[HBR] Pullen mixer

Fri Nov 25 16:29:52 EST 2011

Don

Interesting stuff !

I was checking on pricing for the 6GM8. The audio folks have taken this tube since they can build tube preamps using low voltage and not hurt themselves. These folks are also seeing good signal to noise with these tubes.

Bad news is the audio folks have driven the price of these tubes very high.

The data I have so far has the 6GM8 at 2500 micromho's for transconductance. The 6ES8/6DJ8 has 12,500 micromho's. But, I think the key to low tube noise is low plate resistance.

I plan to get back into my HBR and my 75A-4 using the Pullen Mixer, but haven't gotten the time just yet.

I see that the 6ES6 is the low voltage pentode used for RF/If amplifier in the same car radio applications. May be a good mate for the RF amp to your Pullen Mixer.

Best 73's

Paul - WB2NMI

----- Original Message -----
From: donald haworth <donmhaworth at gmail.com>
To: HBR Receiver List <hbr at mailman.qth.net>
Sent: Fri, 25 Nov 2011 03:07:03 -0500 (EST)
Subject: Re: [HBR] Pullen mixer

Paul
The tube I am using is a 6gm8 dual triode which was designed to run with 30
on the plate(s).  IIRC, the transconductance is around 7800 @ 30 volts.  My
radio is based on the Collins r392, so my B+ runs at 25-26 volts.  Same for
the heaters.
As this was somewhat uncharted ground (at least for me), I was not sure
what The 6gm8 plates should run at.  The 6es8 calls for 150 on the 'signal'
triode and 50 volts on the "LO" triode.  I tried to keep the same 3:1 ratio
on the 'gm8.  I ended up with a 50k pot-wiper to the second plate.  Top of
pot to 25volt B+.  Bottom of pot to ground.  The 'signal' plate B+ was fed
via a 1.8k resistor, 20 volts on that plate, and the plate is at RF ground,
through a .01 cap.
The grid of the 'signal' triode has a 120k resistor to ground.  The 'LO'
triode grid has a 120k resistor to the CATHODES-not to ground-.  The
cathodes are tied together and go to ground via a 300 0hm resistor.
I forgot to mention-the output is taken from an IF transformer.  The
primary is between the plate (LO triode) and the B+ supplied from the wiper
of the 50k pot.  Secondary to IF stage. There is a .01 cap to ground on the
B+ side of the IF transformer.

With an antenna connected to the grid of the 'signal' triode (via a .01
cap), and a signal from a signal generator set to 5455 kc--assuming your
using a 455kc IF-tune the pot until you hear WWV!
As I mentioned, I am using a tube designed for low voltage B+.  I see no
reason why the pot voltage 'tuning' would not work with other tubes.  I did
find that .5 volt made all the difference in the world, so the pot was left
in the circuit.  The adjustment is rather critical-at least with the
6gm8-but once you find the 'sweet spot' it is very stable, however..........
TRY IT!!  I have yet to feed it with a tuned rf stage--RF stages can hide a
bunch of mixer problems--so if it works well-and it does! w /o any
tuning....it should really shine coupled with a good front end.
--let me know how yours works, and if I can answer any other questions.
Hope the above makes some sense....am full of turkey!
73's
de wb0rai Don

On Fri, Nov 25, 2011 at 12:33 AM, Paul Dulaff <pdulaff at embarqmail.com>wrote:

> Don
>
> Could you post details on the mixer circuit such as the plate load
> resistor values, cathode resistor value, etc.
>
> Some literature states that the 6ES8 and the 6DJ8 can be substituted.
>
> Best 73's
>
> Paul - WB2NMI
>
> ----- Original Message -----
> From: donald haworth <donmhaworth at gmail.com>
> To: HBR Receiver List <hbr at mailman.qth.net>
> Sent: Wed, 23 Nov 2011 20:06:44 -0500 (EST)
> Subject: [HBR] Pullen mixer
>
> Some time ago I posted some questions about a dual triode mixer, as
> described by Walt.  A very simple circuit, indeed, with the plates tied
> together, a common cathode to ground via a 470 ohm resistor, and the signal
> and LO fed on the grids.
> -Long story, short--I built it and it worked, as  advertised.  Noise was
> about same as the pentagrid convertor, however, and gain was much
> lower-that was no surprise-but I did expect the noise to be a bit better..
> I must add that I am running only 26 volts on the plate.(26d6).  The twin
> triode I was using is a 6gm8.  I then tried it in a Pullen configuration.
> One plate with 20 volts and second plate (with LO to grid)  2.2 volts--this
> via a 50k pot so I could adjust the voltage.....WOW!!  What a difference!
> All kinds on gain, NO noise---sounds like the antenna has been disconnected
> -then you tune across a signal and it pops out.....!  Really impressive.
> And the best part----no spurs --no birdies---I am using a N3ZI DDS for the
> LO.  Injection is 250mv.  Antenna running into the other grid-no
> tuning-nothing-and it blows away every mixer circuit I have tried, to
> date.  I have found the best way to test a mixer circuit is to input a
> signal to the (signal) grid--no rf amp-no tuning...this will show up a bad
> / good design in a hurry.  A good RF front end will cover a multitude of
> mixers problems, so eliminating it will give you a good idea of how well
> the mixer preforms.....
> --I am now a believer.  All the things they say about the Pullen are
> true---and all it took was three resistors added to Walt's dual triode
> circuit!  I must say that the Plate voltage ratio is critical--it will make
> the difference between no signal and Grins and Giggles-but a pot in a
> voltage divider circuit to feed the second grid, and just tune for the best
> output......for the 6gm8, I have 20volts on plate 1 and only 2.2 on plate
> 2............
> ----Give it a try!
> ---and a Happy Thanksgiving to all
> Don (wb0rai) in Nashville
> ______________________________________________________________
> HBR mailing list
> Home: http://mailman.qth.net/mailman/listinfo/hbr
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:HBR at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>
> ______________________________________________________________
> HBR mailing list
> Home: http://mailman.qth.net/mailman/listinfo/hbr
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:HBR at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>
______________________________________________________________
HBR mailing list
Home: http://mailman.qth.net/mailman/listinfo/hbr
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:HBR at mailman.qth.net

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html