Fixed bias failure B+ switch off protection

[Brit01]

Looking for a simple circuit that will switch off the B+ if the fixed bias fails or falls too low.

Planning to put the relay on the primary of the B+ tranny.

If the fixed bias falls too low then it triggers the relay.

In theory will this work?

Any suggestions/opiniones/experiences?

[tomchr]

The NPN will always be off as it's base voltage is lower than its emitter voltage. So no, the circuit will not work.

Another issue is that the base current is not limited. You'll need a resistor between the base of the NPN and the bias voltage.

To make this circuit work, I suggest changing the NPN to a PNP and "hanging" the circuit from the ground rail. The bias voltage pulls the base of the PNP low, thereby turning it on. If the bias voltage isn't low enough, the PNP will turn off.

You'll have issues at start-up, though as the bias voltage will be out of spec at that time.

[kevinkr]

Originally Posted by Brit01

Looking for a simple circuit that will switch off the B+ if the fixed bias fails or falls too low.

Planning to put the relay on the primary of the B+ tranny.

If the fixed bias falls too low then it triggers the relay.

In theory will this work?

Any suggestions/opiniones/experiences?

The transistor symbols are incorrectly drawn, and in fact the base of the 2N3906 (PNP!) needs a resistive voltage divider ahead which is not shown here.
I used a very similar in fact almost identical circuit to protect against bias failure in one of my designs, and not everyone was able to get it to work reliably. (Start by changing the direction of the emitter arrows and you'll be heading in the right direction.)

In terms of B+, it will not be applied if bias is not present, and usually the bias supply should be arranged to come up to full voltage long before B+ is present.

Simulating a circuit in LTSpice should allow you to come up with a design that actually works. There are lots of ways to do this. You might even consider an LM311 comparator or similar. (watch input common range!)

[Brit01]

Yes you're right. The arrows were incorrect. My error.

Let me look into this further in the evening.

Thanks for the responses guys.

[Brit01]

needs a resistive voltage divider ahead which is not shown here.

I placed a divider here (1K/10K), re-designed the pnp/npn arrows and it seems to work in the simple simulation. Relay cuts in anything below -12 volts.
I can always set up a simple proto on a breadboard.

Simulating a circuit in LTSpice should allow you to come up with a design that actually works

I'll look into this. Is there a freeware copy available?

[kevinkr]

Originally Posted by Brit01

<snip>


I'll look into this. Is there a freeware copy available?

LTSpice is gloriously free, extremely powerful, and easy to get here: Linear Technology - LTspice IV Downloads and Updates

Runs in Windows and Linux (under Wine)

[SpreadSpectrum]

I have taken the approach of sampling cathode current and feeding that to a comparator. This will save the tubes in the case of a bias supply failure and save the output transformer in the case of a tube short. Two birds, one stone.

[Brit01]

I have taken the approach of sampling cathode current and feeding that to a comparator.

Would this work in a circlotron design? Do you have more details?

Thanks