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

[frequencycentral]

So Rick, do you have a layout for the Nixie?  Would be much appreciated.

I did this the other night, I haven't been back over it to double check yet. Maybe you could do that? Note the jumpers and wires under the IC socket. The jumpers are under the board, the wires on top. There is no inductor symbol in DIY Layout Creator, so I used a transformer symbol. You should also check that the caps fit, they're a bit bigger than your average caps.



Another thing about the SMPS, I've been leaving it running for hours hooked up to the amp. I went back to it yesterday to find it humming slightly at 195 volts, I turned it down to 190 volts and the hum went away. I don't seem to be able to get it back after hours of running at slighly lower voltage, which is a good thing!

.....and I got three more 5902 in the post today! 


EDIT: That 2.2uf cap is rated at 250 volts. Please don't anybody build this without referring to this PDF: http://www.ledsales.com.au/kits/nixie_supply.pdf

.........and be sure you know what you are doing - these voltages can kill you!

[frequencycentral]

Here's a quick soundclip. The amp is still on breadboard, and I've got a few more ideas to try, but I'm pretty pleased with the sound.

http://www.aronnelson.com/DIYFiles/up/Sub_Zero.mp3

Signal path:

Squier Telecaster Custom (HB)  >  Sub Zero  >  12" / 8 ohm speaker  >  Shure SM58  >  Event Echo Gina 24 soundcard  >  Wavelab

The first half is the guitar directly into the amp, the second half features my Russian Germanium (GT404) Fuzz Face. No other processing.

[gnognofasciani]

From a quick look, it seems that the 555 pinout is wrong

[frequencycentral]

From a quick look, it seems that the 555 pinout is wrong

How so? I just double checked - seems OK to me.

[solderman]

From a quick look, it seems that the 555 pinout is wrong

I dont know what chematichs the rest is from but the 555 pin layout is correct See this chem Obs the number of the pins does not correspond to there individual place on the DIL



//solderman

[Ripthorn]

Sounding great Rick.  Is that a little bit of grit on the amp or is that just the jangli-ness of the tele?  I look forward to watching this thing come along.  Another question, did your Murder One quiet down some after putting it in the enclosure (in terms of background noise)?

[frequencycentral]

Sounding great Rick.  Is that a little bit of grit on the amp or is that just the jangli-ness of the tele?  I look forward to watching this thing come along.  Another question, did your Murder One quiet down some after putting it in the enclosure (in terms of background noise)?

Thanks, Yup a bit of grit. My Tele Custom is humbucker equipped, I was using the bridge humbucker. Also bear in mind that the amp's running flat out. I guess I could design out the grit but I quite like it, I would expect it wouldn't be there with single coil pups.

The background noise I reported with Murder One was actually a dodgy guitar lead! They build really quiet now. I'm just starting #0004.

[solderman]

Hi all
I need to power the heaters some way. Since 2x 12Ax7 and one EL90 will use just over 1A and the LM317 will fix this I decided to go with this and I made this layout for both the HP Nixie and the filament power. The filament part is working just fine and heating my tubes. It delivers exactly 6,3V with the trim pot at the end. I'm just waiting for the delivery from Banzai with the UF1N4007 (4007 was the only UF diode they had its overkill but...) and the 555 and IFR740 since I smoked the ones I had in my goofed up perf board layout. That's why I always make PCB:s all my perf board fails.

I'm also curios about what's gone happen if I add 24V instead of 12V. I have seen some Nixie supplies that use 18V. Probably would have to step up the IRF740 to something hotter. The LED current limiting resistor R11 would have to change also. I used 1N4004 cause I had them already. The 22uF 400V is to source enough current for the amp. by using rectefied DC to filament the motorboating wold not be a problem.

If this Principe can work with 24V amp up to around 5W can be made with a cheap 24V transformer. yippeeey then its only the cost for the OT but...... you cant win them all 

MIND YOU THIS LAYOUT IS NOT VERIFIED



BOM
http://solderman.fatabur.se/Future%20project/Nixei.txt

//Solderman

[frequencycentral]

I'm also curios about what's gone happen if I add 24V instead of 12V. I have seen some Nixie supplies that use 18V.

Mine gave out exactly the same voltage with inputs of 9 volts and 12 volts. I think its the way the BC547 is set up to regulate the output, it switches off the oscillator at the voltage set by the trimpot.

[solderman]

I'm also curios about what's gone happen if I add 24V instead of 12V. I have seen some Nixie supplies that use 18V.

Mine gave out exactly the same voltage with inputs of 9 volts and 12 volts. I think its the way the BC547 is set up to regulate the output, it switches off the oscillator at the voltage set by the trimpot.

OK so you have allready tryed this. Interesting find. Think I need do some Googling. There aught to be a way to get it up to around 250V

[kurtlives]

Trying to get a sense for the sounds and design ideas behind the different output sections seen in submini amps.

Judging by the sound from the clip this amp seems to have the most headroom? What vol/gain level was that clip recorded at?

[amptramp]

Just looking at the design in the original post, the split-phase inverter drives the output tube grids whaich may go positive with respect to the cathodes and therefore draw grid current.  This may cause "gulp distortion" where the amplifier momentarily drops out due to excessive negative grid bias.  It may not be an issue with the medium-mu triode input stage but will be a problem with the high-mu triode.

With gulp distortion, the positive voltage on the grid pull electrons from the cathode.  When the grid goes negative, they have nowhere to go except down the grid resistors in the output stage.  The voltage declines with a time constant of RC, or in this case > 0.1 uF x 1 megohm or 0.1 seconds since the drive impedance is added to the grid resistance but is much lower.  With sufficient drive, you can cut the output stage off for a noticeable amount of time.  The output appears to be swallowed, hence the name gulp distortion.  If you are pulling grid current, you need a DC-coupled low-impedance drive like a cathode follower.  This remains true whether you are running class AB2 or A2.  (In class A, the output tube is never cut off.  In class AB, it is cut off for less than half of the input signal range.  The "2" indicates that grid current is drawn during the positive signal swing to the output stage.

Adding a bias resistor to the output cathodes will add negative grid bias which will tend to move the operating point from class A2 to AB1 where no grid current is drawn.  Adding cathode followers between the inverter and the output with no coupling capacitors between them will prevent gulp distortion even if the amp is operating in class AB2 or class A2.

[frequencycentral]

Trying to get a sense for the sounds and design ideas behind the different output sections seen in submini amps.

Judging by the sound from the clip this amp seems to have the most headroom? What vol/gain level was that clip recorded at?

Yeah, good headroom. The clip was recorded with the amp maxed out. I never built one, it sat on one of my breadboards for months. I eventually used most of the parts to make Superfly, which seemed to me to have as good a clean sound and headroom at mid volumes, but can also do cranked, so more versatile. One thing to note with Sub Zero is that there's only actually one triode in the preamp. I guess I'll come back to Sub Zero when I get around to trying some other SMPS designs, which might give higher B+.

My take on these very small amps without real PT's is that any more than two tubes in the design and the object of the exercise is defeated, as you're going to need a hefty wall wart which might as well be a PT.

Fo the ultimate expression of a submini tube amp have you seen this? http://ampgarage.com/forum/viewtopic.php?t=6159 I've got some 5902 pentodes, but need to track down a suitable PT suitable for UK mains and capable of providing -33v for the grid bias.

[kurtlives]

Thanks for the info...

Yeah I have read through that thread and watched it's development. Not really what I am into. It's just a scaled down clone, kinda boring imo. I wanna try new ideas and stuff...

My take on these very small amps without real PT's is that any more than two tubes in the design and the object of the exercise is defeated, as you're going to need a hefty wall wart which might as well be a PT.

What do you mean by this? The small factor? The heater requirements?

For the idea/design I am working on it's lots of tube and a unique PS that does use a small xformer...

[frequencycentral]

My take on these very small amps without real PT's is that any more than two tubes in the design and the object of the exercise is defeated, as you're going to need a hefty wall wart which might as well be a PT.

What do you mean by this? The small factor? The heater requirements?

Yeah the small factor and the heater requirement. What's the point of doing a palm-sized amp if the wallwart is massive. I guess like the 2w Sub Miniature Express - what's the point of using subminis if the enclosure is standard sized.

[kurtlives]

Ya that was kinda my thoughts on the mini TW as well.

I do think it is cool to have big complex designs using subminis though. Cause you can have smaller(ish) PSUs and lower output while retaining a lot of character.

[frequencycentral]

I'll be fascinated to see how your project turns out. And the nature of the 'unique' PS.

IMO submini amps should be submini-small. So SMPS's really go hand in hand with them. I'm really looking for a SMPS that can reliably put out 300 volts before I develop the Sub Zero idea any further, those 7327 tubes are the highest voltage rated I've come across @ 300 volts.

[Ripthorn]

One thing you can do, Rick, to get your -33V bias is to use the classic "voltage doubling" circuit used on AC voltages.  You could maybe run this off a heater tap or something from an off-the-shelf PT.

[frequencycentral]

One thing you can do, Rick, to get your -33V bias is to use the classic "voltage doubling" circuit used on AC voltages.  You could maybe run this off a heater tap or something from an off-the-shelf PT.

That's a good idea to use the heater tap. You could even rectify the AC and use it to drive a charge pump I guess. I really need to do some research and study on PT's.

[kurtlives]

A lot of the early tweed Fender's did not have a dedicated bias tap and used various methods to get a -ve bias voltage. Might be worth it to check them out.

IIRC they tapped a portion off the HT AC secondaries are dropped the voltage quite substantially with a dropping resistor. Then went on and did the normal bias stuff...