"Sub Zero" Low Voltage Submini Tube Push/Pull Amplifier Project

[frequencycentral]

I'm really pleased with the way my Murder One amp turned out, but this time I'm looking to build a cleaner amp using a PP configuration. The idea behind my Murder One was to build a SE quiet cranked amp - which it does very well, the clean sound is too quiet to be useable however.

The other design concept is to power it from a 12 volt wall wart. Murder One used a charge pump to get up to ~70 volts, Sub Zero is going to use a SMPS like this: http://www.ledsales.com.au/kits/nixie_supply.pdf

At the moment I'm breadboarding this using two 6111 submini dual triodes and the same charge pump I used in Murder One - I don't have the parts yet to make the SMPS. The final version will use two 7327 instead of the 6111's, as they have a higher voltage rating. I guess two 12AU7 could also be used.

It sounds really cool and clean, but underpowered - as I expected, as the charge pump doesn't deliver enough volts.

Any thoughts, comments, suggestions?





6111 data sheet: http://frank.pocnet.net/sheets/127/6/6111.pdf

7327 data sheet: http://frank.pocnet.net/sheets/138/7/7327.pdf

[Ripthorn]

This looks really cool.  After the first two stages on my submini amp, the sound goes from pristine and sparkly clean to mild crunch, so I think I know what you are talking about here.  My only question is if the push pull pair is seeing the optimal impedance (I noticed you are using the same tranny as in the Murder One).  That could have something to do with power output.  I also bet that there is a lot of low power feeling going on because you only have a single gain stage, then your cathodyne PI and then your power amp.  With only one stage and 6111's you probably aren't getting a gain factor of much more than 13 from the preamp.  Those are just my thoughts, just remember that I don't have a whole lot of experience with tubes either...

[frequencycentral]

The transformer is different one to Murder One, it's a Hammond 125A PP with centre tap and multiple ratio taps. I think I've got it set up optimally - I'll have a little tweak and see! Thanks for your other comments - food for thought, I'll play about with it some more..........

[Renegadrian]

Rick, I think I'd need another life to experiment with your creations...unfortunately life is asking for more time, so less time to dedicate to experiment/build pedals...   thx for your efforts and the way you share them with us...
Adriano

[frequencycentral]

I also bet that there is a lot of low power feeling going on because you only have a single gain stage, then your cathodyne PI and then your power amp.  With only one stage and 6111's you probably aren't getting a gain factor of much more than 13 from the preamp.

Yeah, for this design I really want to use just two dual triodes, as running the heaters in series keeps the milliamp draw low. The challenge is to get a decent amount of clean volume from this setup. My hope is that when I get the SMPS running at ~250v I'll get what I want.

The aim is also to fit the whole thing (including the OT, which is really small) into a Hammond 1590B, without any tubes, caps or transformer sticking out of the top - though it'll need a light emmiting border to add enough depth to the enclosure for the OT.

Rick, I think I'd need another life to experiment with your creations...unfortunately life is asking for more time, so less time to dedicate to experiment/build pedals...   thx for your efforts and the way you share them with us...
Adriano

........I still have a number of goals to achieve with submini tubes - phaser, wah, envelope filter, sample rate reducer - but I wonder what I will do in a year or two when they're all done!

[Boogdish]

why no cathode resistors on your output triodes?

[frequencycentral]

why no cathode resistors on your output triodes?

I tried it first with 100 ohm cathode resistors on the output triodes, it's louder without. Is there any advantage to having them, what function should they perform? I'm still learning this stuff.

[kurtlives]

why no cathode resistors on your output triodes?

I tried it first with 100 ohm cathode resistors on the output triodes, it's louder without. Is there any advantage to having them, what function should they perform? I'm still learning this stuff.

Your limiting your clean headroom. It easier for your signal to clip.

Pretty cool design. But when are you going to start playing around with some real high voltages Rick? 

[Ripthorn]

You know, I just had a thought.  On the MosFET follies page at geofex, RG said he would do a mosfet phase inverter.  It's not posted, but if there is one floating around, that could buy you another triode stage for more output.  As far as I am aware, the phase inverter does very little to the tone, if anything, so putting a mosfet there probably wouldn't compromise your quest for tubiness.  It would also allow for a little bit of preamp crunch if you wanted it.  When you get the higher voltage levels, you will have to let me know how much more output you get and just how loud it will go.  Also, in the way of other methods of getting high voltages, Renegadrian posted a method he used using common transistors and caps to get big voltages.  Not sure if they are as high you are looking for, but might make a good stopgap or placeholder until the stuff for your nixie supply come in.

[frequencycentral]

why no cathode resistors on your output triodes?

I tried it first with 100 ohm cathode resistors on the output triodes, it's louder without. Is there any advantage to having them, what function should they perform? I'm still learning this stuff.

Your limiting your clean headroom. It easier for your signal to clip.

Ok - when I get the SMPS working I'll put the cathode resistors back in. I really want zero distortion in this design - hence the name "Sub Zero". Do you think 100R is an appropriate value? I guess the higher the value the more clean headroom? Cathode resistors on output stages tend to be less than 1K from what I've seen.

Pretty cool design. But when are you going to start playing around with some real high voltages Rick? 

After this amp I'll do some more tube effects - see reply #4 - then it will be time for a HV amp. Part of the attraction for me of low voltage tube amps and tube effects is that it's a little niche of electronics that doesn't seem to have been explored fully - how many low voltage tube amps, tube vibratos or phasers do you see? So there is room for new designs. There are hundreds of HV tube amps to choose from already.

You know, I just had a thought.  On the MosFET follies page at geofex, RG said he would do a mosfet phase inverter.  It's not posted, but if there is one floating around, that could buy you another triode stage for more output.  As far as I am aware, the phase inverter does very little to the tone, if anything, so putting a mosfet there probably wouldn't compromise your quest for tubiness.  It would also allow for a little bit of preamp crunch if you wanted it.

I just couldn't put a MOSFET in there. I have nothing against solid state, which I have used extensively in synths. I have set myself the criteria of all-tube for this project.

Also, in the way of other methods of getting high voltages, Renegadrian posted a method he used using common transistors and caps to get big voltages.  Not sure if they are as high you are looking for, but might make a good stopgap or placeholder until the stuff for your nixie supply come in.

I think that circuit is really interesting due to it's use of common parts, but it ultimately depends on lots of diodes and caps just as the MAX charge pump does, so to get up to ~250v would make it bulky. The SMPS seems to be the logical way to get up to those voltages with a low parts count. By the way, I blew a MAX1044 today when I added two more stages (that's 4 more caps, 4 more diodes), I'm sure I didn't make a wiring error - I assume that I reached above the limit of what the chip can do - the little MF melted part of my breadboard!

[Ripthorn]

Well, then it looks like you've got about the best design that fits all of your criteria, so I say have at it and let us know how it turns out!

[frequencycentral]

I just couldn't put a MOSFET in there. I have nothing against solid state, which I have used extensively in synths. I have set myself the criteria of all-tube for this project.

Mosfet presents a solution to the current challenged supply..Ripthorns suggestion is a good one I think. It'll still sound plenty tubey...

Your suggestion has got me thinking.......I really want to keep the audio path purely tube, but it might be possible to implement a MOSFET as a CCS (constant current source) like I did in my Pentadriver Turbo Mod, this would replace the first stage triode's plate resistor, which would mean more voltage is available to the plate instead of being used up by the anode resistor. This would increase the gain of that stage, but the MOSFET would not be in the audio path.

The Pentadriver Turbo Mod used a PNP transistor as a CCS, but forum member intricatecode posted a similar idea on the Tube boost + overdrive running off a 9 volt battery thread using a J201 as a CCS, so it should in theory be possible to do something similar with a MOSFET. I don't know much about high voltage MOSFETs, but perhaps the same high voltage MOSFET I'll be using in the power supply, an IRF740 could be used in this application. For me this would be a neat solution, as you'd get the extra boost with an all tube audio path. Here's intricatecode's CCS:

[intricatecode]

just be shure to use a 1k resistor with those j201, lower values thant that can give some inconsistent results due to jfet variability and it doesn't get any louder if you go lower than 1k.

been using this with 6111, works just fine.

to use with higher voltages http://headwize.com/projects/cavalli2_prj.php probably try using lnd150 version of the ccs on that page.

[frequencycentral]

just be shure to use a 1k resistor with those j201, lower values thant that can give some inconsistent results due to jfet variability and it doesn't get any louder if you go lower than 1k.

been using this with 6111, works just fine.

Thanks - I'm thinking a similar idea but with high voltage MOSFETs. You're enjoying the 6111's then?

[intricatecode]

ordered a few, haven't had all the time in the world to play around lately but i've built and active mixer with one to mix the wet dry signal on my pedalboard. I'm going for an all tube signal path on the drive stage of my pedalboard 

[Dragonfly]

You might get some good ideas by looking at Fred Nachbauers MiniBlok II

[Johan]

with nearly 1watt max dissapation possible on each tube and 70volt in the voltage supply, you might want to strap your OT for lower primary impedance if you want more output.
j

[frequencycentral]

with nearly 1watt max dissapation possible on each tube and 70volt in the voltage supply, you might want to strap your OT for lower primary impedance if you want more output.
j

Ok thanks, I'll try it, do you have an idea what impedance I should be looking at? I'm not sure how to work this stuff out. And will the ratio need to change once I raise the supply voltage to ~250v? Is there an equation I should be using?

[frequencycentral]

Hmm, I've tried every possible ratio - there doesn't seem to be much difference between them. 

I did fine this calculator: http://www.sengpielaudio.com/calculator-ohm.htm

[Boogdish]

why no cathode resistors on your output triodes?

I tried it first with 100 ohm cathode resistors on the output triodes, it's louder without. Is there any advantage to having them, what function should they perform? I'm still learning this stuff.

A much better explanation on this than I can give can be found at the valve wizard site:  http://www.freewebs.com/valvewizard1/gainstage.html
It's somewhat of a long read and it has math in it ( or "maths" as you would say) but it's definitely worth it.

You know, I just had a thought.  On the MosFET follies page at geofex, RG said he would do a mosfet phase inverter.  It's not posted, but if there is one floating around, that could buy you another triode stage for more output.  As far as I am aware, the phase inverter does very little to the tone, if anything, so putting a mosfet there probably wouldn't compromise your quest for tubiness.  It would also allow for a little bit of preamp crunch if you wanted it.  When you get the higher voltage levels, you will have to let me know how much more output you get and just how loud it will go.  Also, in the way of other methods of getting high voltages, Renegadrian posted a method he used using common transistors and caps to get big voltages.  Not sure if they are as high you are looking for, but might make a good stopgap or placeholder until the stuff for your nixie supply come in.

The problem I see with this is that in a push pull amp you need to have an even number of tube sections amplifying otherwise it becomes a lot more push and a lot less pull, if this were a single ended output amp he could easily use three triodes in his output section.