Tube Heater

.

80hinhiding said:

I tried to turn the heater on with a voltage divider last night, using two resistors but it didn't work.  I'm guessing it's not that simple.  

Click to expand...

It's even simpler. You only need one series resistor for the heater. Assuming your 30V supply can supply enough current and you know ohm's law to calculate correct resistor value.

80hinhiding said:

I tried to turn the heater on with a voltage divider last night, using two resistors but it didn't work.  I'm guessing it's not that simple.   My hope was that I could use the 30V supply I am using for the other test circuits and incorporate a tube for comparison.  Put 30V on the plate and 12V on the heater.

The only way I've gotten it to work is by using a specific power supply for the heater.

Click to expand...

Voltage dividers work fine as long as you're not drawing any current from them (from the midpoint). If you are, then don't use a voltage divider! This is an important lesson for the future. In some cases where the current draw is low enough, and it's AC current, you can get away with it as long as the midpoint of the divider is bypassed with a suitably sized capacitor. The capacitor will slow down the rate of change, so the voltage will only change a tiny bit with the signal.

12AX7 heaters require 12.6V (not 12.0V) at 150 mA when connected in series, so they act much like a 84 ohm resistor (when hot).

But of course, that heater can be one of the resistors of a voltage divider! The other one would have to drop the remainer of the voltage 30V - 12.6V = 17.4V and the current is still 150 mA, so that works out as 116 ohms. That's not a regular value, so try 120 ohms. Watch out for the power requirements: This works out to be 2.6 watts ! That resistor will get hot. I would use a 5W part.

> using two resistors but it didn't work.

Three resistors. R1, R2, and Heater.

Heater is 12.6V/0.15A or 84 Ohms.

If R1 is 133K and R2 is 100K, yes it reads about 12V.

Now put Heater across R2. It is no longer 100K but 83.9 Ohms. You get a few mV, not 12V.

If the load is *steady*, just use it for R2 and compute your R1.

1.33 and 1.0 Ohms will divide 28 to 12 even with a 84 hanging on the 1 Ohm. But the current is over 10A and the power is over 300 Watts! Kinda a waste for a 2-Watt load.

100 Ohms will be close-enough. However what is the dissipation? Well over 2 Watts. A 2W part may burn-up in minutes. A 1/4W part will just go ffffpht. 100r 5W will be good for decades.

80hinhiding said:

At what point is using a voltage dropping resistor or resistors not a good idea in heating a tube filament?
...
Adam

Click to expand...

In the same moment you start to care about efficiency but I guess you wouldn't be messing with tubes at all in that case.

  Other, and the most important factor probably, is it's just not practical, you need a much bigger transformer and you are heating more the resistors than the tube heaters, most power in tube gear (excepting power amps) goes to the heaters, if then you add a lot of waste you end with a much bigger transformer than you would need and ends being more expensive than getting a custom transformer.

  A very common approach to this is using more than one transformer, 12V transformer to feed the heaters with it isn't hard to find. Then from it you could feed a second 12V transformer but backwards and get your mains voltage again, rectify and you get a pretty usable plate voltage. If you are in a 220V or 240V country it might be too high (or not), if in a 110V you could get away with the 150VDC you have or even use a doubler in some cases.
  I'm with 220VAC mains (hopefully) andI had to use 220V:12V-->9V:220V to get what I needed for the preamp I was building, I don't recommend doing so, but it was the only way of getting where I was going without a custom transformer. I guess I could use a 18V transformer and a doubler to get close to that but the space was tight and the transformer selection small.

JS

80hinhiding said:

Well, the heaters looks nice ha but the tube amp I just setup was an epic fail.  Signal severely attenuated and filtered. 

Click to expand...

With low plate voltages you can't usually avoid grid current and input resistance of the grid will be very low. If you're designing with Spice I don't think there are any tube models that will give results that are anywhere near real world with low plate voltages. Behringer and ART have used bunch of low voltage tube circuits in their products. Study their schematics and steal.

80hinhiding said:

I tried to turn the heater on with a voltage divider last night, using two resistors but it didn't work.  I'm guessing it's not that simple.   My hope was that I could use the 30V supply I am using for the other test circuits and incorporate a tube for comparison.  Put 30V on the plate and 12V on the heater.

Click to expand...

I wouldn't do that. The reason is the apparent resistance (U/I) of teh heater changes significantly with temperature. When you start the heater is cold and shows about half of its operating resistance; being in series with a resistor (which doesn't change that much), it receives initially a lower voltage than expected, about half, so it takes a long time to rise in temperature. When it has finally reached its operating temperature, it will receive more voltage, in a cumulative way, so the tube may be overheated, which is not good for its life expectancy and its performance. Using dropping resistors on heaters (or incandescent bulbs) is OK when dropping a fraction of the voltage, but not for so much difference.
This is in addition to the fact that your voltage-divider calcs are inadequate and your PSU not properly dimensined, as others have mentioned.