Using a cathode Zenor instead fixed grid supply for class C protective bias?

[K1JJ]

Hi All,

I wonder if there is any problem using a simple 50W 100V zenor for protective bias in series with the cathode circuit instead of a fixed grid supply?  The 4-1000A in class C plate modulated service needs about -200V total bias. I plan to use grid leak for the other 100V, of course.

In the past, for linear use, I've found the fixed supplies would charge up and slowly rise under carrier on conditions. This required a regulated grid supply.  Can't remember if charging was a problem with class C.  I was hoping a simple zenor in the cathode would solve this problem, and simplify things by eliminating the whole fixed grid supply.

How do you protect the zenor from spikes? Put some capacitors across it?  What value?

I have a few 100V, 200ma power zenors -  I could put three in parallel to get the 600 ma spec needed for the 4X1 in class C.  Only one 75V unit will be needed for the single 813 class C rig.

Any suggestions or potential problems doing this?  The only problem I can see is if it fails/shorts. It is more likely to go than a standard grid supply, from my personal experience.

73,

Tom, K1JJ

[nu2b]

Tom,
There may be a few reasons that a zener would not be good.
1) The zener voltage even though only about 4% of the plate supply does subtract from the swing. 1.04*1.04=1.08 means that approx 6 to 8% power loss compared to the cathode being at ground. The 4-1000 operating at 3KV is already not that great in efficiency since they were designed to really start singing at 4 to 5KV.
2) The peak plate current for class C during the half cycle is 4 to 5 times(1.6A)  the dc average (.333A) plus the peak screen current of maybe another 20%=1.6+.4=2A. A zener in the cathode with 50 ohms effective rs would spike up to another 100V unless heavily bypassed.
3) What about your current SG power supply design? Is it about 600v at 100ma?. You might be able to use choke input(in the neg leg) for good regulation and float the power supply, with the zeners in the neg side which would supply the -100V but only require maybe 100 mils or so to support the screen current (+500V@75ma) and the grid drive current(-100V@25ma).(guesses)
4) How about a clamp tube? Simple and they seem to take a lickin and keep on tickin. The screen supply transformer may have an unused fil winding which could be used for the purpose.

Like most designs, there are about a million solutions. Just keep asking around.
Regards,
BobbyT

[WA1GFZ]

Tom,
You will have to increase the screen voltage by the zener bias. You can parallel zeners but put a resistor in series with each one to help them share current.
100V at 600 ma is 60 watts extra loss and less voltage on the final. But it will be very effective protection bias if you don't mind the loss. I would put an effective RF bypass across them.

[K1JJ]

Tnx, Bobby and Frank -


OK, I see... for a GG linear, cathode zenor bias is OK cuz the required bias is small, thus the zenor loss is small. But when using a grid driven amp with screen voltage, it takes lots of bias and the higher voltage zenor makes a bigger loss.

Looks like it wasn't such a good idea then for a class C amp... the 60W loss across the zenor is too much for me.  I mean, if it's preventable, then it's worth doing something else. I'll just make up a simple fixed grid supply for each and call it a day....

The clamp tube is interestsing too, Booby. I'll have to think about that vs: the grid supply and decide which is the best approach.

What tube wud you recommend and how wud you configure it?  This wud permit 100% grid leak bias - something I've done in the past with good results.

Thanks much for the opinions.

T

[WA1GFZ]

Tom,
Why not use those zeners to shunt regulate your grid bias supply? just use a series resistor off your bias supply with the zeners to ground. The point between resistor and zener goes to grid. Solid bias and shunt regulator to hold it solid. 

[K1JJ]

Yeah, that's a good idea Franz.

Looking at the curves, I've decided to use 24V transformers for the fixed bias, giving roughly 34VDC with cap input.  This is enuff to keep the plate current at a reasonable level if drive is lost on both the 4X1 or 813 finals. Plus it will permit about 75% grid leak bias, which is what I like.

I will run some tests before I regulate these protective fixed bias supplies, but do I need to regulate them?   I remember an isolation circuit I used involving a steering diode and 10H choke that kept the bias supply from charging and creeping up in a grid driven linear. But I can't remember if I needed regulation in class C grid supplies in the past.

Does anyone do that?

T

[WA1GFZ]

I would think you terminate the grid leak resistor to the fixed bias so the bias can only increase due to current through the resistor while the low side of the resistor is
held in regulation by the shunt zener and negative source also resistor isolated from the zener. You need the fixed source to the zener when the drive goes away to
provide protective bias. Grid current falls to zero so there is no voltage across the leak resistor only the voltage of the zener. 

[Tim WA1HnyLR]

Yo Thomas, A protective bias zener would work in cathode return. but consider the following  1, you are shaving off 100v of valueable plate voltage. at 500MA plate cuntrent you are pissing away 50 watts in heat. 2, If the tube farts the spike may take out the zener losing protection.. If you do go this route place at least a 1000  mfd cap across zener, De Tim WA1HnyLR

[WA2IXP]

I have used zeners(and vr tubes) across bias supply for protective bias. When the tube is cut-off there's no current flowing so regulation is unimportant. When tube draws current the zeners hold and rest is developed across grid leak. In my case for an 812 I used 50 volts of zener and the rest developed across resistor to get around 150 volts.  Works fine   Jay-

[K1JJ]

Yo Thomas, A protective bias zener would work in cathode return. but consider the following  1, you are shaving off 100v of valueable plate voltage. at 500MA plate cuntrent you are pissing away 50 watts in heat. 2, If the tube farts the spike may take out the zener losing protection.. If you do go this route place at least a 1000  mfd cap across zener, De Tim WA1HnyLR

Nice surprise to see you herer, Tron!

Yes, looks like a grid supply is the way to go. I've pulled out the little transformers and caps already.

Though, I plan on only about 34V of fixed bias, so that wudn't be so bad as a zenor - but every bit counts.... :-)

Always looking for ways to short cut and simplify the conventional here.

73 and talk at ya soon on the air, OM,

T

[Bill, KD0HG]

One mo thing, Thomaso..It's a good idea to use some sort of shunt regulation of that bias supply, you want to make sure that yer grid current is sunk somewhere. Or bleed the heck out of it. Necessary for Class C amps. Run some numbers to see how much the voltage across the bleeder is going to rise with your rated grid current going through it
All easier done wth a low drive 813 than with a big grid current triode.

A clamp tube on the screen supply is a good way to protect all, too. You could do 100% resistor bias and ditch an extra bias supply that way.

Just some thoughts.

-de hg

[K1JJ]

Thanks, Bill -

Think I will build up the grid supplies and use shunt zenor regulators. Simple enuff.

I've got so much work here to do it's laughable. Busy all winter for sure. Five rigs to build. I have five 28" X 17" aluminum plates with frames and casters sitting on the floor filled with parts to mount. All broadcash style mounting construction. The tube chimneys / air systems are finished... making the bracketing now for other parts..   Two big linears, two plate modulated rigs and one PDM.  All tubes, of course. [of course]   The parts cellar is showing big gaps for a change.

T