300Vdc from car battery(ies)?

Hi,

This is a total mooch question.

How does one go about converting 12 to 60 Vdc to a 300Vdc B+ supply that will deliver about 250mA?

I want to rig up a deep cycle car battery(or maybe as many as 5...more than that would be too dangerous, I think!) to power some tube preamps. I'm going to apply the same idea to a 12Vdc filament supply. Maybe phantom power as well. I want a very pure, rock solid dc supply.

I want to bring all that into a rackmount chassis with meters, switches, jacks, etc..

I just have no idea how to design a dc-dc converter to get my 300V B+.

Help please?

Dean

Bob Danielak sketched out a tube stereo rig for his car that he built a number of years ago:
http://www.geocities.com/bobdanielak/caramp.html

Unfortunately, he didn't think the switching power supply he designed was ready for prime time DIY, and didn't post the schematic.

You can ask him by email.

Steve Bench published several schematics for switching power supplies. Start here:http://members.aol.com/sbench101/#Power

-Jon

Thanks!

That switching business looks like a tradeoff as far as getting a clean/stable supply.

I might be better off rectifying AC from the wall for B+ and using straight batteries for phantom/filaments.

Dean

[quote author="deanp920"]Thanks!

That switching business looks like a tradeoff as far as getting a clean/stable supply.

I might be better off rectifying AC from the wall for B+ and using straight batteries for phantom/filaments.

Dean[/quote]
You're well on the way to electrocution. Don't ever, ever use AC straight from the wall without a transformer. Anything else in the real world that usually is properly grounded will provide a nice dead short for your project, not to mention you might become the dead short. In the old days they had transformerless AC / DC 110 V radios, but the manufacturer took precautions to ensure there weren't any metal parts exposed that were connected to the "hot chassis". Even so, people still got nasty shocks, or worse by tinkering around or trying to fix their own radio.

Build yourself a DC boost converter if you need this from battery power, they aren't hard to make. I't'll take a lot of car batteries in series to get 300V, and the current available would be highly lethal as well. Again, avoid the 120V mains idea, I would sure like to see future posts from you, not about you.
By the way, do you really think you need 250ma at 300 V for some preamps? This is really a lot of power at that voltage - 75 watts.

> B+ supply that will deliver about 250mA... maybe as many as 5... tube preamps

50mA is a HELL of a preamp. The classic RCAs drew about 10mA. Be sure of your numbers: overdesign of a DC/DC converter is very expensive.

The straightforward way is to get 12VDC/120VAC inverters sold to run household toys in a car, and feed a conventional power transformer. Most of these inverters make very nasty output. You could roll-yer-own, winding to the desired voltage, and including the final AC/DC conversion and filtering. You can save some iron by picking a frequency higher than 60Hz. But efficient DC conversion starts from square-waves, and it is awful darn hard to keep the leading-edge spikes from bleeding through.

Oh, another way: 12V motor spinning a 240V alternator and rectifier. This was actually the standard way to power police and military radios, except without good rectifiers we used generators. Yes, this rig makes an annoying acoustic whine: we kept it in the trunk. Yes, the DC motor and generator make a nasty electric whine: you could usually spot a motor-generator transmitter by the whine in the signal. Modern alternators would be somewhat better, but overall this is an un-fun way to convert DC to another voltage.

> rectifying AC from the wall for B+

Through a power transformer, OF COURSE.

And if you have wall-power, starting from rotary alternator sine-ish power is a lot easier than working with square waves to get clean DC.

If you want 300V DC battery-clean, take twenty-five 12V batteries. Charge them in parallel, then wire them in series. That's how the BBC did it before good capacitors were available (except they used a 300VDC generator instead of re-wiring to use 12V chargers). A 7VA emergency-lamp battery will carry 250mA for about 10 hours (you don't get rated capacity unless you let the voltage sag well below 12V). That's around $600 of batteries.

Tube preamps are not fussy what the exact voltage is, as long as it only changes very slowly. IMHO, regulating power for preamps is lily-gilding. Use an R-C or L-C filter, taking advantage of the very low cost per V*uFd of modern Asian electrolytics.

Heater power does not have to be so clean either. Yes, DC heat does allow sloppier wiring, if it is "reasonably clean" DC. If you follow general AC-heat layout (keep heater wiring away from low-level terminals) then 0.1V of ripple should be inaudible. You can get there with R-C filters and mega-caps, or reduce buzz past zero with regulators. The main benefit is in the lowest-level stages: using DC to heat output tubes is probably a waste of good silicon, since they should not hum with raw AC heat.

And remember that AC heat was used for all audio, some of it very top-notch, for decades. If the Dead Men could do it, so can you.

Thank you Bill and PRR,

Yes, that sentence about rectifying AC from the wall was typed in a hurry. What I meant to say was Build a PSU for B+ the conventional way, using wall AC as a source, through a power transformer and then rectified and filtered.

Each of 8 to 10 pres will be 2 stages of ECC99 with a total current draw somewhere around 20mA. Some may have 2 stages of 12AX7 with a total current draw of around 2mA. Maybe 250mA is a bit big, but my motivation to use batteries is really dwindling fast.

I really want to nab that last bit of filament hum I can hear when I use really heavy compression on a drum kit with lots of mics. I just thought using a battery for b+ as well would be a nice touch, but it looks to be a waste of time and money and a safety concern at that.

PRR, I get about the same audible hum with my super-filtered dc filament supply as I do with AC, though the dc is a bit better, but not much. The battery dc kills it completely, though.

A lot of it is the extreme ITB compression I'm using multiplieid by several channels. I'll still have soft hiss, but I want the soft hum gone. One deep cycle car battery from wal mart fixes it. But what a pain to haul it out to the garage for a charge! Do you still think I should/could pursue AC heaters?

[quote author="PRR"] If the Dead Men could do it, so can you.[/quote]

:grin:

Quote Of The Year so far PRR!

:thumb:

Mark

[quote author="deanp920"]Thank you Bill and PRR,

I really want to nab that last bit of filament hum I can hear when I use really heavy compression on a drum kit with lots of mics. I just thought using a battery for b+ as well would be a nice touch, but it looks to be a waste of time and money and a safety concern at that.

PRR, I get about the same audible hum with my super-filtered dc filament supply as I do with AC, though the dc is a bit better, but not much. The battery dc kills it completely, though.

A lot of it is the extreme ITB compression I'm using multiplieid by several channels. I'll still have soft hiss, but I want the soft hum gone. One deep cycle car battery from wal mart fixes it. But what a pain to haul it out to the garage for a charge! Do you still think I should/could pursue AC heaters?[/quote]

It sounds as if the hum might not just be ripple in the supplies---those can be made arbitrarily quiet in terms of across-the-terminals, particularly if you use active regulation. You might also be dealing with some common-mode coupling, which the isolation of the battery avoids almost completely (since it still has capacitance to the outside world no supply is completely floating in that sense). Batteries are not the ultimate low noise power source either btw, although they are definitely hum-free ;-)

In any case, for one potential improvement look at biasing the filament supply a bit positive with respect to the voltages on the most sensitive stages' cathodes. Often the filament-cathode system acts like a litttle direct-heated diode, particularly when there is cathode material deposited on the filament. There is a brief discussion in RDH 4th pg.1197. The effect is most obvious when you have some unbypassed cathode resistance.

In the c800 microphone thread there were some posts about the microphone having less noise with the use of supercaps on the heater supply. The sony c800g schematic looks like it used super caps.

I built what I thought was a very clean fil supply. When I was working on getting the noise as low as I could get it I could hear low level noise in closed back headphones in a quiet room. I added two supercaps to the fil output and it went away.

I did not have any extra wires in the cable to raise the fil about ground otherwise I would have tried that experiment.

I think it was "noise" from the diode bridge.

Noise on the fil supply seems to have more effect than the B+ noise sometimes?

The Dead Guys were hampered by the difficulty of getting good low voltage d.c. from the mains, so explored the most effective techniques of using a.c. The traditional a.c. filament supply has a potentiometer across with the wiper to common (or the bias voltage if used in that fashion), and the a.c. coupling from heater to cathode is nulled out as best as can be done.

The other problems from E and B field emissions are dealt with by routing and twisting conductors judiciously---even shielding foe the E fields if necessary.

The nice thing is that if the mains are not too noisy the hum is mostly fundamental frequency and doesn't radiate much E field. But with solid-state rectification inside and lots of noisy loads outside, the noise on the a.c. is nowadays likely to be quite rich in harmonics which both radiate/couple better and are more obnoxiously audible when they do.

Also there is inductive coupling within the tubes depending on the filament construction.

I think we're probably more careful with the B+ as it is easy to see the effects of hum and noise. The heater supply stuff is a bit subtler, and also I think there's sometimes a reluctance to believe that the Dead Guys knew what they were doing---the "I'm going to fix all that old-hat stuff with a three-terminal regulator" attitude.

[quote author="deanp920"]I really want to nab that last bit of filament hum I can hear when I use really heavy compression on a drum kit with lots of mics. I just thought using a battery for b+ as well would be a nice touch, but it looks to be a waste of time and money and a safety concern at that.

PRR, I get about the same audible hum with my super-filtered dc filament supply as I do with AC, though the dc is a bit better, but not much. The battery dc kills it completely, though.[/quote]

Are you sure the hum is 120Hz (100Hz outside USA) and not 60Hz (50Hz)?

Peace,
Paul

Hi Paul,

Are you sure the hum is 120Hz (100Hz outside USA) and not 60Hz (50Hz)?

Click to expand...

It's 120Hz ripple last time I measured, which, honestly was a while back.
How would using a battery eliminate 60Hz hum?

P.S. I thought your "Big Mike and the Jimmy" was outstanding. I hope to hear/build that preamp someday.

Bcarso, I'll have to try biasing the filaments like you suggested.

Thanks,

Dean

I think it is cool you are using
batteries to power tubes.

But you need to be aware lets say you use
a 12 amp hour motorcycle battery.

If that battery can put out 50 A peak to start
the bike then remember that number.
You will need 22 batteries to get 303.6VDC.
You must recharge lead acid batteries when the volts per
cell drop to 2.0 volts, charge then do not wait.
So your pack will be down to 264VDC when you must recharge it.
Lets assume the pack is 300VDC and it can deliver 50A then
you have 15Kilowatts of power you are playing with.
If you short that pack or any of your wiring or
devices you will easy have a fire on your hands.
You must use a DC fuse that is capatible of breaking that
peak current. Thay make them but you will pay
about $50USD. If the fuse tries to break the fault and it cannot
and it starts arching over your load is still not disconnected!
You will still have a fire. Your problem is not
the current as a 12V car fuse works good (at 12 volts).
It is finding a 300V dc fuse that can interrupt 50KVA or more.

Now if I have not scared you yet here comes battery gases.
Remember you need to float charge wet lead acid batteries
about every 2 weeks for several hours. You will boil the acid in the batteries and convert all the Lead sulfate back to lead.
That is good but charging batteries will generate Hydrogen.
Do we need to say Hindenburg?
When the rouge spark ignites the gas the batteries may
explode and now you have sulfuric acid everywhere.
Not a good sight. Place your batteries in plastic tubs.

Every day large battery rooms work for Solar homes and Telephone
central offices. The big UPS industry uses car size batteries to get
440VDC to run the magawatt UPS's till the generators have time to start.
It can be done but know what you are getting into and follow
all safety rules for working on batteries.

Now you can use Nickel Metal Hydride "D" cells of old cell phone battery packs. You can even use gell cells. But charging this string in series will be a pain in the butt. Use constant current charging (trickle) remove the charger to use the audio equipment then replace the charger. Keep the batteries always on the charger and you may get away with it.

It can be done but you have just started a big project be carefull.

For grins just build a choke loaded supply with RC filtering.
A friend you lurks here make a tube RIAA preamp. With the
power supply in the box and he has no hum! it can be done.
You just have not done it well enough.

Thanks,

If I do go the battery route, i'm down to one deepcycle 12V car battery to power 1.5A to 4A worth of filaments. At this point, its about how far I want to tear into things and how much time I want to spend solving this tiny bit of hum.

Around this topic, my dad told me about some guys he worked with that had a several thousand volt truck mounted battery supply short out. All they could do was run like hell from the sparks and fire.

The safety factor was the whole reason I came here asking about a DC-DC converter to go from no more than 60 volts and preferably 12volts to 300Vdc.

With that in mind , explore the DC converter route. They aren't as hard to build as what it appears. There are lots of resources on the 'net on this type of info. A converter operating at 200kHz filters pretty well, and all you have to worry about is emi radiation. A well shielded box for the power supply with feed-through caps should work well.

I just designed my first buck converter, I was always scared of switchers...

We'll see how well it works when PCBs get here

> If that battery can put out 50 A peak to start the bike then remember that number.

Well, no, it is worse than that. But let me shift to a car-battery just because I know the numbers.

A car battery will supply 200A-400A while staying above 10V-11V. Just for round numbers, say 200A at 2V drop, 10.6V still available to light the sparks and keep the starter relay pulled-in.

So the internal resistance is 2V/200A= 0.01Ω, 1/100 ohms. If you drop a perfect short-circuit on it (even a yard of battery cable, as it is used in a car), you can pull 1,200 Amps!

Going back to your "50A" motorcycle battery, say 300 Amps into a short.

And when that happens, you have 12.6V*300A= 3,700 Watts mostly dumped into a small bucket of strong acid. 3,700W is bigger than any two burners on my stove, and a cycle battery is a very small pot. How long does it take to boil and spew? The bursting may not be wildly spectacular, but the hot acid eats things fast, and skin very-fast.

And as you say: near the end of charge, a lead battery makes Hydrogen gas. It may not be as dangerous (IMHO) as gasoline, but is invisible and odorless so you don't know when you have too much hanging around.

Between electrical, acid, and fire issues, building codes confine large battery racks to specially prepared rooms. "Sealed" cells get a little slack, because they are less likely to spill or make gas, but I don't think they are allowed in nice places without proper fusing to reduce electrical heating (which would burst any "sealed" cell).

Common household fuse-box fuses (which are no longer common) can easily stand the peak current, and are not seriously overvoltage (a 120VAC line with motor load can kick to 300V when broken at high current). They don't come smaller than 15A, which is still too big for what is only a 0.25A load. 3AG fuses like we put in boxes are not rated for interrupting huge (10,000A) peak current, because they are nominally protected from feeder currents by the basement fusebox and house wiring. A 3AG will "probably" fail cleanly at 300A, though it may shatter; I don't think it is rated for such work. Fusing a 90KVA source is indeed tricky.

A 12V 5A battery supply for heating is more reasonable. A 12V 10A car fuse is ample protection. A single cycle/car battery would last a session. Modern semi-sealed batteries on gentle charge don't make much Hydrogen. Keep it in a cat-litter box in case it seeps acid (stuff happens). I still think that heater-hum is curable without such heavy-weight measures. (Maybe I just don't run enough compression....)

[quote author="deanp920"]Hi Paul,

Are you sure the hum is 120Hz (100Hz outside USA) and not 60Hz (50Hz)?

Click to expand...

It's 120Hz ripple last time I measured, which, honestly was a while back.
How would using a battery eliminate 60Hz hum?[/quote]

If it was 60Hz hum rather than 120Hz, the cause would very likely have been a ground loop rather than inadequate filtering.

Let me ask this: are you regulating the filament supply now? And how is the whole thing grounded? In many setups, there's a main grounding point, usually at or near the bottom of the main filter capacitor, and both the regulator and load are grounded to that. If that's how you're hooked up, try instead connecting the ground terminal of the regulator to the load's ground point but not to the main grounding point. The load is then grounded to the main grounding point. That has eliminated hum in some projects I've built.

P.S. I thought your "Big Mike and the Jimmy" was outstanding. I hope to hear/build that preamp someday

Click to expand...

Thanks for the good words!

Peace,
Paul

Dean;

After all the scare issues I do plan on a string
of 4 gell cells 12 volt at 7 amp/hour to get me a
clean phantom supply.

Yes I will place the "UPS" batteries in a plastic tub
and fuse the whole pack for excessive currents.

That's pretty safe and sane. Not much gas to worry about and they rarely leak electrolyte either.

Get a good charger and charge the packs in parallel---constant (accurate!) voltage with current limiting works well.

Avoid Li-ion unless you really know what you are doing :roll: Most manufacturers won't let you get your hands on the raw cells anyway, so there's little worry there.

By the way, there is a belief that batteries are inherently quieter than any other power source. This is a myth, one rarely challenged. There is no denying the isolation though.