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: [1] 2 3 4 5 Next   Go Down
  Print  
Author Topic: Copper Strip For Grid Grounding in TL922A  (Read 21794 times)
N1UK
Member

Posts: 2211




Ignore
« on: July 23, 2015, 07:46:50 AM »

I am repairing a TL922A which looks to have had a tube anode to grid short. The 470 uH grid grounding choke has been vapourised.  I like the way that W8JI has grounded the grids on this amp using copper strip. This seems to be the way to go since the original Kenwood coils don't seem to be available anymore.

http://www.w8ji.com/tl922_mods_parasitic.htm

I have some 1/4 inch 25 mil copper strip but it seems a bit thick to work easily. I found a site that sells 1/4 inch copper strip in a variety of thicknesses such as 1, 2, 3, 4, 5, 6, 7, 8, and 16 mils.  Without previous experience what would be the best thickness of copper strip to use for this task?

A second question...if I ground the grids can I use the original Kenwood anode suppression assemblies


thanks  Mark N1UK
Logged
G3RZP
Member

Posts: 1159




Ignore
« Reply #1 on: July 23, 2015, 08:00:48 AM »

1 skin depth in copper at 3.5MHz is 1.39 mil. You need several skin depths, because one skin depth is where the current is 1/e or roughly 37% of the surface current. So you ideally want a strip that is ten skin depths thick or about 14 mil.

Now the length involved is very small, so you would probably find in practice that there was no measurable difference between 8 and 14 mil. But I wouldn't go thinner than 8 mil.

At 28 MHz, one skin depth is 0.59 mil.
Logged
KM1H
Member

Posts: 4792




Ignore
« Reply #2 on: July 23, 2015, 08:25:56 AM »

Plus the amp goes to 160.

Since many thousand TL-922's are still running fine without benefit of mods then why bother with trying to second guess the designers. Chokes are available many places with a little looking and any value from 470 to 1000uH at 300ma or more is OK.

The only 922 mod I feel is worthwhile when running a fast switching SS rig is to get rid of that clunky relay if using VOX or any other rapid TX/RX operation. A foot switch or PTT is OK to prevent hot switching which is the real cause of burnt band switches, arced caps, tube arcs (other than residual gas which should have been regularly regettered) and blown grid chokes.

If the original parasitic suppressor resistors are within tolerance they are OK; otherwise replace with Ohmite OY type from Mouser, etc.

Carl
Logged
N1UK
Member

Posts: 2211




Ignore
« Reply #3 on: July 23, 2015, 08:41:25 AM »

I did find these chokes, which I think would work OK

http://www.mouser.com/ProductDetail/Bourns/5800-471-RC/?qs=sGAEpiMZZMv126LJFLh8y4rtocMwlKUK4usuRa1xM5o%3d


The 220pF silva mica caps would need replacing. I am not sure of the working voltage

Kenwood  CM93D2H221J   220pF +/- 5%


I guess that I will have to lift one end of the suppressor resistors to be able to measure the resistance


Mark N1UK
Logged
WA7PRC
Member

Posts: 2314


WWW

Ignore
« Reply #4 on: July 23, 2015, 10:34:07 AM »

On my SB-220, I hard-grounded the grids using 0.032" thick soft copper. Each socket terminal is connected to chassis:

(click for larger image)
You can't make the connection any shorter (the XL is the lowest possible). Hardware is #4-40.

Quote
The only 922 mod I feel is worthwhile when running a fast switching SS rig is to get rid of that clunky relay if using VOX or any other rapid TX/RX operation. A foot switch or PTT is OK to prevent hot switching which is the real cause of burnt band switches, arced caps, tube arcs (other than residual gas which should have been regularly regettered) and blown grid chokes.
While MOX (Manual Operated Control) is OK, use of FAST relays is the BEST way to prevent hot-switching. They switch in about 1 millisecond (open-frame relays take about 25 milliseconds to toggle and stop bouncing). Many modern solid state transceivers take 5 to 10 milliseconds to produce RF after PTT is toggled. I use W7RY's PCB in my SB220. Jim has similar boards for other amplifiers (including Kenwood).

Quote
If the original parasitic suppressor resistors are within tolerance they are OK; otherwise replace with Ohmite OY type from Mouser, etc.
New old stock Allen-Bradley 2W carbon composition resistors can still be found. Ebay seller 'bigsmythe74' (link) has them for about $2.00 each.
Logged
WB2WIK
Member

Posts: 21836




Ignore
« Reply #5 on: July 23, 2015, 10:52:12 AM »


While MOX (Manual Operated Control) is OK, use of FAST relays is the BEST way to prevent hot-switching. They switch in about 1 millisecond (open-frame relays take about 25 milliseconds to toggle and stop bouncing). Many modern solid state transceivers take 5 to 10 milliseconds to produce RF after PTT is toggled. I use W7RY's PCB in my SB220. Jim has similar boards for other amplifiers (including Kenwood).

Some rigs allow you to set this timing (RLY KEY closure vs. RF output, both when keyed and when unkeyed), my Ten Tecs do and you can set it to whatever you wish.

Not all rigs have that option.  I use open frame amp relays but set the timing to ~30 mS so nothing can ever hot switch.  Not good for QSK, though. Tongue

Logged
WA7PRC
Member

Posts: 2314


WWW

Ignore
« Reply #6 on: July 23, 2015, 11:21:39 AM »


While MOX (Manual Operated Control) is OK, use of FAST relays is the BEST way to prevent hot-switching. They switch in about 1 millisecond (open-frame relays take about 25 milliseconds to toggle and stop bouncing). Many modern solid state transceivers take 5 to 10 milliseconds to produce RF after PTT is toggled. I use W7RY's PCB in my SB220. Jim has similar boards for other amplifiers (including Kenwood).

Some rigs allow you to set this timing (RLY KEY closure vs. RF output, both when keyed and when unkeyed), my Ten Tecs do and you can set it to whatever you wish.

Not all rigs have that option.  I use open frame amp relays but set the timing to ~30 mS so nothing can ever hot switch.  Not good for QSK, though. Tongue
Yeah. Until I changed to vacuum & reed relays, I had a delay programmed into my logger/keyer program (WriteLog), and used MOX. The trick with MOX is to remember to toggle PTT first.

Hermetically sealed relay contacts will NEVER oxidize (and cause intermittent operation). They're also DEAD quiet. I have to listen closely in a quiet room in order to hear the vacuum and reed relays toggle. Though they switch fast, you can quickly blow through their 1- or 2-million guaranteed lifecycle if you use full QSK. Better to limit use to VOX (fone) and semi break-in (Morse). If full QSK is desired, we're talking diode switching, and as you are well-aware, that's a different set of concerns.
Logged
KM1H
Member

Posts: 4792




Ignore
« Reply #7 on: July 23, 2015, 11:57:03 AM »

Quote
The 220pF silva mica caps would need replacing. I am not sure of the working voltage

A generic 500V.

My 1986 LK-500ZC came with the QSK option which included the vacuum relay but in the thousands and thousands of contest and DX QSO's I never used QSK and it certainly didnt keep from winning most of the contests entered or missing a new DXCC contact. VOX was used for SSB and CW.

Other amps used in the second station or for backup use the stock open frame relays with a speed up circuit, no hot switching in several decades going right back to the NCL-2000 of the 60's.
Logged
W1QJ
Member

Posts: 2927




Ignore
« Reply #8 on: July 23, 2015, 02:36:45 PM »

Mark, ground the grids with braid from RG-58.  Keep run short to ground.  It works fine.
Logged
K8AXW
Member

Posts: 7033




Ignore
« Reply #9 on: July 24, 2015, 10:26:34 AM »

My question is different but in line with the present thread.

The 922 and 220 have coils/caps from grid to ground.  When there is a fault they fail.

The questions I have are, "What happens when the coil/caps are removed from the grid(s) and are then tied directly to ground and a fault occurs?"

I've had grid resistors fail due to high grid current.  I replace the resistor and I'm then back in the game.  I feel that if the grid was tied directly to ground, the grid would then vaporize instead of the resistor.  Kiss off one expensive tube!  Right or wrong?



Logged

A Pessimist is Never Disappointed!
W1BR
Member

Posts: 4163




Ignore
« Reply #10 on: July 24, 2015, 11:29:46 AM »

Well, I've read what Tom has to say:

http://www.w8ji.com/grounded_grid_amplifiers.htm
http://www.w8ji.com/arc_protection_al572.htm

I've read that direct grounding increases IMD generation by a small amount, and there are pro and con arguments that have been posted on here in the past. I thing I know, if the tube arcs, there should be surge resistor in the B+ to take the peak current, and that I would suspect once the grid resistors vaporize the cathode biasing devices would be toast.

Pete
Logged
KM1H
Member

Posts: 4792




Ignore
« Reply #11 on: July 24, 2015, 03:50:34 PM »

No grid resistor is used in Drake, Heath, Kenwood, or others using that scheme; instead a RF choke of 470 to 1000 uH or so at 300 ma will often serve to protect the tube during an arc. Sometimes they fail open but often the inductive kick slams the pies together. The Drake chokes were a much higher current and very low resistance while the Heath chokes are 25 Ohm current limiters.

The average size current limiting resistor will not always protect the tube from a grid short before the amps fuse or breaker acts.

All I see on that site Pete mentioned is one persons opinion and IF I was building a 3-500Z amp from scratch Id ground the grid as Amp Supply did in the early 80's since the complete design and layout is inherently stable.

I cant say that about the Heath, Kenwood, and some directly grounded commercial amps.

Carl
Logged
WA7PRC
Member

Posts: 2314


WWW

Ignore
« Reply #12 on: July 25, 2015, 12:04:23 AM »

Keep run short to ground.  It works fine.
That was my intent. You can't get any shorter than the way I did it,
using soft copper strip that I cut using scissors:

(click for larger image)

Afterward, I found my SB-220 easier to drive and dead stable.
I also use a glitch resistor (Ohmite pn B20J10RE) in place of RFC2.

vy 73,
Bryan WA7PRC
Logged
KM1H
Member

Posts: 4792




Ignore
« Reply #13 on: July 25, 2015, 08:02:01 AM »

Quote
Afterward, I found my SB-220 easier to drive and dead stable.
I also use a glitch resistor (Ohmite pn B20J10RE) in place of RFC2.

Replacing RFC-2 with a 10-15 Ohm WW resistor was something I wrote up in the mid 80's plus on the old AMPS reflector in the mid 90's and certainly added to the stability.
Logged
WA7PRC
Member

Posts: 2314


WWW

Ignore
« Reply #14 on: July 25, 2015, 09:58:38 AM »

Quote
Afterward, I found my SB-220 easier to drive and dead stable.
I also use a glitch resistor (Ohmite pn B20J10RE) in place of RFC2.
Replacing RFC-2 with a 10-15 Ohm WW resistor was something I wrote up in the mid 80's plus on the old AMPS reflector in the mid 90's and certainly added to the stability.
The concept may have been your idea but, I used the selection info provided by Tom W8JI (link).

My SB220 was stable after I hard-grounded the grids, and before & after I added the glitch resistor. The resistor measures 1 μH on my AADE L/C Meter. RFC2 was 8.5 μH.
Logged
Pages: [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!