Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: Prev 1 2 3 [4] 5 Next   Go Down
  Print  
Author Topic: Copper Strip For Grid Grounding in TL922A  (Read 21789 times)
SM0AOM
Member

Posts: 238




Ignore
« Reply #45 on: August 07, 2015, 01:21:02 AM »

If lifting the grid of a gg triode has so many advantages in getting low IMD, how come those building BIG linear amplifiers - 30kw and up - never used it? Companies like Marconi, Standard Telephones and Racal?

GG final stages in big amplifiers have, as I recall the story in a German engineering journal, been largely avoided due to the general problem of making convergent servo-mechanisms for automatic tuning.
As the input and output circuits essentially are in series, their adjustments interact in a way that makes servo-tuning extremely difficult. It appears that the conventional grounded-cathode circuit with neutralization and RF negative feedback were far more docile.

As far as I know, only one or two servo-tuning large GG commercial amplifiers have been produced, a 20 kW from Rohde&Schwarz in the early 60's, and another from Siemens.

The R&S I have seen "in the flesh" at Grimeton and it was soon relegated to fixed frequency
service due to the general "jitteriness" of its servo control system.
It was described as a reliability PITA by the operational staff, as compared to the more conventional Telefunken 30 kW transmitters that came a few years later.

73/
Karl-Arne
Logged
G3RZP
Member

Posts: 1149




Ignore
« Reply #46 on: August 07, 2015, 01:54:16 AM »

Both Marconi and Standard Telephones and Cables used a servo tuned gg stage in their 30kW tx (Marconi) and 80kW tx (STC), preceded by a servo tuned driver using a grounded cathode tetrode. There are papers on both in the proceedings of the IEE Convention on HF Communication in 1963. The Marconi tx was the H1100: it was quite successful, as was the STC one, which I saw at Ongar Radio back in the late 1960s.
Logged
K9AXN
Member

Posts: 441


WWW

Ignore
« Reply #47 on: August 12, 2015, 09:10:04 AM »

Thoughts regarding the Collins originators and Bill Orr implementations of Super Cathode Drive in grounded grid amplifiers.

We haven't heard any comments from any of the numerous engineers that approved that design including the Collins personnel who notably were the genesis; nor any of the numerous MFRS who included it in their amplifiers.  If you search on super cathode drive you get 6 Jillion hits from a hand full people, none positive, with as many theories as to why it fails in triodes as why your toe nail gets caught on your ragged PJ’s.

I disagree with the assertions previously written, excluding the anode to control grid inter-electrode capacity in the triode from the feedback bridge.  It is the most significant part of the negative feedback scheme.  It has significantly more negative feedback than the cathode to grid capacity.  

Plate to control grid feedback is blocked in the 30S1 tetrode with the screen grounded but replaced by the 5pf capacitor added between the plate and CG.  When the 5pf capacitor is added it renders the shielding effect of the grounded screen meaningless.  They added what appears to be a parasitic suppressor in series with the 5pf capacitor.  This appears to be used to mute parasitic content from reaching the control grid through the added 5pf capacitor.  

The grid in the triode configuration doesn't have this protection.  The triode configuration may not be capable of reaching as stable a state as the tetrode even if it is was run in AB1 mode.  There is no way to decouple the oddities of the plate to the grid in a triode, however the values selected for the grid to ground capacitors seems to be appropriate.

Each of the three 200pf capacitors with ½ inch leads are series resonant approximately 70Mc.  With 2 inches of lead they are series resonant at approximately 56Mc --- close to a direct ground – skin effect. 

With 2 inches of lead the capacitive reactance would decrease up to 56Mc where there is no reactance, above 56Mc it would become increasingly inductive but not exceed the inductance of the length of the wire or materials used to connect the capacitor. 
 
Last thought.  Struggling with an occasionally heard coincidence from several other SB220 owners.  SB220 running fine, hear a snap --- the Zener and grid choke are gone.  Later find one of the 3-500’s with an open filament.  The filament is not open but the solder has melted.  Seems to happen in laterally cooled 3-500’s.  The radios fitted with the Eimac chimney seem not to be effected.

What is the real problem?  Some blame the instability of the super cathode configuration and others have a variety of other reasons.  It would seem to me that the three 200pf capacitors from grid to ground would be adequate to quench most parasitic content and counter some of the lead inductance to the grid.

I would be interested in any comments especially regarding the SB220.

This has been rewritten for clarity.

Regards Jim
« Last Edit: August 13, 2015, 05:29:52 PM by K9AXN » Logged
W1BR
Member

Posts: 4163




Ignore
« Reply #48 on: August 12, 2015, 02:02:10 PM »

Last thought.  Struggling with an occasionally heard coincidence.  SB220 running fine, hear a snap --- the Zener and grid choke are gone.  Later find one of the 3-500’s with an open filament.  The filament is not open but the solder has melted.  Seems to happen in laterally cooled 3-500’s.  The radios fitted with the Eimac chimney seem not to be effected.

Snap was probably a flash over, which took out the cathode biasing Zener. Were the grids floating, or directly grounded??

Laterally cooled amps were generally used when the FCC power limits was at 1000 watts input.  Exceeding design limits, or dirty fans, or fans that needed lubrication were the most likely culprits.  


 "It would seem to me that the three 200pf capacitors from grid to ground would be adequate to quench most parasitic content and counter some of the lead inductance to the grid."


I'm also curious about your comment that 200 pF caps on the grids would snub parasitic oscillations... I am bit confused as to how that work.  The grid leads seem to be extremely short, and consist of three stout parallel leads to individual tube pins.  They should exhibit very low inductive values.  Perhaps you could clarify for all of our benefit?

73

Pete
« Last Edit: August 12, 2015, 02:05:43 PM by K1ZJH » Logged
AF6LJ
Member

Posts: 572




Ignore
« Reply #49 on: August 12, 2015, 03:03:27 PM »

Last thought.  Struggling with an occasionally heard coincidence.  SB220 running fine, hear a snap --- the Zener and grid choke are gone.  Later find one of the 3-500’s with an open filament.  The filament is not open but the solder has melted.  Seems to happen in laterally cooled 3-500’s.  The radios fitted with the Eimac chimney seem not to be effected.

Snap was probably a flash over, which took out the cathode biasing Zener. Were the grids floating, or directly grounded??

Laterally cooled amps were generally used when the FCC power limits was at 1000 watts input.  Exceeding design limits, or dirty fans, or fans that needed lubrication were the most likely culprits.  


 "It would seem to me that the three 200pf capacitors from grid to ground would be adequate to quench most parasitic content and counter some of the lead inductance to the grid."


I'm also curious about your comment that 200 pF caps on the grids would snub parasitic oscillations... I am bit confused as to how that work.  The grid leads seem to be extremely short, and consist of three stout parallel leads to individual tube pins.  They should exhibit very low inductive values.  Perhaps you could clarify for all of our benefit?

73

Pete

A couple of notes.
The SB-220 is marginally cooled under chassis by the fan. If the amplifier has had the front spacers removed from the front feet the cooling becomes even more marginal.

Your Ziner usually shorts out, which ends up with the tubes running at zero bias.

The cause of the flash-over is always a point of controversy.
Those caps are not there to snub parasitics, they are part of a feedback network used to lower IMD (by a small amount). As for the grid leads being short, that is a matter of poinion. Consider the physical distance between that ground lug and the actual grid in the tube. About as long as your ring finger. Smiley

Out of curiosity; what brand was the tube that lost its filament??
Logged

Take Care
Sue,
AF6LJ

Don't Kalifornicate My Life
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #50 on: August 13, 2015, 03:46:03 PM »




 "It would seem to me that the three 200pf capacitors from grid to ground would be adequate to quench most parasitic content and counter some of the lead inductance to the grid."


I'm also curious about your comment that 200 pF caps on the grids would snub parasitic oscillations... I am bit confused as to how that work.  The grid leads seem to be extremely short, and consist of three stout parallel leads to individual tube pins.  They should exhibit very low inductive values.  Perhaps you could clarify for all of our benefit?

73

Pete

It won't stabilize the amp, Pete. It also won't "counter some of the lead inductance" in a triode.

Grid to ground capacitance is sometimes used in a VHF amp for neutralization or improving screen shielding, but those are always tetrodes or pentodes. It is a bad idea in a triode, especially one with grid current. It also is only useful in a monoband amp. It can be used, in a multigrid tube, to for a bridge neutralization circuit.

That circuits is well known, and even appears in handbooks like Eimac's Care and Feeding of Power Grid Tubes.
Logged
K9AXN
Member

Posts: 441


WWW

Ignore
« Reply #51 on: August 13, 2015, 06:02:17 PM »

Sue,

I rewrote note #47 for clarity.

I agree that the primary purpose of the capacitors is to implement the feedback bridge.  What I was trying to articulate is the values selected were the best choice for stability.  Not that they were specifically to stabilize the tube.  Pay special attention to the the values and where the series resonant points are.  It's obvious when analyzing the 30S1 circuit that they took stability a step further by adding the parasitic suppressor in series with the feedback line.  Can't do it with a triode.

----------------------------------------------------------------------------------------------------------------------------------------
Tom,

I don't have a horse in this race ---- just looking for correct reliable information with explanations not just pointing to some reference.

I've asked this question several times.  Is the plate to grid interelectrode capacity in the 3-500 triode a part of the negative feedback circuit or is it not?  Please explain, need to know if I'm headed down the wrong logic path.

Have a super day Kindest regards Jim

   
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #52 on: August 13, 2015, 07:41:48 PM »

--------------------------------------------------------------------------------------------------
Tom,

I don't have a horse in this race ---- just looking for correct reliab
le information with explanations not just pointing to some reference.

I've asked this question several times.  Is the plate to grid interelectrode capacity in the 3-500 triode a part of the negative feedback circuit or is it not?  Please explain, need to know if I'm headed down the wrong logic path.

Have a super day Kindest regards Jim

   

That has already been answered several times.

It is grid-filament capacitance. Anode-grid at radio frequencies would be unstable regenerative feedback. Look at any neutralization circuit in the world. Anode to grid is always inverted phase externally. The exceptions to this would be some stuff at VHF or higher, but this is an HF application.

Look at the grid to cathode impedance through the RF cycle. Look at the Miller effect on capacitance. Look at the amount of grid-cathode capacitance in tube data sheets.

It is pretty easy to see why it is a silly system.

It measures poorly, it does not improve IMD in a triode. It tends to destabilize the amp when used in a triode. It hurts the gain flatness with both drive power changes and operating frequency changes.

It was a good idea in the 30S1. It was a terrible thing to use in triodes. It was pushed into systems outside Collins by Orr. I know for a fact he wrote and called Heath, Ameritron, and Dentron promoting that circuit. It went so far as he installed it in a AL80A and sent it to Ameritron. When I reported back it had worse IMD with the super-cathode, he changed the 3-500Z to a 4-400A. I'm not sure how it worked with the 4-400A, because he never sent it back in. As far as I knew, he ran it with a 4-400A in his AL80A after the 3-500 did not work so well.

Dentron got that goofy sweep tube circuit the same way. They tied the control grid to the cathode, and grounded the screen and beam forming plates. This destabilized the sweep tubes because the anode-cathode feed through capacitance went way up.
Logged
W5WSS
Member

Posts: 2283




Ignore
« Reply #53 on: August 13, 2015, 08:21:19 PM »

Hey Pete, Tom teaches you ey? That's not nonsense. Grin
Logged
W1BR
Member

Posts: 4163




Ignore
« Reply #54 on: August 13, 2015, 09:12:36 PM »

Hey Pete, Tom teaches you ey? That's not nonsense. Grin

Who?  Me?  If I have a question about amps I'd probably ask Tom or Carl.

Pete
Logged
K9AXN
Member

Posts: 441


WWW

Ignore
« Reply #55 on: August 14, 2015, 02:08:14 PM »

Tom,

Thanks for taking the time to reply.  I'm uncomfortable with the answer but will be moving on.

Have a great day --- Kindest regarsd Jim


That has already been answered several times.

It is grid-filament capacitance. Anode-grid at radio frequencies would be unstable regenerative feedback. Look at any neutralization circuit in the world. Anode to grid is always inverted phase externally. The exceptions to this would be some stuff at VHF or higher, but this is an HF application.

Look at the grid to cathode impedance through the RF cycle. Look at the Miller effect on capacitance. Look at the amount of grid-cathode capacitance in tube data sheets.

It is pretty easy to see why it is a silly system.

It measures poorly, it does not improve IMD in a triode. It tends to destabilize the amp when used in a triode. It hurts the gain flatness with both drive power changes and operating frequency changes.

It was a good idea in the 30S1. It was a terrible thing to use in triodes. It was pushed into systems outside Collins by Orr. I know for a fact he wrote and called Heath, Ameritron, and Dentron promoting that circuit. It went so far as he installed it in a AL80A and sent it to Ameritron. When I reported back it had worse IMD with the super-cathode, he changed the 3-500Z to a 4-400A. I'm not sure how it worked with the 4-400A, because he never sent it back in. As far as I knew, he ran it with a 4-400A in his AL80A after the 3-500 did not work so well.

Dentron got that goofy sweep tube circuit the same way. They tied the control grid to the cathode, and grounded the screen and beam forming plates. This destabilized the sweep tubes because the anode-cathode feed through capacitance went way up.
« Last Edit: August 14, 2015, 02:10:35 PM by K9AXN » Logged
AF6LJ
Member

Posts: 572




Ignore
« Reply #56 on: August 14, 2015, 02:13:07 PM »

Hey Pete, Tom teaches you ey? That's not nonsense. Grin

Who?  Me?  If I have a question about amps I'd probably ask Tom or Carl.

Pete
Same here.
Logged

Take Care
Sue,
AF6LJ

Don't Kalifornicate My Life
K9AXN
Member

Posts: 441


WWW

Ignore
« Reply #57 on: August 19, 2015, 09:12:03 AM »

Tom,
My curiosity is always peaked by incredible statistical anomalies.
Reading the endless dissertations that you’ve written regarding the abject horror of Super cathode drive and the effect that Bill Orr has had with your psyche, I wonder what his power source was.  Used properly it could bring peace to all of mankind.

You infer that Bill Orr was able to exert some kind of uncanny influence on so many of those designers that the concept became nearly universal.
 
Sounds like a visit to a séance where uncle Harry blew smoke through a now hear this tube torn from a 1904 wind jammer and quietly whispered Beware the ides of March --- Bill Orr – Bill Orr.

Probably a good idea to give your fingers a rest and leave the computer for a while.
Take your wife dancing, share a bottle of Champaign, crack a smile, take a long breath of the night air, search the night sky for a meteor tail, and if you’re lucky make a wish with your wife.
You’ll find there are more important things in life than sitting on you glutes getting fat and lazy.

I wish you peace my friend.
Kindest regards Jim K9AXN
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #58 on: August 21, 2015, 10:53:04 AM »

I can't help it you won't understand very simple basic things, Jim. I tried to explain it.

Logged
K9AXN
Member

Posts: 441


WWW

Ignore
« Reply #59 on: August 21, 2015, 06:57:30 PM »

Tom,

I have run out of time to tutor you.  You say the plate to grid capacity has no effect on the negative feedback, and assert that the cathode to grid capacity is the only participant in the grounded grid super cathode drive configuration.  With that, you disqualify any further discussion with you.
  
You might take a moment to think before you say that other folks just don't understand.  I'll see what can be done to find a local Elmer for you.

Good luck and Kindest regards Jim
« Last Edit: August 21, 2015, 08:28:50 PM by K9AXN » Logged
Pages: Prev 1 2 3 [4] 5 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!