Pullen Mixer

[N4LTA]

A buffer can't hurt and a 6J6 will fit into the same hole - It is eating up my heater current budget though. I ill probably install it instead of the 6C4.

I have built quite a few stable SS VFOs with the 6 material in a better temperature situation than a tube VFO -  No core should be better than a type 6 core - I have some of the zero material and might try it in this LO.

Thanks

Pat
N4LTA

[KM1H]

A bit OT but Ive been waiting to have a 6C4 problem in a Hammarlund and give the 6J6 a shot with sections in parallel.

Carl
KM1H

[N2DTS]

This is for AM right?
I just used the 6C4 LO right into the mixer, no pulling...

What would it be pulled by?
I can do stuff (tune the antenna) without any pulling.
Strong signals dont change the frequency, so I assume I have no pulling.

It sort of sounds like you guys are worrying about problems that wont be problems?

In a VFO where the load changes, I can see using buffers, but in a receiver the LO just looks into a grid of the mixer, which is fixed output (filter or IF trans).

Am I missing something here?

Brett

[N4LTA]

Right now my oscillator has a tapped coil with the tap feeding the cathode and the plate bypassed to ground.

If I read you right - you are saying that I can add a  100 K to ground on the second grid and a 47K or so from the second plate to B+ and capacitively couple off the second plate?

[N4LTA]

LO pulling is pretty common at higher frequencies - especially with high dynamic range mixers.

Dr. Pullen suggested that a buffer be used - it probably may not be necessary up to 20 meters but it is never a bad idea - especially with a free running oscillator. That was what my original question was about - I had decided not to use one for this receiver - then W3JN came in and messed things up  - and I fully agree with him - it can never hurt - I probably will do it if my heater budget allows - I have to come up with another .3 amps.

OK guys  - I need you to make me doubt my design again - this is AM only and I want to have good Hi Fi performance - What detector is best   - a true diode 6AL5 like HH Scott used or an infinite Z detector?

Pat
N4LTA

[N2DTS]

I tried a number of different types, the only one I could get to work well with a wide range of signal strengths and agc was the one I am using.

For a simple tube circuit, distortion is very low.

Maybe try few different designs...after all, you may end up with whatever you can get to work well.

My entire receiver is a bunch of circuits I could get to work, I tried different LO designs, different mixer designs, different IF amp designs, different antenna input designs, and what I wound up with was all the designs I could get to work well or at all!

Once I got the basic design down, I built another receiver with all the improvements and less extra holes....

The experimentation was the fun part, all the different LO designs and parts value changes, watching the signal on the spec-an for amplitude over the freq range, and stability.
I was lucky to have a real good spec-an that allowed me to watch the drift, frequency range, output level, etc.
I did all sorts of tests, even running the LO plate voltage down to something like 10 volts (it still worked!).

I think you might be disapointed if you think you will have a final design, then build it and have it work well.

Brett


 


Brett

[KM1H]

Most vintage commercial radios with free running LO's pulled on strong signals or when the RF gain control was run up to copy a weak signal and a strong station fired off nearby.  Some did it as low as 40M. Ive stuffed several 6C4's on brackets under the chassis to fix.

Yes, Ive mentioned modular construction on here and elsewhere before. Its something Ive been doing since the early 60's and it makes changes and trouble shooting a lot easier.

The best diode AM detector Ive run across so far is in the HRO-60. Jay, W1VD, had good things to say about its performance recently also in his review series. Im still up in the air what to use in the HB project. ER also had some mention of the subject but with the sometimes questionable articles they run Ive put any serious analysis aside for now. I havent reached that stage yet.

I dont know if any of the real high end consumer radios going back to the Scotts, McMurdo Silvers, Zenith 1000Z, and later ever really addressed the issue. More reading I guess.

Carl
KM1H

[WU2D]

Here is a HFO that I originally put into my Super Pro 400 using a 6SN7 but I modified the schematic for a 12AT7.

Mike WU2D

[w3jn]

LOL - sorry for the thread hijack - didn't mean to wreck the consensus 

Build the osc as you were going to using the first section o f the 6J6.  If you're using a design with the tank coil tapped to the cathode, this will provide enough coupling to the second stage.  You don't need to feed the grid of the second stage, you can ground it or just put a grid leak resistor to ground.  Then feed the second plate thru a 5K resistor or so (experiment here) and capacitively tap off that.  Or you can go with a wideband toroid transformer off the plate as well.  I wound a toroid with a split secondary to feed the deflector plates of a 7360 180 degrees out of phase.

[KM1H]

There is nothing new about the cathode follower oscillator either, CQ for July 1947 had a 6J6 version that claimed 50 cps drift on 80M in an hour after a short warmup. Im pretty sure they were used in the late 30's also. Cant get any simpler than a 6J6 for sure.

Carl
KM1H

[N4LTA]

I like the idea of the 6J6 with my existing oscillator. I'll try to get it built in the next few days.

The 12AT7 would do the same and be less of a strain of the heater supply though.

I am not interested in building a simple S38 or equal - did that a long time ago.

Pat
N4LTA

[w3jn]

The 12AT7 would give you the advantage of having separate cathodes, and thus you could plate-grid couple the buffer rather than cathode couple it.  I thought maybe you had a 7-pin hole punched already for that 6C4.

[N4LTA]

I do already have it punched - and the 6C4 working. That's why I will probably go with your solution.

I am going to try your method this afternoon - Your saying put a 47K or so on the second grid to ground.

Then put a 5 k or so plate resistor (second plate) and tap the RF off the second plate with a 100pf cap or similar.

Or maybe a T50-43 with 30 or so bifilar turns in the plate circuit - high z toward the mixer input?


What's your opinion on a good decent am detector?

Thanks,

pat
N4LTA

[w3jn]

Yeah, maybe try it breadboard style first on a scrap chassis if you have the 6C4up and working to your satisfaction already.  I'm overseas and can't look at the project to see how I did it, it was almost 10 years ago.

All of this is from memory so you may have to experiment.  You could try putting the primary of a transformer in series with the plate of the 2nd tube and you can wind the secondary for the appropriate voltage.

Best AM detector is the infinite impedance detector in the ARRL handbook.  You need a separate AVC detector for this though, preferably a whole nother 6BA6 IF stage to drive the AVC detector separately.  But this might blow your fil current budget.  If not, just copy the HRO50 or 60 detector, it does very well.

[N4LTA]

Sounds like I need to add a 6.3 volt filament transformer. The HRO 60 detector is very similar to the Scott Hi Fi am detector using a 6al5




Pat
N4LTA

[W1DAN]

Guys:

I have both an early and late HRO-60. The early '60 has lotsa detector distortion. The later one is pretty clean. Dunno the circuit differences.

I built an infinite impedance detector using an MPF102 for a SS rx. It was one of the cleanest detectors I have used.

Dan

[W1VD]

The HRO-60 did a respectable job of low distortion am detection...too bad the i-f bandwidth is so restrictive. Good for battle condx but not so good for hifi listening.

http://www.w1vd.com/NationalHRO-60.html

JN touched on an important point...separate AM and AVC detectors. This is one of the biggest problems with the BA receivers and causes increased distortion at the low audio frequencies. Some receivers are terrible in this regard...others are okay. This is one area where the SDR stuff really shines - essentially complete separation of recovered audio and agc.

I'd even consider giving each detector it's own last i-f amplifier toob...

[WA1GFZ]

My last HB I did a phase splitter as the last IF stage and used 1 phase for AGC and the other for detection.

[N4LTA]

How did you split the phase - with a wideband transforerm center tapped? I might give this a try - I have seen seperate AGC with a small cap to a diode but worry this will load the stage.

pat
STF

[K1ZJH]

How well are untuned buffer stages going to work say much above a few MHz?  I'd think phase splitters would even be more problematic?  JN, how did that untuned toroid work for you?

Pete

[N4LTA]

I have used bifillar toroid at 50 Mhz with a type 61 core and had no problems. Impedances were less than 1000 ohms though. A tuned splitter with type 3 material would work with a tapped secondary I guess. I have some T130-3 - 75 turns gives 200uh - would take about 600pf or so to resonate it at 455 Khz. An ARCO mica would work fine. An untuned one would be easier. Or find a 455 Khz transformer with a center tapped secondary.

Pat
STF

[WA1GFZ]

I used a 2N4416 AC couple into the gate with a 10 K to ground off the gate. 1 K 0n source to ground , Drain connect 1 K to 12 volts.  AC couple Source and a second cap on the Drain for the two phases. Very easy to do with a tube but might need higher resistor values.

[w3jn]

The untuned toroid worked OK.  As I recall about 10 turns primary and 20 turns each secondary, bifilar wound, each driving a 7360 deflector plate.  A real PITA to wind.  But this was for a single-band RX.

Check out the skizmatic for the Squires SS-1-R.  It uses a 1N34 for the audio and a 6AV6 in a triode amplified diode detector for the AVC.  Never seen one like this before.  http://bama.edebris.com/download/squires/ss1r/SS1Rsch5.gif

[N4LTA]

That's an interesting circuit. Do you have any idea how well that radio works on AM?

Looks like they are forward biasing the diode and grid at 1.7 volts or so at the anode?

Pat
N4LTA

[K4TLJ]

Gives me a headache to look at it....