Lil' Bumble-B - 'micro' Bassman with OD channel

[Ten Over]

This is another point. 4x 12AX7 plus one ECC99 make 3.2 A.

I get 2.0A

[roberto]

My fault, you are right on this point.
But again, no safety margin is not a good idea, especially for heaters that are cold on startup, so their resistance is lower.

[romberg]

In addition to all the great suggestions already posted, I'd take a look at the entrance to the overdrive stages. The ODS has a trimmer and voltage divider here that dumps like 90 to 99% of the signal before the two overdrive stages work with it. You might wanna do something similar before your "lead gain" pot.

When an ODS clips, it usually only clips 1/2 of the wave. The other half still looks sinusoidal. If you hit these two stages with too large a signal, then you will probably clip both sides and get square waves.

Take a look at a few ODS schematics in the Dumble section. Those input trimmers really do cut the signal by a huge amount. When I built my ODS I was thinking I'd need to modify this voltage divider a bit to make the signal stronger. But I was wrong . It works really well just as Dumble did it.

Mike

EDIT:
Just looked at the ODS schematic I built from. There is a 470k - 200k - 25k pot - 4k7 voltage divider network leading into the overdrive stages. The 470k is bypassed by a 47p pot. So you can see that this cuts the signal (of the lows not bypassed by the cap) to between 4% and 0.6% of the clean channels output. The highs get bypassed around that 470k so they only get cut down to between 12% and 2%. In either case, it is significant.

The other thing I'd do is move that "lead gain" pot to be after the first overdrive stage and not before it. This is how it is done in an ODS.

[Lynxtrap]

Thanks for the advice so far!

And stop worrying about the power supply, there is nothing to see there, move along
I see that there is a line looking like a centertap sticking out there, a leftover from the Robinette file that I forgot to delete.
AX7 and AY7 each draw a filament current of 0.3A, the ECC99 draws 0.8 = 2A total.
The output stage is not "soft", it's not like that ECC99 loads down the PSU.

I will surely look into the suggestions already given here! Some of them I had already planned, like dumping way more gain before V2A. I will also try to cut both lows and highs to make it more mid focused.

A few things might be worth mentioning about the design:
The OD channel is meant to work as an independent channel, as mentioned. BUT, there is also the option to use it as a 4 stage + CF high gain amp, using the TMB stack while bypassing the OD Tone control.
But my main concern now is getting decent sound out of V2A and V2B, since I don't want to mess much with the Bassman circuit.

Another thing: this is built in an Marshall 18W chassi on an 18W turret board. Every turret and hole is occupied in there, so the possibility to add more components is quite limited.

[johnnyreece]

It appears that the CT is unconnected, and it's wired as a bridge rectifier, which would take .707(ish) times the total voltage (400), which would make it 283-ish volts, unloaded. Feel free to correct me if I'm wrong.

You are indeed. Try with this if you do not believe me:
https://www.falstad.com/circuit/e-fullrectf.html

Vmax is what we call Vp.

Thank you. I had it backwards.

[Lynxtrap]

A couple of updates and some weird experiences.

When I turned it on tonight it seems like the noise I had from the start has gotten worse. It's the "rrrrrrr" kind of noise, like when your gutar needs shielding or something. I tested continuity from all groundpoints and that seemed fine.
I've also gone over it with the chopstick. Grid resistor on the input to V1 was microphonic, so I redid the wiring with a new resistor. No change.

It seems like the amp goes silent when I break the signal flow at any point downstreams of V1. But pulling V1 does not help.

Moved the PSU grounding from the safety ground lug and grounded it with the rest of the circuit. Same thing.

Two other weird things. I was going to connect the NFB and put it on both + and - on the jack to find out where it would howl. But I just couldn't get any positive feedback going. How is this possible?

Last weird thing: I played the amp through a very different speaker this time, Eminence Texas Heat. The OD channel sounde much more as I had envisioned it, almost like i could use a little more highs...

So, something funny is going on here, and the noise has first priority to be sorted out.

[10thTx]

I apologize for being late to the discussion. Yrs ago, I wanted sort of a bassman/marshall preamp topology going into an overdrive channel. So, I made the D'Mars ODS. I built several of them and really liked the amp alot. I thought it had a reasonably transparent tone.

I don't know why this approach couldn't be used with the "micro-bassman" power amp topology. One could also use ECL84 for about 3-4 watts (search So-Low Watt) or 6BM8 tubes for around 7-8 watts.

I might add that I have tried inserting an OD channel between stage 1 & 2 of a Marshallish topology without success which is why I changed it to the OD channel after the preamp.

I've attached how I approached the idea for it. Hoffman amps sells a layout board for this.

Here is a sound clip of the amp. https://soundclick.com/r/s71th8 and another soundclip https://soundclick.com/r/s71g2i

With respect, 10thtx

[10thTx]

Here's how I would probably approach the idea ................. not saying this is "better" simply stating this would be my thinking on it based on what's worked for me in the past.

with respect, 10thtx

[roberto]

If you want to add a SS device, you can also use it as load for the second triode in gyrator configuration, and then go to the EQ, instead of adding it as source follower.
On top of that, I'd use something more recent instead of the IRF820 that goes more into the Mosfet Follies territory (quite outdated nowadays).

[10thTx]

if you want to add a SS device, you can also use it as load for the second triode in gyrator configuration, and then go to the EQ, instead of adding it as source follower.

I'm not familiar with what you're proposing. Can you kindly provide a schematic? So far, I've been pleased with using the mosfet as a cathode follower and it's been successful.

With respect, 10thtx

[Lynxtrap]

Here's how I would probably approach the idea ................. not saying this is "better" simply stating this would be my thinking on it based on what's worked for me in the past.

with respect, 10thtx

Thanks a lot for input and taking the time, interesting food for thought there! And closer to an ODS topology. I will absolutely take a closer look at that, and certainly give it a try if I don't get what I want out of the config I have now.

First I have to sort out the noise issue and the almost miraculous change of sound I experienced yesterday when playing through a different cab. Something funny is going on and I want to find out what it is before making any further changes to the circuit.

As for the noise, I'm more used to dealing with "hmmmmm" and "ssssss" than with "rrrrrrrr"

[Lynxtrap]

I thought an update on this might be due, I am very close to the finish line.
Been tweaking the OD channel and the output stage.

There is some ridiculous voltage dividing in the schematic, I'm waiting for delivery of 100k audio pots for the OD Gain and OD Level and hope that those will do, so ignore all those resistors...

Might put a bright cap on the gain pot, and change the 1uF cathode bypass to something like 3.3uF on V2A.
There is a little lack of bass in the OD channel.

I don't find much use for NFB here, so I'm going to use the pot for a Vox style tone cut instead.

The clean channel is what it is, and the OD channel can do everything from Keef-like tweed breakup to a mid-focused ODS-type of thing, and scooped metal or SLO style sounds with the right amount of OD Level to overdrive the stages downstreams.

If I may say so myself, I would probably buy this amp if someone presented it to me

I'm almost starting to think that it's a pity it only puts out about 2.5W (rough estimation).

Plate dissipation on the ECC99 is 3.47 and 3.28W. Limiting value is 3.5W according to the datasheet.

As always, any comments and suggestions are very welcome!

Lil Bumble-B_rev7.png

[roberto]

I'm not familiar with what you're proposing. Can you kindly provide a schematic? So far, I've been pleased with using the mosfet as a cathode follower and it's been successful.

I just noticed this post, because you didn't quote it was me writing. The first that proposed this solution has been (to my knowledge) Gary Pimm to be able to have the gain of the pentode and the low output impedance of the triode, to directly drive DHTs with one single gain stage before. We can use the low output impedance to drive tonestacks. Another interesting solid state load for the pentode is the CCS, but it needs to be thamed otherwise you will run into oscillations, as the gain could be close to infinite. It was used in old medical stuff from the 50s where high gain was needed, and a triode as CCS on top of a pentode has way more gain than cascading the two.

[10thTx]

Roberto,

Thanks for taking the time to respond and provide more information. I appreciate the effort.

With respect, 10thtx

[10thTx]

The clean channel is what it is, and the OD channel can do everything from Keef-like tweed breakup to a mid-focused ODS-type of thing, and scooped metal or SLO style sounds with the right amount of OD Level to overdrive the stages downstreams.

I would love to hear some soundclips if that's convenient and possible! I've got a friend that has expressed wanting a low watt overdriven amp sometime in early 2023.

With respect, 10thtx