Linear voltage regulator failing - Questions and help

[TheBathroom]

Hello everybody, I know I don't post often, but I am a frequent browser on here.

I have a 7806cv voltage regulator (6v, 1.5A) that I am using to power this project of mine that uses 4 tubes. I decided to wire them all in parallel, but with all four tubes in I got a weak glow when it was on, then nothing. I knew something was wrong with the regulator of the circuit, but I can't figure it out.

My circuit information: I have 9v going in, then 6v going to pin 9 of each tube, 4-5 to ground, and I have 150nf going from input pin to a Schottky diode to ground (figured this would be good for polarity protection, but its probably redundant) and 100nf on output pin to ground. Also I have a heatsink on there, this guy here:
http://www.mouser.com/Search/ProductDetail.aspx?R=577102B00000Gvirtualkey53210000virtualkey532-577102B00

I expected it to get hot in the first place, so I got that heatsink. When I checked it after my tubes 'died', I nearly burnt myself. I took two tubes out so that there were only 2 left in there, it should only draw 600ma, and they worked fine. However, my DMM read 5v or so at pin 9.

So if someone can kindly give me advice on what I should do, I would be very grateful. Is it too much of a load for the regulator to handle, or is overheating you think?

[samhay]

What are you powering the regulator with?
Is it giving you 9V on the voltage regulator's input pin with valves (sorry tubes) plugged in?

[merlinb]

You are asking the regulator to dissipate 1.2A*3V = 3.6 watts. That little bolt-on heatsink is probably borderline.
Also, some regulators will refuse to start into the cold load of heaters. You can sometimes fix this by putting a power diode in series with the regulator output, but that will lose you about 1V (which can be made up by putting another diode in series with the 'adjust' leg of the regulator).
What are you powering this from? Not a battery, right? Do you even need a regulator or would a power resistor do the job more simply?

[amptramp]

Heaters take several times their rated current when cold.  If the regulator cannot supply enough current, the heaters never get warm enough to go to high resistance and the current regulator and thermal shutdown circuitry in the regulator drops the current before it can come up.  The simplest answer in your case would be to split the heater circuit and add another regulator.

And yes, that heatsink is not likely to make much difference.  You need one of the extruded aluminum ones, even with two regulators.

[PRR]

Which is why you want RESISTORS. (As Merlin said)

9V drop to 6V is 2.7V

The most-pop small tubes are 0.3 Amps.

2.7V/0.3A is 9 Ohms (10 will work).

2.7V times 0.3A is 0.8 Watts (use 2 Watt part).

Per tube!

Yes, you could do math for four tubes and arrive at 2.25 Ohms 3.24 Watts. Which is such an odd value, that you will probably cobble it up from other more-standard values. Such as four 10r 2W parts.

And if you use a common resistor, but leave a tube out (or it fails), the other tubes run hot, short-life. Individual resistors, each one can run alone, or not.

Resistor also reduces the start-up surge. IMHO, not enough to make a darn bit of difference, but it feels good.

Do NOT fret about 6.0V on small "6.3V" tubes. All of these are over-heated at 6.3V 0.3A. In earlier days they were often used at 5.5V when that was convenient. You have to take 12AX7 and kin near 5.0V before emission falls-off enough to matter.

[TheBathroom]

Which is why you want RESISTORS.

9V drop to 6V is 2.7V

The most-pop small tubes are 0.3 Amps.

2.7V/0.3A is 9 Ohms (10 will work).

2.7V times 0.3A is 0.8 Watts (use 2 Watt part).

Per tube!

Yes, you could do math and arrive at 2.25 Ohms 3.24 Watts. Which is such an odd value, that you will probably cobble it up from other more-standard values. Such as four 10r 2W parts.

And if you use a common resistor, but leave a tube out (or it fails), the other tubes run hot, short-life. Individual resistors, each one can run alone, or not.

Resistor also reduces the start-up surge. IMHO, not enough to make a darn bit of difference, but it feels good.

Do NOT fret about 6.0V on small "6.3V" tubes. All of these are over-heated at 6.3V 0.3A. In earlier days they were often used at 5.5V when that was convenient. You have to take 12AX7 and kin near 5.0V before emission falls-off enough to matter.

Thank you so much for your response, and for everybody else who took the time to respond. I'm still learning my Ohm's law though, and I appreciate your suggestions. Yeah, to use a regulator would need a colossal heat sink, and I don't have room for that. I should have thought about researching how I am going to heat that many tubes at once. Anyways, thank you all again very much.

[PRR]

> regulator would need a colossal heat sink

With either (linear) regulator or resistor, it is the SAME heat either way.

But a regulator will fail (or shut-down) when "hot". A resistor can run HOTT!, because it is a stupid thing and doesn't have to "think" like a regulator does. You can pack more heat in less space. However that heat will still warm everything else in the box.

An alternative is a "switching" regulator. Young-guy (and a few oldsters) are very comfortable dropping 9V to 6V with hardly-any loss using chips and coils. You can even find them for $4 on eBay. Myself, I am not comfortable enough to be confident buying from the slim descriptions and specs that these Asian suppliers give. Maybe someone else here is.

[PRR]

Also check your "9V" (as Samhay said). Those regulators need near 3V of "headroom" to do their job. 9V-6V is 3V barely. And 1.2 Amps is high for a "9V pedal supply". Some go to 1.7A, others can't hold 0.6A. If the "9V" is falling 8V 7V 6V, the regulator can't do its thing.

If the heaters were the only issue, and you had 8V or so full-load, resistors can be made smaller so they drop less and you end up in the 6V area. However in this case I assume you have Plates to feed, 9V is gross under-volting, and less is worse, so you may need a beefier supply to avoid supply droop.

[TheBathroom]

Yeah no worries there. I'm using a One Spot that measures 9.24v unloaded, and supplies 1.7A. Plus the plates don't draw that much current at that voltage anyway, and I have isolated the plate feed with diode and caps with minimal voltage drop.

I'm not exactly building a valve pedal, so I guess in that respect I'm cheating  but I can use this page for reference when building something that needs a regulator.

[miech]

Doesn't the LM7806 datasheet have a high current example in the typical appplications section? Most voltage regulators I know have at least one mentioned in the datasheet. It will obviously increase the footprint, because more components are needed. There is a second transistor in the design, which obviously will need a large heatsink.

Tracopower makes some LM78xx replacements with switching regulators inside. They have the same footprint and pinout, and 95% efficiency. They might latch up however.

Tube filaments should be powered by their design voltage +/- 5% for maximum lifespan. For a 6.3 volt heater, a 6.0 volt supply is just barely within that specification.
I don't know how that translates to new production tubes however, that info comes from a book from the 1950-ies.