Battery amp from National Dobro

[PRR]

Battery power tube guitar amp from National Dobro.

National Dobro Model 31

Note: in this plan, "M" resistors are what we call "K" today. "Meg" is Meg.

OT CT not shown, a typo. OT is probably 20K CT.

Power output from two 1C5 tubes about 0.4 Watts:
1C5 datasheet

1G6 data

Taking volt-amp stage gain as about 20, input sensitivity is about 7V/(20*20)= 20mV, in-line with classic large Fenders.

Should run many hours on a ten-pack of 9V batteries and a D cell or two.

[DummyLoad]

cool! thanks for sharing.


--pete

[jeff]

Cool. bring yer tube amp to da beach man.

Is the "-7" -7V bias voltage? If so how are they getting that, I don't understand. Is there another battery just for bias? Would that be the C+?

10 9Vs and 2 Ds...wow!

[PRR]

> how are they getting that

350 Ohm resistor in the negative battery lead.

[jeff]

Oh, ok.
I took the schematic to mean the 90V battery was sitting on top the 350R resistor. I get it now.

[jeff]

Revitavly speaking, measuring from the battery's - this looks like a cathode biased amp but with all the "grounds" tied to the top of the cathode resistor. Am I understanding that correctly?

[DummyLoad]

measuring to ground from the battery (-) you see -7V virtual zero volt is in the battery? 


build a battery eliminator. skip down to section 16.


http://www.dos4ever.com/battery/battery.html



--pete

[kagliostro]

Nice Amp PRR and interesting link Pete

Thanks for Sharing

Franco

[jeff]

Wow, that threw me for a loop. Are you saying it's a 97V battery? (-) terminal at -7V, (+) terminal at 90V and 0V is a virtual point within the battery? Whew!

So is this a fixed bias or a cathode biased amp?
 Looks cathode biased to me but instead of saying the grids(which are tied to the battery's -) are at 0V and the cathdoe is at +7V, the measurements are taken from the cathode so the cathode reads 0V and the grids read -7V.

I guess with a battery, ground is where you assign it.
 A 9V battery could be 0V on the - and 9V on the +,
 or 4.5V on the + and -4.5V on the -.
 Or 0V on the + and -9V on the - (like a fuzz face).

Just blew my mind for a sec.

[HotBluePlates]

Wow, that threw me for a loop. Are you saying it's a 97V battery? (-) terminal at -7V, (+) terminal at 90V and 0V is a virtual point within the battery? Whew!

So is this a fixed bias or a cathode biased amp?

This is a back-biased amp. Or you could call it fixed bias, but developed in a back-bias manner.

The battery is 90v. The output tube cathodes (which are directly heated, so they are the filaments too) have one side connected to ground. So do all the other tubes. A resistor is placed between ground and the battery - terminal. That means all current for the amp is drawn through that 350Ω resistor, and creates a 7v drop across the resistor.

However, the location of the circuit ground and the direction of current through the 350Ω resistor make it appear as 7v "below ground" or -7v on your meter when the black lead is touching ground. So the 90v battery - terminal also looks like it's at -7v, and so the + terminal will measure 83v when your meter's black lead is on a ground point.

After all, "ground" is only a reference point you choose to call "0v."

The -7v developed across the 350Ω resistor is applied to the output tube grids to bias them. But there's not a true fixed bias supply, just the back-bias resistor. Looks like the entire amp's B+ supply only draws 20mA from the B battery.

[DummyLoad]

But there's not a true fixed bias supply,

really? i guess the battery and the || RC combo don't count as being a bias supply?

--pete

[HotBluePlates]

It's current-dependent, just like a cathode resistor. So it's not "fixed" in the way a power supply with its own rectifier and filter are "fixed."

I did say "call it fixed bias, but developed in a back-bias manner."

[PRR]

It is self-bias.

Traditional "back-bias" runs the audio AND the radio tubes together through a resistor. The sum of the radio stages may be similar to the audio stage current. In that case a change in audio stage current does not cause a large change of bias.

But here we have like 19mA in the power tubes and 1mA in all the rest. It is 95% self-bias.

The plate-cathode voltage is 83V. That's why there is a 83V condition on the data-sheet. 90V batts were very common, carrying a third batt just for bias is more trouble than it is worth, so they take-off the bias and find the output for 83V plate-cathode.

[kagliostro]

Thanks to all for the explanation


K

[HotBluePlates]

Assuming you want to lug a 12 v or 6 v battery around, you could replace the A and B batteries and use a vibrator and rectifier.  ...  I guess you could purchase some solar panels and power this 1 amp wonder.  (harbor freight has some 45 watt panels).

If you lug a heavy battery and a vibrator, you'll need a rectifier and filter caps. Which misses the point of a 0.4w amp run on common light batteries.

The 20mA average draw results in 20 hours or better from the (400-565 mAh) 9-volt batteries. The 400mA draw for the filaments yields 20 hours of play time from a single (8000mAh, Zinc-carbon) D-cell, even longer for an Alkaline battery.

Cheaper, lighter, all-around better.

[Ritchie200]

This will allow you to scare the bears in your remote mountain campsite for days!

Jim

http://www.hi-powercycles.com/90v-13ah-hpc-elite-90v-13ah-li-nmc-ultra-high-performance-battery/

[Ritchie200]

Although at $1K less, this might be more appropriate.  No rating listed however.

Jim

http://www.batteriesinaflash.com/radio-microphone-meter/exell-490-alkaline-90v-battery-neda-204-a490-60f40

[HotBluePlates]

This will allow you to scare the bears in your remote mountain campsite for days!

A month. Non-stop. 13Ah / 20mA = 650 hours = ~27 days.

[HotBluePlates]

So How do you calculate 0.4w?    The method I used was 4 times the SE amp output. (0.96w) or ~1 watt.

Original Post (and linked data sheet)

http://el34world.com/Forum/index.php?topic=16366.msg161078#msg161078

[sluckey]

Your attachment, about field coils,   My research indicates that when the original schematic was published, field  coils were stated as DCR, so determining Z in the A/C ohms equations become necessary.

I believe you missed the point of that link... or chose to ignore it.

[DummyLoad]

I for one am done debating with you from now on, because you refuse to let info sink in (which could only benefit you). From now on, if I'm inclined to refute any future bad info you present, I will save myself time by linking back to this post. I regret only that I've wasted my time in trying to help you where help is obviously not wanted.

i believe that the excerpt above is what was being conveyed.

--pete

[RicharD]

Rather than bicker about this kookie old circuit, how about design it forward.  I've spent the last year and a half eyeballs deep in energy research.  Like it or not 48VDC is coming into the home in about a decade.  Our lab runs 48VDC lighting already.  Your grid source will still be good ol AC but home storage is right around the corner.  Design an amp using cordless drill batteries, +/-48VDC.  :)

[kagliostro]

Ciao Richard

is a long time we don't hear you here, happy to read you 

Franco

[RicharD]

Thangs are good in my world albeit very different from a couple of years ago.  I haven't built a tube circuit in ages hence why you don't see me here.  I now work for an energy research company.  We have compiled the world largest database of residential power use.  We've captured a lot of data in commercial settings too but there's not as much interest (funding) for that research.  I am currently in Boulder Co. installing real time energy monitors.  When I get back home I will resume work on the control circuitry for the battery storage cabinet we are developing.  I cannot say too much about it because it's not being developed on my dime.  It does use transformerless switching inverters.  The trick here is making things switch precisely at the moment of zero volt zero current, or as close as possible, therefore the big honking transistors are dissipating almost no power.   Life expectancy is forecast by electrolytic caps (sound familiar) and environment.  It's pretty crazy stuff.

[kagliostro]

A new job and a lot of commitments, it was to imagine, made to feel when you can


Ciao


Franco

[PRR]

> 48VDC is coming into the home

I'd be happy to have 240VAC. I have 250V no-load, 230V when the dryer runs, I got the smallest A/C compressor I could find so it doesn't get dim on humid evenings. And we were at zero Volts for 33+ hours after an ice-storm. (Others were dark for a week.)

48V domestic power seems low. Yeah, I get a lot of light from a 12 Watt LED fixture, so maybe I could light with small wires. But I just re-wired a lot of 117V runs, and am running out of fishing spaces.

Home Storage.... color me dubious (but ready to be wrong). Maybe in a decade battery prices will tank and significant storage will become cheaper than sizing transmission lines for peak loads. OTOH I am still waiting for Nuclear Power "too cheap to meter". And personal jet-packs.

Tube amps can be happy on 48V. 50L6 will push 0.2W SE at 48V. Farm-radios used to run on 48V WinChargers and such.

[jeff]

Don't get me started on Edison.

This thread did get me wondering, what are the 1Meg resistors between the inverter plates and the power tube plates?

[sluckey]

That's a local NFB loop. Thinking of it as a resistor from power tube plate to power tube grid may make it seem more logical.

[Ritchie200]

Yeah and Tesla is shaking his head calling us all idiots!  We should have had wireless power by now - or 100 years ago.....

I agree with PRR.  That was one thing the Brits got right, higher voltage - lower current.  It only tickles a little bit more when you touch it to your tongue...  I too am waiting for the Jetson's flying car and space needle homes.

Dubious Jim

[RicharD]

My post was pretty vague and misleading.  First off traditional AC transmission & distribution is not leaving anytime soon if ever.  The home storage thang is probably a decade out and it's not going to be limited to +-48VDC.  This voltage has been chosen because there is already a ton of equipment out there.  Ma Bell has been running 48VDC forever I think.  National Electrical Code starts at 50 volts to "ground".  You do the + - thang and you've got a solid 90 volts to use without the constraints of an authority having jurisdiction. 48V is plenty for LED lighting.  Our lab lighting can run either at 48VDC or 120VAC.  I don't really see DC wall outlets becoming a big thing because rectification is so easy.  Shoot, you don't even need a transformer anymore. 

Right now, battery storage is only legal when installed in a traditional UPS format.  This means the batteries are tied to an inverter that feeds loads only.  No push back to the grid.  Great if you have a room full of computers you need to keep running 24/7, not very beneficial to the family with a rooftop covered with solar panels who wants to store their excess production for later.  A typical solar installation has no storage and if the grid goes down, so does your inverter because it has no frequency to sync to + it would now try to backfeed the grid, quickly overload die.  Right now you get to sell your excess electricity back to the utility at a reduced rate.  I guess this is OK but it should be better.  This is what we are working on.

"I'd be happy to have 240VAC."
It's not just the folks who live at the edge of the world who suffer from voltage drop.  I've seen some very sloppy voltages in Austin during peak demand times.  Ponder this:  Suppose you have an electric car and you charge it slow during the night when you sleep.  You have telemetry on your electrical system that tell you your current is up, your voltage down, and your car is fully charged.  This could in turn switch the car battery secondary to an inverter which backfeeds the grid thus smoothing out your peak demand.  This mod works and could have been deployed yesterday yet it is illegal.   

[HotBluePlates]

Did a little followup on this amp.   Radio Museum has the model 31 at 0.8w

Gotta be smarter than your sources; it's not "rocket surgery" or "brain science".

1st Post:

Power output from two 1C5 tubes about 0.4 Watts:
1C5 datasheet

1C5GT data sheet gives Class A output for 1 tube as 200mW, with a 90v total supply, 7v of bias, 83v plate-to-filament (exactly what the schematic shows).

2x tubes in push-pull Class A can be counted on to deliver 2x the individual-tube output, when loading and supply voltages are similar; 1-tube condition specified 9kΩ load impedance, hence PRR's suggestion the OT probably presents a (9kΩx2=) ~20kΩ plate-to-plate load. Maybe even a bit more output when you account for the reduction of even harmonic distortion in push-pull.

So it's a 400mW amp; could conceivably make 800mW square waves, but for half the time as clean-sine (battery-drain imposes the limit).

Same Old Response