The AM Forum

Hello all!

I have a question on this old push-pull 809 transmitter circuit.

What is the specific function of the 40 ohm center tapped resistor?  Is it only there because of the cathode bias resistor?  Could it be omitted if the cathode bias resistor is not used (OK if you keep it down to 600 VDC on plate).

Thanks!!!

http://n4trb.com/AmateurRadio/RCA_Ham_Tips/issues/rcahamtips0101.pdf

The 809 does not have a cathode.   The resistor, along with C18/C19, is there to provide a ground reference and prevent 60 cycle AC being impressed on the output of the amplifier, it has nothing to do with having, or not having a self bias resistor in the filament circuit.   

Yes R9 does help set bias for plate voltages higher than 750 as shown in schematic and as written in the schematic's footnotes.  Yes the 809 has a cathode. The hot filament is the cathode.  Filament voltage is only used to heat the cathode ( filament wire) to incandescence so that the filament as cathode can emit electrons, such thence to be controlled by the grid and onwards to the plate or anode.

It is also there to allow or act as a channel for electrons from ground to get to the cathodes, split both ways for two tubes, if there is no center tap on the filament transformer secondary.  If there is a center tap, R9 may be omitted and RF bypasses capacitors placed appropriately.  That will work fine for the 600 VDC that 4WAX proposes.

If you do want R9, two 20 ohm resistors will work fine since the tap on the 40 watt one is probably not exactly centered. -- Tolerances and all that.
 

I sure hope you build this rig!  What a great old buzzard transmitter. 

The 809's are what is call a directly heated tube since it has a filament, grid and plate.  Unlike tubes with a separate cathode, one must provide a return for the filament.  That is the primary function of the WW center tapped resistor.  I have a transmitter with a 210 in it and its the same arrangement.  Anytime you have resistance in the return path, you will get a voltage drop based on the current through the tube, so yes that is a form of bias.  This transmitter has cathode bias along with grid bias provided  by the resistor and fix bias supply.   

Joe-W3GMS     

Oops Joe, I was editing, typing essentially the same as you posted. Sorry.

Like minds Rick!  Hope all is well....

73, Joe-GMS

Dear Rick:

I believe R8 is the cathode bias resistor, and not R9.......though R9 might contribute a little smidge.

Have a look at the 6L6/809 transmitter in the '47 West Coast book.  No 40 ohm CT, but two caps establishing an RF neutral point.

See page 290 on pdf reader.

Matt

http://americanradiohistory.com/Archive-Radio/Handbooks/Radio-Handbook-11-1947.pdf

The R9 setup could be a low cost or convenient substitute for a center tapped filament transformer. I do this with 3-500Zs (using two 10 Ohm resistors in series) since the transformer has no CT and never have had a problem, and have found no weird things going on even after 10 years. Go for it!

Matt,
Yes "R9 helps set the bias" as originally written and necessary for higher plate voltages.   ;D. --And whether pseudo filament center tap is required or not.

Nice rig from yesteryear and I also hope he builds it and has fun.

Also Much Thanks for the great " West Coast"  Manual link. I have a later edition but will look through the linked one to see what's changed.

I really enjoy these along with ARRL 's versions.

Joe,
pretty much same over here. Appreciate your kind thoughts. Best to Martha and you too, as always. I see your making the fester rounds in good form.

It seems to me that you would want the center tapped resistor, or a pair of 20 ohm, if you did not have a CT filament transformer.

If you run 600 VDC with no cathode bias, the two capacitors would get it done.

My current scheme involves driving a pair of 809s with an 807 in class A.  I surmise the driver circuit in the RCA sheet is in class A.

For modulation, I am going to use the Dick Genaille shunt feed scheme, with a solid state hi fi amp and a Hammond hi fi transformer.  By my reckoning, 40 to 45 watts of audio will get it.

My next question concerns harmoni attenuation.  How much from swinging link circuit?  My plan is to have the swinging link tank coupled to a second swinging link circuit which will serve as a transmatch.

I expect the two would get you 45 dB or more.

If you're running push-pull, the second harmonic suppression is less critcal.

Yes, push-pull takes out even if done well.

I am wondering about the numbers for odd.

It has a tuned output network which by definition limits all frequencies but the L/C/R fundamental.  Suppression of all other freq's. depends on the loaded Q of this  parallel tuned network.   There are equations and charts showing such with a search. The circuit as shown is adequate for amateur communication.  ;D

Particularly See bandwidth section at end of following:

http://www.electronics-tutorials.ws/accircuits/parallel-resonance.html