wd3r]

Is this the schematic you followed? Did you separate V1s cathodes as instructed?


Layout and wire routing become more critical once you cascade the preamp triodes. Show us a hi-rez pic of your actual amp.

Yes, on the RobRob Deluxe the Lead Channel already has a separate cathode bypass, however with a 2.7K resistor and a 0.68uF cap in contrast to the 1.5K and 22uF of the normal channel. So my cathodes are separated.

So I'm already not starting from a standard 5E3, and just adapted his schematic to cascade the lead and normal channels usingthe RobRob original specs.

[mensur87]

The resistor values sound way out of spec. The 470K could be too little and the 1M too much, which damages our gain. Come on in and bring it closer to 680K Perhaps also the wiring and cathode bypass caps, they also should sound off.

[johnnyreece]

Interesting take. I recently built a tweed Tremolux-ish amp using this preamp, and I found a very audible difference between the Normal and Lead channels.

[St4rP0wd3r]

So I found the biggest problem, and it was just a shorted switchcraft jack.

I replaced it and now the gain channel is "mostly" working. The resistor on the gain pot wiper seems to be intended to "tame" a bit the gain before the second gain stage.

I tried 470k, 470k+220k, and 1M. The later produced the best results but I'm still having screeching sound when normal volume is near max. That only happens when using the gain channel and all behaviours are perfect when using the normal channel.

I still believe this is something related with excessive gain, but will live with it for a while, keeping volume knob below 80% when using the gain channel.

[johnnyreece]

I'm able to dime mine without it screaming, although I believe that's due to the way my chassis is designed, not so much any talent I may have. Got any pics? I'll show ya mine if you show me yours! You can see on my pic, the jacks are right by the tube socket (minimal signal wire runs).

IMG_6828.jpg

[GlideOn]

So I found the biggest problem, and it was just a shorted switchcraft jack.

I replaced it and now the gain channel is "mostly" working. The resistor on the gain pot wiper seems to be intended to "tame" a bit the gain before the second gain stage.

I tried 470k, 470k+220k, and 1M. The later produced the best results but I'm still having screeching sound when normal volume is near max. That only happens when using the gain channel and all behaviours are perfect when using the normal channel.

I still believe this is something related with excessive gain, but will live with it for a while, keeping volume knob below 80% when using the gain channel.

I had the same initial issue on my own build - drove myself crazy as everything in the circuit checked out, but there was no signal getting through anywhere. Brought in to my tech. It was a tiny blob of solder from my input jack shorting to chassis, literally tech and I shook it loose and it freed up. We put in a Switchcraft 12 instead as the existing cheapo Cliffjack which was kind of a tight fit there the begin with and feels like a more secure connection personally.

I've done my own cascade switching schemes which uses a 4PDT and series in a coupling cap to make the cascade sound tighter, but Rob's is great too.

I know yours is no Marshall, but those JMPs and JCM800s are also famously cascaded and it helps to study those on Rob's website as some of those tricks can apply here. Doing the "One Wire Mod" I've found I can go as low as 100k with the series grid stopper after the gain pot before it gets too crazy. Remember that a Grid Stopper or any resistor in series with the signal retards all frequencies. The lower you go, the more signal you retain and more gain you get, warts and all. So why not send some of the highs through using a cap in parallel with said resistor?

You can try the JCM800 "peaker" where you have the grid stopper, but put a cap over the resistor. Values such as 470k/470p sound classic for rock, but also try 220k/330p which is a little less aggressive and smoother sounding, all while emphasizing highs a bit more. I did the latter of which in a plate-fed Bassman circuit and it sounds like a friendlier, "Fendery-er" Marshall.

Technically you need to block out more of the lows from getting through in a Cascaded setup to prevent blocking distortion in later stages, less so of the highs so don't be afraid to experiment with some resistor and cap values as it's a custom amp anyways.

Also, lead dress, length of wire and using shielded wire for the inputs starts to factor in with cascaded gain stages to mitigate oscillation. Highly recommend shielded wire from your input jack to the first stage grid with series grid resistor of 33k or 68k directly onto the pin. Use a terminal strip bolted on the screw used to hold the tube socket in place if you need to mount it all.

Next step is trying snubber caps of 100pf or more from Plate-to-Cathode on input side, Plate-to-plate works too. Marshall started to do that mid 70s to prevent theirs from oscillating.