The AM Forum
March 29, 2024, 04:25:49 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: 811 project  (Read 8833 times)
0 Members and 1 Guest are viewing this topic.
K9ACT
Guest
« on: August 25, 2007, 08:39:17 PM »

I am working on a new project, an 811 based AM rig for 40M.  I am at the breadboard stage and it is basically working but with lots of problems.  The exciter is the one discussed as my last project for driving the 8000 rig.  It is also intended to be a 40 m CW rig which is the reason I chose to include the fixed bias supply.  As it sits, it puts out about 100W.

There is a schematic of it at http://schmidling.com/811.emf

1.  The most serious problem is that if I look at the grid tank with the scope, I see a nice clean sine wave and sharp peak at resonance until I either light up the 811 or if it already is, when I connect the bias.  At this event, the sine wave turns into mush and there is no longer a sharp peak at res.  In fact there is no peak at all but just increases in the direction of min cap.

2.  In spite of this, it seems to neutralize properly and put out about 100w on the proper freq.

3.  I have read that one should start by removing the tube and look at the output with a scope and measure coupling before worrying about neutralizing.

When I do that I get as much RF at the output as I get at the input.  Hunting for the problem, points to the neutralizing cap.  It is a direct couple between input and output so how does one learn anything this way?

4.  My 100 watts out means into a light bulb.  In fact, I can get about 150 watts into a light bulb but when I switch to the dummy load position on the MFJ, I only get about 50 watts.  I have a Bird in series so I believe the numbers.  Why this difference between a dummy and a light bulb?

Thanks,

js
Logged
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #1 on: August 25, 2007, 10:06:48 PM »

Do you have rf chokes on the grid coupling?   Is the power supply lead(s) for the grid driver tube(s) bypassed for rf?
Is the grid circuit isolated Physically or shielded from the output circuit?
Logged
Pete, WA2CWA
Moderator
Member

Offline Offline

Posts: 8154


CQ CQ CONTEST


WWW
« Reply #2 on: August 26, 2007, 12:04:46 AM »

4.  My 100 watts out means into a light bulb.  In fact, I can get about 150 watts into a light bulb but when I switch to the dummy load position on the MFJ, I only get about 50 watts.  I have a Bird in series so I believe the numbers.  Why this difference between a dummy and a light bulb?

Thanks,

js

Pumping RF through a Bird meter and then into a light bulb will generally give you bogus results. If you heat the nichrome wire as the filament of a light bulb, the temperature changes, and therefore, the resistance changes (so your resistive load changes). I would also suspect there is reflected power on the line causing the "bird watts" to read even higher.

Pumping RF through a Bird meter into a non-inductive 50 ohm dummy load should generally give you an accurate indication of power output, given little to no reflective power. Light bulb loads are good visual indications in a pinch but are not very accurate loads.
Logged

Pete, WA2CWA - "A Cluttered Desk is a Sign of Genius"
K9ACT
Guest
« Reply #3 on: August 26, 2007, 01:06:22 AM »



Pumping RF through a Bird meter and then into a light bulb will generally give you bogus results. If you heat the nichrome wire as the filament of a light bulb, the temperature changes, and therefore, the resistance changes (so your resistive load changes).

I guess I do not understand that.  Once the light bulb reaches temp, the resistance is more or less constant.

> I would also suspect there is reflected power on the line causing the "bird watts" to read even higher.

That has been puzzling me.  I have seen the 100 W bulb get so bright it looked like it would burn out and  the needle  hit the peg 250w element.  Where on earth is all this power coming from? It's only a 1000v dc supply.

There is also an MFJ tuner and meter in line and I get the same numbers on it.

js


Logged
WBear2GCR
Member

Offline Offline

Posts: 4135


Brrrr- it's cold in the shack! Fire up the BIG RIG


WWW
« Reply #4 on: August 26, 2007, 10:27:12 AM »

Jack,

The link: "shesanowork"

Where is the power coming from?
Ohms law?

P = I x E 
250 = 0.250 x 1000 ??

       _-_-bear
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
K9ACT
Guest
« Reply #5 on: August 26, 2007, 02:36:12 PM »

Jack,

The link: "shesanowork"

Where is the power coming from?
Ohms law?

P = I x E 
250 = 0.250 x 1000 ??

       _-_-bear

I caught the link error and did an edit but guess it was ignored.  The schematic is at
http://schmidling.com/811rf.emf

Problem with the latter is, it's only drawing 150 ma and certainly not 100% effeciency.

js


Logged
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #6 on: August 26, 2007, 06:52:17 PM »

THe impedance of a light bulb is not gonna be 50 ohms, so any reading from the Bird at any other impedance will be incorrect.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #7 on: August 26, 2007, 07:55:10 PM »

  "  THe impedance of a light bulb is not gonna be 50 ohms, so any reading from the Bird at any other impedance   "



But they will be BIRD watts, the best kind.... ..         klc
Logged

What? Me worry?
kf6pqt
Member

Offline Offline

Posts: 530


« Reply #8 on: August 27, 2007, 12:42:57 AM »

What kind of file is an emf file?
Logged

W6IEE, formerly KF6PQT
W3GMS
Contributing
Member
*
Offline Offline

Posts: 3063



« Reply #9 on: August 27, 2007, 11:08:01 AM »

Jack,
Good deal on your 811'A rig.  You may want to simply the input network by ungrounding the center tap of the secondary input link and put in an .001 bypass cap to gnd at that point.  At the same point put a 2.5mh rf choke and insert the bias voltage on the low side of the choke with another .001 cap to ground.  Then gnd the center junction point of the input dual section variable.  Then you can remove the series input cap going to the grid of the 811A.  Not sure if that will solve your problem, but its a bit cleaner.  Concerning the wattmeter dialog, yep it all has been said.
Regards,
joe W3GMS     
Logged

Simplicity is the Elegance of Design---W3GMS
K9ACT
Guest
« Reply #10 on: August 28, 2007, 12:06:50 AM »

What kind of file is an emf file?

It's an image file like jpg but if compressed, much less.  I know not what emf stands for but you can open it in the windows photo/fax viewer like any other image.

js
Logged
K9ACT
Guest
« Reply #11 on: August 28, 2007, 12:24:37 AM »

....You may want to simply the input network by ungrounding the center tap of the secondary input link and put in an .001 bypass cap to gnd at that point.  At the same point put a 2.5mh rf choke and insert the bias voltage on the low side of the choke with another .001 cap to ground.  Then gnd the center junction point of the input dual section variable.  Then you can remove the series input cap going to the grid of the 811A.  Not sure if that will solve your problem, but its a bit cleaner. 

I willl give it a try.  I drew it out to make sure I understand what you said.  What I like best about it is that I do not have to isolate the cap which is a big pain.

I just rewound the coil for less L and put a dual 60pf cap in it but it still isn't right and I don't have the right cap to get the Q to 12.

I wish there was some way to do this with a single gang cap as I have lots of them.

The pi output is the show stopper.

What I have so far is the way I did the 8000 grid and it works well so it seemed a good place to start.

BTW, this is literally on a board with Masonite panel so some of these problems may go away when I go to aluminum but I want to get as close as possible to the final configuration before I start that.

Thanks,

js
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #12 on: August 28, 2007, 04:45:01 PM »

2 zig zags make the 811 more stable
Logged
WU2D
Contributing
Member
*
Offline Offline

Posts: 1800


CW is just a narrower version of AM


« Reply #13 on: August 28, 2007, 10:20:02 PM »

Hi Jack,

That is a nice amplifier design. I think it will act differently when it is on aluminum and you have some shielding between the grid and plate circuits.

I like to check neutralization backwards because it is easier to do most of the time - this is a bridge and it works both ways. A simple check on the neutralization would be to remove drive and attach a 50 Ohm dummy load to the INPUT. Remove B+. You can Bias and Fill on.

Connect a generator with a 50 Ohm pad (if it is not 50 Ohms or you do not trust it) to the OUTPUT connector and set it for something like 1 Vp-p.

Attach a scope with a 10X probe to the top of the input grid tank to ground. Set the scope so you can see something. You can use almost any high Z RF sensing device like an RF voltmeter too.

Peak the grid and the plate circuits (they should be close already if the amp is tuned up). This is not that critical since we are trying to null the tube's capacitance and strays.

Now rock the neutralization cap for minimum on the scope. Are you getting a good null?

Mike WU2D
Logged

These are the good old days of AM
KU6X
Guest
« Reply #14 on: August 28, 2007, 11:26:38 PM »

Most PP tank circuits I've seen use a dual gang capacitor with the rotor  grounded. Typically you *don't* want the coil CT to be at RF ground, it's difficult to get a true CT  with the non linear loading of a single grid. Don't bypass the coil CT to ground, let it float.  You don't need a grid RFC with the 1 K grid leak bias resistor.  Of course watch out for  parasitics.  The VHF type are easily detected by a holding a neon lamp (on a stick) close to the plate. A purple glow (vs orange) is a positive indication of VHF energy.   A high Q (High C) tank may help to  minimize grid waveform distortion and make neutralization more complete.
If your efficiency remains low you might increase the grid leak value  to run at a smaller conduction angle.
Sounds like a fun project!
73, John, KU6X
Logged
K9ACT
Guest
« Reply #15 on: August 29, 2007, 01:12:35 AM »

Don't bypass the coil CT to ground, let it float.

I reworked it as Joe suggested and it seems to work although it did not seem to want to neutralize at first but that problem went away.

I had it on the air today and made a few CW contacts.  Great fun.

>You don't need a grid RFC with the 1 K grid leak bias resistor.

I will try floating it but I presume that you mean to leave the bias hooked to the CT but leave off the cap and RFC?

>Of course watch out for  parasitics.  The VHF type are easily detected by a holding a neon lamp (on a stick) close to the plate. A purple glow (vs orange) is a positive indication of VHF energy.

That's a pretty cute trick.  Can't wait to try it.

Thanks,

js

Logged
K9ACT
Guest
« Reply #16 on: August 30, 2007, 10:23:12 AM »

I added the 811A project to my Radio page.  There is a picture of the breadboard and schematics at the bottom of the page.

It actually works.

http://schmidling.com/radio.htm

js
Logged
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 0.045 seconds with 18 queries.