Warning: Creating default object from empty value in /homepages/11/d132647312/htdocs/Amfone/mkportal/include/SMF/smf_out.php on line 47
AM noob; looking for help with underperforming DIY transmitter




 
The AM Forum
May 23, 2019, 09:35:13 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: AM noob; looking for help with underperforming DIY transmitter  (Read 997 times)
0 Members and 1 Guest are viewing this topic.
K8DI
Contributing
Member
*
Offline Offline

Posts: 24


« on: December 06, 2018, 02:04:37 PM »

Background: ham for a long time, but only playing on HF for the last few months. DIY electronics for 40+ years, but mostly audio, both tube and solid state.

About a year ago I became intrigued by the Heil pine board article in QST.  I decided to give it a try. I did it a little differently; I used junkbox parts combined with a few eBay finds, and bought some stuff new. My PSU was much stiffer. I used a preamp I already had. But, it didn't oscillate. I got other parts, crystals, etc. and researched LOTS of 6AG7 circuits, then tried again. I got an oscillator that modulated poorly and had low output. Swapped to a pi output network, a 6L6 for modulation (twice the plate dissipation/plate current capability) and a bypassed resistor arrangement to put the 6AG7 plate at a slightly lower DC voltage than the modulating tube to allow 100% modulation.  Now, that transmitter sounds good on a monitor receiver, modulates fully, etc., but in reality, the 1.5 watts it makes are insufficient for making contacts on 40m with a low-mounted (20') dipole these days. No CQ was ever answered.  

So I've moved on to another transmitter build, this one to have an oscillator, maybe a buffer/driver (but I feel like it shouldn't be needed), and one of either 807, 809, 811, or even 814 for a final plate modulated amp. I've found some decent enough iron, the audio modulator amp is easy (I've got half a dozen tube-based 30-60 watt amps laying around here I can re-purpose), so here I am, stuck on the first stage.  I cannot get my circuits to oscillate half the time, and when they do, the output voltage swing is tiny, at best 20 Vpp, usually 1/4 that.  With 300+ on the plate, I expect much more.  They all amplify but even then the gain seems low.  Below is today's circuit with values for two tube types in a chart.  I would love to hear some commentary and/or suggestions on this. I have decent test gear; if you want to know waveforms/voltages/currents, I can measure things.

Goal of this circuit section is 35-75 Vpp output, to drive the grid of an 807 for starters.

Ed Walters, KB8TWH

* AM40-Osc.pdf (34.17 KB - downloaded 63 times.)
Logged

Ed, K8DI (fmr. KB8TWH)
PA0NVD
Member

Offline Offline

Posts: 529


Nico and Chappie (Chappie is the dog...)


« Reply #1 on: December 06, 2018, 02:39:15 PM »

Hi Ed
There are many designes of simple 2 tube transmitters, but they are in general crystal controlled. Here https://www.robkalmeijer.nl/techniek/electronica/radiotechniek/hambladen/qst/1946/07/page16/index.html
is one. If you leave the crystal out, you can drive the grid of the first tube using a VFO
When you like to make the first tube to oscillate freely, the frequency stability may be a big problem, making a stable VFO at 7 MHz can be quite a challenge.
I normally use the DDS vfo's that are sold at Ebay and alikes from China. Very low cost and they work nice. Than you can select the frequency you want and it is dead stable. The VFO is really the heart of every transmitter (provided it is not crystal controlled)
You need approx 50Vtt -70 Vtt to drive the grid of the first tube. So the VFO needs an amplifier. Attached is the diagram. The voltage is transformed using a LC tuned circuit, also needed to clean-up the output of the DDS unit. I did post the information of the VFO attached as My DDS vfo) in AMFone. If you like to use a DDS vfo from China, I advice that you read it
Attached is an example of such a vfo with a drive amplifier that is sufficient to drive the first tube of the example transmitter above.
You can simply anode modulate the transmitter in the anode voltage of ONLY the final amplifier using the iron you have. Put the scope at the RF output to see the modulation and prevent over-modulation.


* vfo diagram.JPG (370.96 KB, 2592x1944 - viewed 80 times.)

* vfo1.JPG (593.41 KB, 2592x1944 - viewed 47 times.)

* vfo2.JPG (763.57 KB, 2592x1944 - viewed 67 times.)
Logged
PA0NVD
Member

Offline Offline

Posts: 529


Nico and Chappie (Chappie is the dog...)


« Reply #2 on: December 06, 2018, 03:04:33 PM »

I forgot the tuning capacitor across the output of the VFO amplifiers coil. I use a 2 x 300 pF foil capacitor with the sections in parallel, than yo can tune the low band from 1.8 - 7,5 MHz. Just tune at max grid current.
Logged
KC2ZFA
Member

Offline Offline

Posts: 369



« Reply #3 on: December 06, 2018, 07:34:49 PM »

Ed, remove the lamp (?) thatís in series with the crystal and directly ground the crystal. Also, remove the cathode resistor and ground the rfc. Remove the cap going to the vfo jack. If the circuit doesnít oscillate turn that 33pf in the grid into a trimmer and adjust so oscillation starts reliably on turn on. Also, you may want to try connecting the grid leak in parallel to that 33pf cap or to the trimmer that will replace it.
Logged
K8DI
Contributing
Member
*
Offline Offline

Posts: 24


« Reply #4 on: December 07, 2018, 10:22:22 AM »

... remove the cathode resistor and ground the rfc....

With my RFC having only 2.1 ohms DCR, there'd be insufficient bias developed; the tube would be in saturation and burn up in a short time.  Please explain the other suggestions; a lamp in series with the crystal to monitor RF current seemed to be standard in the old days, and the VFO coupling cap isn't in circuit when I want the tube to oscillate...   I've tried a variable for the feedback capacitor, I'll try that again.

I've seen some circuits run the B+ through the tank, in there any advantage to that? I'd rather not have a bare coil and capacitor plates to shock myself with if there's no point...

Also, for what it's worth/satisfying curiosity, I've attached the schematic for the final modified Heil design I ended up with.

Ed


* Final Heil PB.pdf (31.21 KB - downloaded 38 times.)
Logged

Ed, K8DI (fmr. KB8TWH)
KC2ZFA
Member

Offline Offline

Posts: 369



« Reply #5 on: December 07, 2018, 12:39:25 PM »

the one immediate change you can do is change the 33pF to a 15pF and see what happens.

also see here https://www.qsl.net/k5dh/6ag7_807.html

Logged
K8DI
Contributing
Member
*
Offline Offline

Posts: 24


« Reply #6 on: December 07, 2018, 09:14:12 PM »

I forgot the tuning capacitor across the output of the VFO amplifiers coil. I use a 2 x 300 pF foil capacitor with the sections in parallel, than yo can tune the low band from 1.8 - 7,5 MHz. Just tune at max grid current.

Eventually I will have a VFO in a separate enclosure to drive this. I have a DDS card a friend brought me when he was working in Germany a few years ago from Conrad Electronic.

This output tuning capacitor, am I correct in assuming you mean across the primary of the output transformer?

Ed, KB8TWH
Logged

Ed, K8DI (fmr. KB8TWH)
PA0NVD
Member

Offline Offline

Posts: 529


Nico and Chappie (Chappie is the dog...)


« Reply #7 on: December 07, 2018, 10:10:15 PM »

No Ed, across the secundary, the output. The primary only has 8 turns connected to the amps output. The secundary has 40 turns and has to be tuned to get the high voltage needed to drive a tube.
I drive my Phillips SFZ395 transmitter with this VFO
Logged
K8DI
Contributing
Member
*
Offline Offline

Posts: 24


« Reply #8 on: December 17, 2018, 06:39:55 PM »

Thanks for the suggestions, guys.  An update:

From some suggestions locally that I may have some parasitic problem, which I was a little doubtful of, I started to investigate things I could find different from the classic circuits and my setup. 

The first obvious difference: RF chokes. Mine are modern parts, 2.7mH, 2.06 ohms DC, 530 milliamps max current. Likely designed for choking hash in a SMPS circuit.  Vintage RFCs were typically pi wound,  lower current, higher DCR parts.  Now, I know that it's not unheard of to use an RFC in a circuit above its self-resonance frequency; plate chokes in big amplifiers have to be tested to make sure their self resonance isn't in-band, etc., but ideally you'd want an RFC to be an inductor at the frequency of interest.  So, unable to find specs on self-resonance for these chokes, I found a way to measure it (I don't have a working grid dip or I'd have used it...) These modern parts' self-resonance is at 60kHz! They're nothing like an inductor at 40m frequencies.  So I've sourced some NOS pi type chokes from Fair Radio, and a couple used ones off eBay. I'm reworking the circuit this week, incorporating a trimmer for the crystal, and also changing the other parts that were "non-classic" -- mica caps instead of ceramic for anything expected to pass RF rather than bypass it to ground.  Here's hoping for lots more RF flying around the workbench!

Ed, KB8TWH


* modern.jpg (24.28 KB, 200x200 - viewed 28 times.)

* classic.jpeg (6.69 KB, 392x182 - viewed 29 times.)
Logged

Ed, K8DI (fmr. KB8TWH)
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 1995


Making AM GREAT Again!


« Reply #9 on: December 17, 2018, 10:30:38 PM »

Running the B+ through the tank allows the RFC inductance to be a whole lot smaller on the low impedance side of a tank circuit.

10 times Z is a lot lower when the Z is 50 ohms instead of 1k to 5k!

Thays the benefit I know of.  Others may chime in with different reasons.

--Shane
KD6VXI
Logged
K8DI
Contributing
Member
*
Offline Offline

Posts: 24


« Reply #10 on: December 24, 2018, 11:05:21 AM »

Update:

I built up another oscillator circuit, largely lifted from another old schematic. Simpler, not really an ECO/tri-tet in the conventional sense. It oscillates.  More importantly, putting in pi-type old-school and vintage RF chokes instead of the modern stuff I had made an enormous difference. The oscillator shown here generates over 400 vpp when the tank is tuned properly and the PA tube is out of the socket. The screen voltage is variable and can be turned down to keep from overdriving the 807's grid.

The 807 makes lots of RF from that, possibly as much as 60 watts, but I am a little fuzzy on that. I am hitting the 807 with a little past the tube book limits, about 668v on the plate at 105mA. Grid drive is 2.1mA. 

My setup is this: coax to LDG AT200Pro tuner, one of it's two outputs is my antenna, the other is the dummy load (A 3GHz rated 150w continuous high quality load that's been verified by a buddy who works in an EMC lab with their calibrated Rohde and Schwarz VNA). The tuner is in bypass mode. While it is showing roughly 50w on its meter, it is also showing SWR of 1.5:1.  I am determining power by measuring the output voltage on the scope (Agilent digital scope) and calculating based on 50 ohms. The SWR being not 1:1 makes me suspect the 60watts may not be accurate.

So, any comments on determining actual power output, without access to a Bird Thruline? Is my concern about the SWR affecting the accuracy reasonable?

And, now to modulate this thing. And possibly figure out a way to lose 10% of my B+.  Comments and suggestions are welcomed; thanks for the help thus far.

Ed, KB8TWH


* AM40-Osc-v6.pdf (32.88 KB - downloaded 23 times.)
* AM40-807-Amp-v2.pdf (27.98 KB - downloaded 24 times.)
Logged

Ed, K8DI (fmr. KB8TWH)
Pages: [1]   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 0.062 seconds with 19 queries.