The AM Forum
March 28, 2024, 08:05:42 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]   Go Down
  Print  
Author Topic: class AB2 for AM tube amplifier?  (Read 14662 times)
0 Members and 1 Guest are viewing this topic.
pe1pal
Guest
« on: July 18, 2009, 05:00:01 AM »

Hello, how has good experience with class AB2 and modulation in the driver.
I use 2x 813 in AB2 but the final modulation is still not 100%  Embarrassed
I use these schematic;


* 813_Amplifier_1_Tube_Drawing_1.jpg (164.6 KB, 576x474 - viewed 2511 times.)
Logged
flintstone mop
Contributing
Member
*
Offline Offline

Posts: 5055


« Reply #1 on: July 18, 2009, 06:27:24 PM »

WOW I hope the experts here can figure out your circuit.
Is that an RF final? Is it part of a linear? I see a pi-network for an RF stage, but can't juggle my brain any more than that.
What are all of those diodes hanging on the screen grid?

Fred
Logged

Fred KC4MOP
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #2 on: July 18, 2009, 06:55:30 PM »

Is this a final stage or driver stage. Need more info, dude.

What is the modulating voltge?

What is grid bias voltage, screen voltage voltage. Where does and what kind of supply does Igs and Isg go to?

It looks like some kind of clipping effect occurs when input RF goes to goes to -, + 6 volts.
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
W2PFY
Contributing
Member
*
Offline Offline

Posts: 13312



« Reply #3 on: July 18, 2009, 07:26:29 PM »

it looks like those diodes rectifies part of the driving power to bias the screen grid positive to me. Quite unique if I am correct. 
Logged

The secrecy of my job prevents me from knowing what I am doing.
n2bc
Contributing
Member
*
Offline Offline

Posts: 290


« Reply #4 on: July 18, 2009, 07:49:02 PM »

G2DAF design from the 50s or 60s, I think.  Rectified the RF drive for screen voltage. Google "G2DAF" you'll get lots of hits.

73, Bill   N2BC
Logged
WBear2GCR
Member

Offline Offline

Posts: 4135


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


WWW
« Reply #5 on: July 18, 2009, 08:05:21 PM »


You can look at it in a simple way:

100% modulation = 2x RF B+ (that being the peak-peak AF voltage

That's because 100% modulation has to be able to drive the voltage on the plate down to zero and up to twice the B+. So, you have to have the B+ on the plates of the 813 and the ratio of the mod iron's primary to secondary has to then permit the required AF peak-peak voltage to be developed.

The ratio on the mod iron may or may not permit you to make 100% modulation for a given B+ on the 813s...

Let us know what you are running in the modulator??

             _-_-bear
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #6 on: July 18, 2009, 08:15:35 PM »

I havent played with the G2DAF in decades but what I see is the input RF going thru a voltage doubler to develop screen voltage and using grid leak bias.

Since the G2DAF is a linear there is no modulator for the 813. What goes in comes out plus distortion/high IMD which the G2DAF is infamous for. Keep it in Europe.

Carl
KM1H
Logged
w1vtp
Member

Offline Offline

Posts: 2638



« Reply #7 on: July 18, 2009, 08:30:37 PM »

I havent played with the G2DAF in decades but what I see is the input RF going thru a voltage doubler to develop screen voltage and using grid leak bias.

Since the G2DAF is a linear there is no modulator for the 813. What goes in comes out plus distortion/high IMD which the G2DAF is infamous for. Keep it in Europe.

Carl
KM1H

I have to agree with Carl.  It looks like an attempt to have a linear amplifier.  The plots that I looked at didn't reveal much.

( http://home.swipnet.se/~w-30548/g2daf/ny_spect.htm )

 Certainly not the way I would check for linear operation.  If you're going to run an 813 as a linear amplification you're not going to get much of a return for your investment in time building it.  Better put the SG and CG to ground and drive the cathode.  Better yet, get a toob with at least 500 watts dissipation. 

As far as not achieving 100% modulation, are you getting there with the rig that is driving it?  If you are and are not with the amp, the answer is simple -- you don't have a linear. 

Properly adjusted would 2 X 813's give you 150 watts carrier? But with that circuit linear operation is going to hard to check or maintain.

My $0.02

Al
Logged
WBear2GCR
Member

Offline Offline

Posts: 4135


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


WWW
« Reply #8 on: July 18, 2009, 09:52:12 PM »


Gee, I didn't even consider that it is a linear, I just looked at the text...

Well, we can infer that he is using it as a linear, CW input from a ricebox perhaps, and then he replaces the RFC with some mod iron and modulates the linear??

Then what?

                           _-_-bear
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #9 on: July 18, 2009, 10:26:54 PM »

In a GG amp you can run Class C with about 15-20% of the audio coming from the driver and the rest from heavy iron. But its tricky to adjust and I didnt have good luck trying to make it play 160-10 with a 4-400 driving with an AF-67 or the Viking I. Its real drive tuning sensitive but would probably be fine in a monobander or duobander. Maybe a true high gain GG tube such as a 3-500Z would be better. The 4-400 is even a bit doggy as an SSB GG linear.

Carl
KM1H
Logged
pe1pal
Guest
« Reply #10 on: July 19, 2009, 05:42:13 PM »


As far as not achieving 100% modulation, are you getting there with the rig that is driving it?  If you are and are not with the amp, the answer is simple -- you don't have a linear. 

Properly adjusted would 2 X 813's give you 150 watts carrier? But with that circuit linear operation is going to hard to check or maintain.

My $0.02

Al

Hello Al and all the others, thanks for reply. I use the TS570 for driving these amplifier. when I give 15W drive, the amplifier gives around 150W carrier at 2200V. With 25W I have around 400W PEP. The only thing is some distortion at the modulation in AM. The drive modulation has no problem at all. Possible someone has some experience with class AB2 amp's. For more information about these amp; VK4YE has published almost the same design on internet. 73 Paul PE1PAL
Logged
w1vtp
Member

Offline Offline

Posts: 2638



« Reply #11 on: July 19, 2009, 08:17:08 PM »


As far as not achieving 100% modulation, are you getting there with the rig that is driving it?  If you are and are not with the amp, the answer is simple -- you don't have a linear. 

Properly adjusted would 2 X 813's give you 150 watts carrier? But with that circuit linear operation is going to hard to check or maintain.

My $0.02

Al

Hello Al and all the others, thanks for reply. I use the TS570 for driving these amplifier. when I give 15W drive, the amplifier gives around 150W carrier at 2200V. With 25W I have around 400W PEP. The only thing is some distortion at the modulation in AM. The drive modulation has no problem at all. Possible someone has some experience with class AB2 amp's. For more information about these amp; VK4YE has published almost the same design on internet. 73 Paul PE1PAL

Paul

I think I understand the circuit.  It appears to me that you are having some linearity problems near the peak of the envelope power cycle.  A couple of things come to mind here.
  • Have you been careful to load the amplifier to FULL expected peak power?  Assuming a 150 watt carrier (just what are you getting for carrier power output?)  you should be loading for 4X the power in CW -- so 150 watts = 600 watts of peak power.  Add to that the unsymmetrical characteristics of the human voice, you should probably add approx.. 20% -- so 600 watts + 20% = 720 watts -- just a fudge factor at this point.   If you cannot do this because of power supply duty cycle problems consider pulsing the xmtr / amplifier with a keyer and adjust for maximum power.  You can calculate peak power by reading the average power and applying the duty cycle as a multiplier factor.  Example, with a keyer sending a series of dits, figure a 33% duty cycle so 720 watts should read about 235 watts (if I have my math right).  I admit that this may be a problem due to your grid leak approach so that option may be out of the question. If keying or loading with full CW is out of the question then consider a modified approach by speaking into the mic and adjusting your amplifier for best possible positive peak deflection on a scope
  • The other thing that comes to mind is that you may be having difficulty with the load presented to the driver rig since some of the power is used to create your screen voltage.  I don't know what modulation peaks would look like with a changing load from the viewpoint of your driving rig.
  • Finally, the grid leak bias approach is, in my opinion, at best a very difficult bias system for linear operation and will take careful operation to make it work.  At worse, it will be near impossible to maintain linearity operation.

Is this just a "grand experiment" or are you looking for the best "bang for the buck" (or Euro)?  Again, you are going to be fighting to get best linearity out of this circuit and will need to be able to adjust your system for maximum expected PEP which I'm estimating to be at ~ 720 watts, given a 150 watt carrier.

Regards and GL, Al VTP
Logged
WBear2GCR
Member

Offline Offline

Posts: 4135


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


WWW
« Reply #12 on: July 19, 2009, 09:54:28 PM »


(this is Al's post with the formatting cleaned up a little... dunno if it is worth it but I did it anyhow- bear)


As far as not achieving 100% modulation, are you getting there with the rig that is driving it?  If you are and are not with the amp, the answer is simple -- you don't have a linear. 

Properly adjusted would 2 X 813's give you 150 watts carrier? But with that circuit linear operation is going to hard to check or maintain.

My $0.02

Al

Hello Al and all the others, thanks for reply. I use the TS570 for driving these amplifier. when I give 15W drive, the amplifier gives around 150W carrier at 2200V. With 25W I have around 400W PEP. The only thing is some distortion at the modulation in AM. The drive modulation has no problem at all. Possible someone has some experience with class AB2 amp's. For more information about these amp; VK4YE has published almost the same design on internet. 73 Paul PE1PAL

Paul

I think I understand the circuit.  It appears to me that you are having some linearity problems near the peak of the envelope power cycle.  A couple of things come to mind here.
  • Have you been careful to load the amplifier to FULL expected peak power?  Assuming a 150 watt carrier (just what are you getting for carrier power output?)  you should be loading for 4X the power in CW -- so 150 watts = 600 watts of peak power.  Add to that the unsymmetrical characteristics of the human voice, you should probably add approx.. 20% -- so 600 watts + 20% = 720 watts -- just a fudge factor at this point.   If you cannot do this because of power supply duty cycle problems consider pulsing the xmtr / amplifier with a keyer and adjust for maximum power.  You can calculate peak power by reading the average power and applying the duty cycle as a multiplier factor.  Example, with a keyer sending a series of dits, figure a 33% duty cycle so 720 watts should read about 235 watts (if I have my math right).  I admit that this may be a problem due to your grid leak approach so that option may be out of the question. If keying or loading with full CW is out of the question then consider a modified approach by speaking into the mic and adjusting your amplifier for best possible positive peak deflection on a scope
  • The other thing that comes to mind is that you may be having difficulty with the load presented to the driver rig since some of the power is used to create your screen voltage.  I don't know what modulation peaks would look like with a changing load from the viewpoint of your driving rig.
  • Finally, the grid leak bias approach is, in my opinion, at best a very difficult bias system for linear operation and will take careful operation to make it work.  At worse, it will be near impossible to maintain linearity operation.

Is this just a "grand experiment" or are you looking for the best "bang for the buck" (or Euro)?  Again, you are going to be fighting to get best linearity out of this circuit and will need to be able to adjust your system for maximum expected PEP which I'm estimating to be at ~ 720 watts, given a 150 watt carrier.

Regards and GL, Al VTP
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #13 on: July 20, 2009, 02:56:04 PM »

OK, after reading the theory from the .pdf file I am starting to understand this creature. That long string of diodes provide screen bias from the input (driving) voltage.

Good info!


The only thing I can suggest is the following:

In addition to checking your AM linearity from the modulating source (your TS) into a dummy load and into the AMP input, I would make the control grid bias and screen grid bias resistors variable (for example, the grid bias resistor might be 600 ohms to 2.5 k ohms; the screen grid resistor might be 10k to 40k),
or changeable (switchable) since it appears this amplifier is one tricky animal toward obtaining linearity vs loading. After finding good values, you might want to construct something similar to a band switch for the grid and screen bias resistors.


Also, I would add a bypass cap to the lower end of the 2.5 mH grid choke.

Phil - AC0OB
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
pe1pal
Guest
« Reply #14 on: July 20, 2009, 05:07:53 PM »

Hello thanks for the information, I will testing the amplifier again. Possible there are good results even in AM modulation Shocked
Otherwise I have to find another trick for modulation on the screen or something Huh
73' Paul
Logged
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #15 on: July 20, 2009, 05:48:49 PM »

Quote
OK, after reading the theory from the .pdf file I am starting to understand this creature. That long string of diodes provide screen bias from the input (driving) voltage.

Good info!

That is obvious from the schematic shown in the first post. It is a classic voltage doubler driven by RF voltage stepped up by the input pi network. IMO a 1N4148 is a poor choice for linearity. Schottkys would be better.

A grid choke bypass may or may not be a good idea with that big a choke. Id at least swamp it with around a 3300-4700 Ohm carbon to minimize the chance of LF parasitics. Id certainly snoop that choke with a GDO in diode mode and see if anything funny is going on.

The 813 socket should also be sunk below deck to shield the grid as well as use a socket with a grounded twist lock base. With a careless layout a 813 can take off nicely even on 80M.

Carl
KM1H





Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #16 on: July 20, 2009, 09:17:38 PM »

controlled carrier distortion generator.
Logged
w1vtp
Member

Offline Offline

Posts: 2638



« Reply #17 on: July 20, 2009, 10:31:00 PM »

Hello thanks for the information, I will testing the amplifier again. Possible there are good results even in AM modulation Shocked
Otherwise I have to find another trick for modulation on the screen or something Huh
73' Paul


OK Paul on all.  Well, if you decide that you've had enough with this circuit, consider switching over to SG modulation.  In this regard, take a look at a previous amfone posting:

http://amfone.net/Amforum/index.php?action=printpage%3Btopic=19018.0

Good info there.  Basically, I think you could realize at least the power you want and at better performance.  Properly adjusted, SG modulation can be an excellent sounding AM modulation system.  You can keep your grid leak approach to the CG.  You will need to have a SG power supply but it will only have to be approx.. 1/2 plate modulated SG requirements (about 200 Volts)..  It would be nice to be able to load up your 813 to full power with full SG volts and then back off the SG volts as discussed in the above thread.

I'm assuming that plate modulation is not a practical option for you.  However, SG AF requirements are far less demanding and you could get by with about 25 - 30 watts of AF power with the right combination but that's another subject of discussion for another post.  This has been a fun thread.

GL es 73, Al
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 0.081 seconds with 19 queries.