The AM Forum
March 28, 2024, 12:39:23 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: 1 2 [3]   Go Down
  Print  
Author Topic: Pullen Mixer  (Read 55008 times)
0 Members and 1 Guest are viewing this topic.
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #50 on: September 29, 2009, 08:53:45 PM »

What are the series # of those HRO-60's?

Carl
KM1H
Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #51 on: October 03, 2009, 03:07:28 PM »

I put together a 6F6 oscillator on a small chassis to test. I have a conventional tapped coil inductor with the tap tied to the cathode and a 47K grid resistor to ground and a 100 pf NPO from the grid to the "top" side of the inductor. The low side is connected to ground. Th e plate is bypassed to ground and has 165 volts on it. I get a good 10 volts or better PK to PK at the grid.


The second triode has a 5.1 K resistor from +165 to the second plate with a .001uF coupling cap to the output. I have a 100K from grid to ground.

I get about a volt output non sine wave - almost looks like it is doubling or trying to.

Doesn't seem like I am coupling via the common cathode to the second stage.

Removed the 5.1 K resistor and replaced it with a wideband transformer wound on a FT37-43 core. Output still bad and low.

Changed the 100K grid resistor to a 47K - Waveform looked a little better. Put a 22pf cap from grid #1 to grid #2 - 20 volts Pk to Peak and a much better waveform but still not a clean sinewave - rounded at the top .
Maybe not enough inductnce in the coil?

Pat
N4LTA
Logged
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #52 on: October 04, 2009, 12:19:56 AM »

Yeah, there may not be enough inductance in the feedback tap of the inductor to get good coupling between the 2 sections.  THat's OK though.  Less feedback is better - use as little as you can for reliable osc operation.

Looking back I don't think I used a L with a FB tap - I used the Colpiss (capacitive divider in the grid)... I may have used a RF choke in the cathode circuit.  Crap, I can't remember now.  But it did work FB.

Where you're at now, use as little coupling between stages as you possibly can.   You're probably overdriving that second toob a bit.  Try a 5 pF or so, and try a 10K resistor to ground?   Dunno.  Wish I had my stuff here, I could breadboard something up and try and replicate what I did  Undecided
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #53 on: October 04, 2009, 09:43:57 AM »

I'll cut back on the drive and use a 10K - That is the direction that seems to look better. When I get this thing working - I need to publish it on the net - as I have searched high and low for a schematic. it is a nice tube and circuit and as easy as a single 6C4 tube but with much better isolation.

Thanks for the help.

Pat
N4LTA
Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #54 on: October 04, 2009, 04:32:51 PM »

Cut the coupling cap from 22pF to 4.7 pf and dropped the gid resistor to 15K.

Much better - almost good enough.

Nice looking sine wave at first look  -  25 V Pk to Pk



[But the tops and bottoms were a little fuzzy with a thick trace. So I put my signal generator on the scope at the same frequency and got a nice clean sine wave with no trace thickening on the corners?

Not good - it's almost like a little FM - in my earlier days I would have said looks good enough. Put in on the spec analyzer and found the 2nd only 12 db down and the 4 th down about 22 db. Everything else was 45 db or more down.

I read that a problem with a Hartly or Colpitts is that they are not clean - yet I see them used as LOs all the time.

Any comments? Is it good enough - Do I need to filter it like I would in a SS rig - I sure don't like the second only 12 db down.

(Maybe I'd be better off without all this test equipment - not really)

Pat
N4LTA

Logged
N2DTS
Contributing
Member
*
Offline Offline

Posts: 2307


« Reply #55 on: October 04, 2009, 10:18:34 PM »

I would experiment with values (operating points) some more.
After playing around with my LO, I was able to get a clean sine wave, with a really flat output level across a wide frequency range, a very low harmonic content, and stable operation with plate voltages as low as 20 volts, or maybe less, cant remember but it was stupid low!

Maybe isolate the sections, the osc, and the buffer and see where the fuzzy peaks are originating?

Dont settle for any trash on the LO....

Brett



 
Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #56 on: October 05, 2009, 11:02:21 AM »

I have a 47K grid resistor on the first stage of the oscillator. I may try to reduce it and see what happens.

Otherwise I may have to use a 12AT7 where I can separate the cathodes.

Pat
N4LTA
Logged
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #57 on: October 05, 2009, 12:52:03 PM »

Less coupling.  Put a 5-20 or so variable cap in there and adjust, see what happens.  You could also mess with the plate resistor value.   Mine was very clean, yours will be too.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #58 on: October 05, 2009, 01:30:30 PM »

Right now I don't have a plate resistor - I replaced it with a wideband transformer. A resistor would be easier to use.

 When looking at the inductor tap with a scope - It has a hitch in the waveform - it is not evident when looking at the grid of the oscillator. I am using a 47K grid resistor. Maybe I should decrease the value.
Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #59 on: October 07, 2009, 08:01:54 PM »

Well, I finally gave up on the 6J6 LO. I tried changing the coupling with a trimmer and changing the bias resistor on both grid but nothing would clear up the am fuzzyness. Maybe I had a bad tube - I had a new one though -

This afternoon I punched a hole for a 9 pin socket and installed a 12AT7 using the same LC parts. I got a nice clean sine wave on the first attempt. I'll wire the other side as a cathode follower and couple it to the mixer which is built and seems to be biased correctly with 155 volts and 51 volts on the two plates. I don't have enough drive voltage on the inductor tap so I can't couple the Cathode Follower there. Plenty available at the inductor "hot" end - so i'll hook up the CF input there via a small NPO cap.

I still have to build two IF stages but all the sockets and IF transformers are mounted and ready to wire.

I still wish I could get the 6J6 oscillator working and documented.

Pat
N4LTA
Logged
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #60 on: October 07, 2009, 09:43:51 PM »

I have a couple of old time 6J6 VFO/LO circuits that I'll scan and post here maybe tomorrow. Been kinda busy so havent replied earlier.

Carl
KM1H

Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #61 on: October 07, 2009, 09:50:19 PM »

I think maybe the tapped coil has caused the problems. Love to see the circuits.

Pat
N4LTA
Logged
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #62 on: October 08, 2009, 10:15:55 AM »

Here are some from 1947-48 CQ's.

I built a few of the PTO version to go with the CE 10A, 10B, 20A and Phasemaster Jr I used in the 60's to drive various VHF/UHF converters. Didnt take too much TC work to get good SSB stability. Having use of a machine shop at work took care of the mechanical stuff Grin

Carl
KM1H


* 6J6 VFO-1.jpg (160.18 KB, 600x1036 - viewed 1818 times.)

* 6J6 VFO-2.jpg (142.97 KB, 600x773 - viewed 1346 times.)

* Cathode Follower VFO-2 001.jpg (62.88 KB, 600x418 - viewed 1949 times.)
Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #63 on: October 08, 2009, 11:50:37 AM »

Thanks Carl,

I had searched high and low for something like this.

I think the hartley circuit does not do well with the common cathode. The untapped coil seems to be th eway to do this properly.

Pat
N4LTA
Logged
N2DTS
Contributing
Member
*
Offline Offline

Posts: 2307


« Reply #64 on: October 08, 2009, 01:20:01 PM »

I also had no luck with feedback coils, or tapped coils for feedback.
I know they are common circuits but I could not get any of them to work in casual experiments.

Brett
Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #65 on: October 08, 2009, 02:08:12 PM »

I have always used the tapped coil Hartley. I usually tap about 20% above ground even though I have seen some old references stating a 50 % tap is correct.

Never had many problems in getting them to oscillate and find them more stable than the Colpitts. The extra caps in the Colpitts add to things that can be thermally unstable.

The problem I was having was  using the 6F6 second triode as a buffer. The common cathode made it difficult.
For some reason, I was getting some small amount of AM or something that looked like AM - a slight thickening of the trace at the corners of the sine wave. Using the same tank components with a 12AT7  - and it is not there.

I switched to a 12AT7 an the output is very clean. The second was down 58 db measured on the coil tap.

I then go through a conventional cathode follower with the second triode. I haven't got the biasing on the second triode perfect yet.  Hope to get that working tonite and get the IF strip wired.

I am cramped for room in the preselector. The dual gang 365 pf cap is bigger than I left room for. Wish I had a couple of 100pf V caps that have a rear shaft so I could separate them and put a test RF amp in the middle.


Pat
N4LTA

Logged
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #66 on: October 08, 2009, 05:43:45 PM »

Quote
I am cramped for room in the preselector. The dual gang 365 pf cap is bigger than I left room for. Wish I had a couple of 100pf V caps that have a rear shaft so I could separate them and put a test RF amp in the middle.


Get creative Shocked

Cut the back off a 2 x 365, strip the back plates and use a coupler and shaft to reach the other cap. Pull plates on the good sections to reach 100pf.

Carl
KM1H

Logged
K1ZJH
Member

Offline Offline

Posts: 299


« Reply #67 on: October 08, 2009, 06:26:32 PM »

I For some reason, I was getting some small amount of AM or something that looked like AM - a slight thickening of the trace at the corners of the sine wave. Using the same tank components with a 12AT7  - and it is not there.

I switched to a 12AT7 an the output is very clean. The second was down 58 db measured on the coil tap.


Pat
N4LTA




Pat, did you check for VHF parasitic oscillations? Sounds
like the osc. may be breaking into parasitics over part of
the operating cycle.

Pete k1zjh
Logged
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #68 on: October 09, 2009, 09:06:19 AM »

I looked on my Spectrum Analyzer out to about the 10 harmonic.  That was about 70 Mhz - I didn't look any further and I should have. I got a 2nd only 12 db down and a 4 th down only 22 db or so. Everything else out to the 10th or so was way down.

That tube would be prone to a parasitic as it is a VHF tube. I still have a test breadboard so I'll check it out.


Thanks ,

Pat
N4LTA
Logged
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #69 on: October 09, 2009, 10:56:59 AM »

The 6J6 made a nice modulated oscillator transmitter at 420 mc back when it was still legal. However I suspect you have AC hum modulation and not a parasitic. Maybe some heater-cathode leakage.

Here is an interesting tapped coil plus cathode follower LO. Its from the Drake 1A. Its the best I can do with the crappy BAMA original PDF, printing to the HP All-In-One, then scanning back to Photoshop. The values are discernible if you squint a lot!

Carl
KM1H


* Drake 1A.jpg (75.5 KB, 660x600 - viewed 1350 times.)
Logged
WU2D
Contributing
Member
*
Offline Offline

Posts: 1800


CW is just a narrower version of AM


« Reply #70 on: October 09, 2009, 07:13:03 PM »

The good old 12AT7 will be hard to beat as an HFO.

The 6J6 is a tough one to tame. It is certainly used even at audio frequencies, but it is really VHF tube. It is hard to tame as a regenerative receiver as well as you can read in the lengthy thread on the regen Yahoo group. One guy started with a 1G6 regen with perfect results and utterly lost it with the same exact circuit using a 6J6.

I assume that you have a regulated plate voltage of 100 -150 volts or so for the buffer.
 
For the follower try 680 Ohm on the cathode, a .01 uf on the plate with a plate feed resistor of say 1K. A 220K will do on the grid. Feed in with a very low value cap like 3.3 pF and out with .001 uF or so.

Mike Wu2D
Logged

These are the good old days of AM
N4LTA
Contributing
Member
*
Offline Offline

Posts: 1075


« Reply #71 on: October 09, 2009, 08:58:44 PM »

Yes,

I am running th e buffer and oscillator on 108 volts regulated with a 0B2.

I decided to clean up the shop this afternoon - that is a major undertaking in my shop and just finished getting it about half cleaned up. I am feeding with a 68pf so I was probably overdriving it. My feed resistor is a 22K

I'll try your values as a start tomarrow. When I get it cleaned up - I'll wire the IF and see what the front end looks like.

I actually found a NOS set of two 140pF Hammarlund MC140S variables (don't ask what I paid for them) for the preselector and should have them mid next week.

Thanks

Pat
N4LTA
Logged
Pages: 1 2 [3]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.046 seconds with 18 queries.