Dumble-ish preamp project

[rootz]

I've been a long time visitor of the amp garage and based on the great information found here, I build a 2xEL84 Rocket reverb some years ago. That little amp is now my main gigging amp for small clubs and is great on its own when (turned up to push the output section) and as a pedal platform.

However, amp building is like an ever returning itch to me. Since building the Trainwreck inspired amp, I wanted something Dumble inspired too. Always loved Fords sound and was a huge fan of Bonamassa about ten years ago.

Now, I had a power transformer, choke, etc. To keep things flexible, I wanted to go the route of rack mounted preamp and power amp. The power amp is more a Marshall than a Fender. 10k/4.7k tail, inductor based resonance (stole that idea from Diesel), switchable feedback, DC coupled MOSFETs driving the output tubes (will do class AB2 with ease).

The main idea behind the preamp was to build something that could be either HRM or non-HRM. Soon after finishing I found that too limiting. The Two Rock Opal gave me the inspiration to make the choice between HRM and non-HRM switchable. Actually, almost anything is switchable. There is switching for rock/jazz, brightness (100p and 330p), midboost (390p//2n2 or 2n2), deep, clean/OD, HRM-bypass and PAB. On the back is switching for local negative feedback in CL2 and two switches to couple either or both midboost and PAB to the clean/OD channel switch. The latter is very useful.

I didn't quite dig the high plate cleans. Pristine clean, but a bit dull with a strat. I changed that to low plate with 10uF1k5 cathodes. Skyliner bass was a bit overwhelming in combination with the deep switch. Changing the slope resistor to 100k and mid cap to 47n solved this.

After that change the preamp was a bit harsh in OD. So I added 220p snubbers on OD1 and OD2 and converted the input trimmer to #102 style (220k/100k pot). The OD is now much smoother, though HRM is still a bit sharp. Might need to change the treble cap to 330p on the HRM board to have a bit less in your face mids; to scoop a bit more.

I've made some recordings over a backing track to find out how the OD now sits in the mix. I think it sounds pretty good already, even with some EMG's (89). Clip will be added shortly.

So, let me know what you think. I already know this is not the typical Dumble build .

[norburybrook]

that looks, to my eyes, very nicely made. I'm impressed at you making your boards like that.

hope it sounds as good as it looks.


M

[ChopSauce]

Yes, great build. It's so neat

Thanks for the reporting, also!

I'm pretty sure I wouldn't be the only one interested in some schematic for that beast, especially the section relevant to the MOSFET driving the power tubes, though.

[rootz]

Thanks Marcus! You should have seen my first failed attempts. After a couple of PCB's I got a feeling for etching, drilling, silk screening. When I get past the prototype stage I frequently order factory made PCB's. Made my whole pedalboard that way from a Deep blue delay clone to a Landgraff OD and all in 1590a mini pedal format.

It's not for me to judge if it sound as good as it looks, so here's a clip.
https://soundcloud.com/user-433489378/d ... n-hrm-mode

The whole clip is in non-HRM OD mode. The dynamic range is pretty good if you ask me. I was a bit afraid the snubbers on OD1 and OD2 would dull out the tone too much when I roll back the volume of the guitar.

[erwin_ve]

That's a interesting builds Rootz, I like it!
I see you used 3 pre amp tubes, V1 clean, V2 od, so what is V3 doing? Cathode follower?
The soundsample you posted sounds very good to me, dynamic and the low end is super tight for a non hrm overdrive.

[norburybrook]

sounds great rootz

do you have to get a minimum number done when doing factory made or can you get one offs?


boards remind me of the early Peavey amps. they were really good quality, still going strong 30-40 years later, I've gutted a few for the transformers and was impressed at the build quality.


MC

[rootz]

Thanks ChopSauce, Edwin and Marcus.

I always try to work neat Chopsauce, it prevents me from making mistakes and makes modding a lot easier. Also, working with pcb's makes it actually easier to work neat. Will post some more about the power amp at a later moment. It's currently not without its flaws: there is a nasty 50 cycle hum that depends on the setting of the resonance control. Seems the resonance inductor gets coupled to the power transformer. So layout problems, but no easy way to test it. The MOSFETS work great though. Tried concept I got from diyaudio, or to be more exact gingertube's 6V6 Baby Huey amp. So yes, controversial (sand in tube amps is like swearing in church to some) high end audio solutions in a tube guitar amp.

The 3rd tube is indeed an AC coupled cathode follower (so standard common cathode and cathode follower inline). To keep output levels under control I wrapped local negative feedback around the whole tube, so from output of the cathode follower to the input of the preceding triode. Much like an anode follower. There is a design flaw though. The master volume pot is behind the cathode follower, so limiting it's capabilities in output impedance. And due to the high levels of feedback it is virtually impossible to overdrive the cathode follower, thus giving up tone shaping possibilities like a dumbleator would give me.

What I'd like to do next as part of an experiment is to put the volume control between the last two triodes and run the first triode as the recovery stage of a dumbleator and the last as a cathode follower. Sounds familiar? That's what Two Rock does in for example the Opal (minus the cathode follower).

The low end is unlike what I heard before from a non-HRM Dumble. It's not mushy, but also not as 3D. The 2n2 cap after OD1 as per HRM design plays a role. The 100k slope resistor and 47n mid cap are another. Both cut massive in lows. In HRM mode the lows are almost overwhelming though! Even with a 47k slope resistor and 330p treble cap on the HRM board.

What do you think Marcus, do you hear Dumble DNA in the sound? And if so, what is it?

Those boards are a charm to work on. Very smooth and much softer than the FR4 boards. So drilling is easy, as is cutting it. I can now cut and break boards! Try that with FR4. Sadly, the copper is not as strongly adhered to the base too, so gets torn of more easily. Not the best base for modding. But very fast and easy to work with, which is great.

Ordering boards always is in larger quantities if you want good prices. Last time it was PCBway in China I believe, OSH Park is great too. PCBway is good quality especially for the price but not super fast. OSH park has those great purple pub's with gold plating. Fast and cheap for small boards. My Hifi amps have PCBway pcb's. I might as well get you some teasers of those amps before I get back on topic again.

[erwin_ve]

Thanks Rootz for giving some insight in your design choices.

Im not familiar with the Two rock opal, so your next step for V3 might be basically a upside down FX loop:recovery triode, volume control, cathode follower?

I hear your comment on the bass content when in Hrm. My Hrm board is almost on it's lowest setting for bass and I usually combine it with Pab only.

[10thTx]

Deleted the schematic I posted for editing to avoid confusion with actual schematics below.

Thanks for sharing your schematic! Very interesting design.

Thanks, 10thtx

[rootz]

I think I could get away with a 100k bass pot on the HRM board at this point. Should give me a much more usable range, because currently it sits at 1 or 2 on a scale of 10, so 25k to 50k.

The Opal is a mix between different Dumble topologies. Two Rock can claim they came up with their own designs based on the sound they wanted, but the similarities between their designs and HAD's are striking. The idea to use a single tube reverb and add the recovery stage of a Dumbleator were new though, at least as far as I can tell.

Here's the current schematic for the signal chain. I've not added the FET at the moment, nor do I feel the urge to do so.

[rootz]

The reworked half dumbleator and cathode follower would look like the top schematic in the next picture.

[rootz]

We are a year further and I finally got to work on the preamp again. I changed the design to a #183 (just love the sounds of the different clips I heard), but off course with switchable HRM stack. I think I had a all (or at least mostly) wrong before by copying the Two Rock Opal solution. In the Opal the HRM stack gets lifted and a 100k//330p combo is switched to ground. Loading of V2b is thus okay-ish for something like a #102, but you still lose a boat load of low end and the high end gets a bit tame with another low pass filter.

So I went another route: optimising for #183 with proper loading of V2b and switching in a HRM stack and dropping the #183 load for V2b simultaneously when switching to HRM mode. The HRM stack has a 250k bass pot and is rather bass heavy due to 10n coupling on V2a and 4u7 cathode bypass cap on V2b. Now both modes sound to my liking and HRM mode can do some pretty convincing hard rock and metal tones too! I should build a complete amp around this concept and put the HRM lift on a foot switch.

[10thTx]

I greatly enjoyed your 2018 sound clip! I thought it had a terrific tone to it. Any chance you have a sound clip with these latest revisions?? I'd sure love to hear it.

With respect, 10thtx

[rootz]

I have some rehearsals to do for a gig tonight. Might as wel record some clips when I’m at it.

[rootz]

There you go! Some fresh sounds 9from the guitar, that is, not me).