The AM Forum
March 29, 2024, 02:46:33 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] 2   Go Down
  Print  
Author Topic: Tetrode Linear  (Read 24931 times)
0 Members and 1 Guest are viewing this topic.
WA1HZK
Member

Offline Offline

Posts: 1104


WWW
« on: December 29, 2004, 11:02:31 PM »

Does anyone have some input on the grid circuit for a linear using a single 4cx5000? I have all of the pieces & want to play with one of these. I know it's a zero grid current thing with the loading set for the normal screen current. What do you recommend for an input circuit? In VHF & UHF I used to just swamp my 4CX250B's with a 50 ohm resistor and that was about it. I assume I'll need more to generate a clean signal with the big brother. If you know where there are any practical designes posted, let me know. I have all of the tube spec's but am looking for fabrication ideas. I figure one cool running fat tube will last forever.
Keith
Logged

AM is Not A Hobby - It's a "Way of Life"!
Timmy, Sometime in 2007 on a Mountain Far Away..
www.criticalradio.com
www.criticalbattery.com
www.criticaltowers.com
www.criticalresponder.com
Official Registered "Old Buzzard"
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #1 on: December 30, 2004, 12:12:29 AM »

Quote from: WA1HZK
Does anyone have some input on the grid circuit for a linear using a single 4cx5000? Keith

Hi Keith,

You knew I'd answer this post didn't you?...  :lol:

Use the grid circuit from the Tesla 360 813 rig.  I've used this SIMPLE link coupled input with neutralization circuit for years in many high power tetrode linears and it always worked well. Needs only a floating variable single-section cap, not complicated split stators or link coupled plate outputs. It works FB with a standard pi-network on the plate output.

I usually use a small 150pf variable neutralization cap mounted and grounded on the front panel for neutralization adjustment. Real cornvenient to adjust on the fly..  Use a small aluminum or copper 1"X1" tab sitting on a stand-off insulator mounted near the plate structure for pickup sampling. The neu cap adjustment will shunt it to gnd as it tunes, so make it sample too much to start.

http://www.amwindow.org/tech/htm/813/813.htm

You want a stable neutralized stage and the neutralization does this plus gives it some neg feedback for better 3rd order IMD.  The input circuit usually needs about 5 turns for 50 ohms. Be sure you wind the input link on the high/hot side of the grid tank coil.   I usually use a roller inductor there too.

BTW, you will generate some good voltage at the grid, so use a 200 pf that can handle at least 1KV or it will arc.

Use a swamping resistor as LAST resort. I have never needed one with this circuit assuming it neutralizes cold with NO grid-plate feedthru on the scope.

73,
T

Good luck, OM.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1HZK
Member

Offline Offline

Posts: 1104


WWW
OK
« Reply #2 on: December 30, 2004, 11:32:49 AM »

Thanks
I tried that circuit with my 833 rig but could never stop the arcing around that input coil. trying to get the input coil built around the roller inductor always caused a fire. I ended up using a standard split stator cap, roller inductor and balanced input circuit with the regular neut. setup. That's the only thing that worked for the 833's. However those triodes took a couple hundred watts to properly drive them. The big tetrodes take no power, just volts, right? How did you construct your input coil around the roller inductor? Did you modify the inductor to do this? I spend so much time fixing blown up rigs I may not be able to hook this up anytime soon but it makes for a good posting.
Keith
Logged

AM is Not A Hobby - It's a "Way of Life"!
Timmy, Sometime in 2007 on a Mountain Far Away..
www.criticalradio.com
www.criticalbattery.com
www.criticaltowers.com
www.criticalresponder.com
Official Registered "Old Buzzard"
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #3 on: December 30, 2004, 11:49:22 AM »

Rick Measures has an interesting grid circuit on his site.....
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #4 on: December 30, 2004, 11:59:45 AM »

Quote from: WA1HZK
Thanks
I tried that circuit with my 833 rig but could never stop the arcing around that input coil. trying to get the input coil built around the roller inductor always caused a fire.  How did you construct your input coil around the roller inductor? Did you modify the inductor to do this? Keith


Hmmm....   I used a standard LaPointe roller inductor and wound the link around the outside frame using HV insulated wire. Never had any arcing. There was about 300VDC bias and maybe 300V RF on top of that there.

The grid tuning cap had to be 1KV spacing, but the circuit always worked FB. Maybe if you put the roller on insulator standoffs it mighta helped.

I think we went thru this thread before when you built it.  You might check out Frank's idea of the Richard Measure's circuit too.

Whatever it takes to neutralize the tetrode and drive it with a high Q grid circuit is all that matters. Many ways to get there.

Zero grid current, recommended screen current and regulated - ..... proper loading, plenty of air and you're golden, OM.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
W2VW
Contributing
Member
*
Offline Offline

Posts: 3489


WWW
« Reply #5 on: December 30, 2004, 12:09:06 PM »

Unless you really need more than 30 db stage gain, my vote is for putting normal grid bias and screen Voltage on the tube and then driving the cathode. When driving the grid, you need to be exactly on the mark with neutralization or the amplifier will take off like it has a solid rocket booster with a fuel leak. Too much monkey business especially if trying to multiband the box. You need either to configure the thing in grounded screen or grab a socket with a very low inductance bypass. If you don't have a good screen bypass, you can fabricate one using some brass stock and Teflon sheeting. Use the underside of the chassis for one plate of the cap, insulate with the Teflon and the brass plate goes to the screen. The whole thing is held together with insulated hardware. The key to fabricating a cap like this is to keep the connection to the screen short. Ideally, the fabricated cap will surround the tube. This will take some room on the chassis.
Any way that you go, figure the correct surge supressors to put across the screen bypass. There are some good units available.
Filament choke can be a couple of paralleled units like the H.D. ones R.F. Parts sells. If you plan use on 160 then the paralleled chokes might be an issue. You could always parallel 2 seriesed pair$. Pull filament meter Voltage connection right off the socket through a couple of small bypassed R.F. chokes. You will want the ability to carefully measure and adjust fil. Volts. Watch out for panel meters that are not designed to be mounted on your type of panel metal. Meters are calibrated according to aluminum or steel of a given thickness. Allow the filament circuit to be adjustable using a variac. You will want to ramp the Voltage up on warmup to save tube life. Also, having extra fil. Voltage available might help "recondition" an old tube with low emission. Econco pays about $35.00 for 4CX5000 duds
    Even if you do stay with grid driven configuration, you can still use a filament choke to facilitate cathode driven service. There's no reason why you couldn't figure a scheme to switch beteween the 2.
     There is some good information in the amps reflector archives. Pay attention to anyone who actually has built an amplifier with this tube verses someone who once saw a guy who's cousin had a working 4CX1000.
Consider applying to the electric utility based on industrial peak demand billing. You still need to burn 600 Watts just to light the thing even in standby. If you don't key the B+ then consider a scheme to half the Voltage during standby. This will reduce the bleeder wattage to 1/4. You can use a relay to switch the primary(s) of your plate iron from 240 to 125 on standby.  :mrgreen:
     Use anything that will keep your signal clean. A swinging choke input plate supply will help. Remember all the stuff we learned about Class AB/B modulators needing excellent Plate Voltage regulation. Well this amplifier is no different. All the parameters of a modulator are shared here except for the operational frequency. Why should somebody expect to get away with a capacitor input or even Voltage doubler supply just because this is R.F. and not audio? Keeping the plate Volts rock stable will also pay off by not allowing the Voltage to soar under low loads. This will keep you further away from parasitic oscillations and even X-rays which might only take place at higher potential. Swinging chokes can be connected in the negative supply lead if you cannot find a choke with the proper hi-pot rating. Series all the chokes you can come up with to cut down on critical bleeder current. Use spark gaps across stuff like this. You might even try to resonate the choke(s). I never did this myself. A stiff screen Voltage is absolutely a must. You also want adjustability as the screen Voltage will control gain. Another thing to keep the amp clean would be degenerative feedback. The Cathode driven configuration has this built in. Grid driven can be set up to do this also but you absolutely need a stabile circuit when introducing cathode resistance for feedback. Don't try to drive this thing on AM with an exciter that uses a crappy modulation scheme. I prefer a ricebox with a good filter to define bandwidth. No need to make friends all over the spectrum.
    Consider using control circuitry that allows overcurrent to shut off the plate supply AND the screen supply. The screen will try to act like the plate if you lose plate Voltage and will not last too long. I've seen a lot of and used a few neat gizmos advertised as being best way to feed screen Voltage to a grid driven tetrode and really have to wonder what is so bad about an ald fashoned stiff supply with overcurrent cutout. I also like run-on sentences.
You should end up with an amplifier that can make the legal limit with plenty of dynamic headroom.
Logged
WA1HZK
Member

Offline Offline

Posts: 1104


WWW
OK
« Reply #6 on: December 30, 2004, 02:51:45 PM »

Good input, Thanks.
I have a "R" Version of the tube that is supposed to be Virgin plus some pulls. I have the right socket & Plate transformer and am collecting ideas. I had first considered Grounded Grid but some folks kept saying drive the grid. I wrecked a few 4cx250B's before and really do not want to do that again. So build a big mother Fil. choke & Bypass the grid to ground with RF caps & apply the neg. voltage to determine the idle current & switch the tube on or off. Sounds kinda simple. All of my transmitters can work into odd loads. Do you still think I'll need a Pi-Net on the input or just drive it through a cap? PS, my 833 rig eats 500 watts just sitting there so whats the difference. I live about 15 miles from the Seabrook Nuke Plant!
Logged

AM is Not A Hobby - It's a "Way of Life"!
Timmy, Sometime in 2007 on a Mountain Far Away..
www.criticalradio.com
www.criticalbattery.com
www.criticaltowers.com
www.criticalresponder.com
Official Registered "Old Buzzard"
W2VW
Contributing
Member
*
Offline Offline

Posts: 3489


WWW
« Reply #7 on: December 30, 2004, 03:49:38 PM »

A tuned input is nice even if only for keeping things at 50 Ohms for instrumentation. A Ranger of Valiant would probably play into it without one. I'm sold on dealing with tuned inputs.
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #8 on: December 30, 2004, 04:36:12 PM »

Keith,
I ran a grounded grid 4CX3000A for many years. I just connected the grids to ground and drove the cathode with a pi network and also use a
La Pointe roller inductor. (Inside joke Tom Vu) I used the smaller one.
BTW I have a stash of the spring loaded parts as spares.
I used a string of 4- 5 volt 10 watt zeners in series with a switch to select
0 to 15 volts of bias on the cathode.  A pair of UTC CG310s provided
4 KV of motivation and the Viking 2 at 80 watts carrier drove it ok fine.
I got a bit more gain from the 4CX3000A than the 4-1000A in the same configuration. The rig was taken apart for tha addition of vacuuum variables and never put it back on.  I had a hum problem for a while but corrected it with a better fil choke.
I suspect the innards of the 4X5000 are very similar so bet it would work fine. You will have a nice room heater lighting it up.
Logged
W2VW
Contributing
Member
*
Offline Offline

Posts: 3489


WWW
« Reply #9 on: December 17, 2010, 02:45:37 PM »

Cathode drive it with control and screen grid voltages still present.
Sounds like there's plenty of drive available to you.
Logged
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #10 on: December 17, 2010, 09:18:29 PM »

Id rather run a YC-156 and not have any of those worries. Bolt the grid flange right to the chassis and use it as a dummy load if you want. Run 2500-8000V depending on the thrills you want. It has a 6KW dissipation with normal rated air flow, a lot more if you run a big blower. No neutralization and no parasitics either yet it has very high gain for a triode. And the filament is only 15V @ 15A.

Makes a great AM linear and almost indestructible.

Carl
KM1H
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #11 on: December 17, 2010, 11:19:20 PM »

Having run both, I agree with Carl.

You can get MORE than enough to get LOTS of YC-156s. 

Gain = close to the same.

Difference in grid dissipation is a moot point, if you bolt the grid flange on the YC tube to a LARGE spreader...  It's the grid, think heat sink.

Grounded grid, I drove my 4CX5000R with close to 2K for full pep into the dummy load.  22 grand input was the norm.  14 - 16K output, dependant on freq.  With it grid driven, I found it was about 20-22 dB gain, freq dependant.

The YC-156 I've only run monoband, and honestly, it's a LOT easier.  NO parasitic suppressor necessary, bolt in, plug and pray!  Slightly longer warm up time, slightly lower gain (in grid drive service on the 5K), but all in all, who needs to produce 14 grand from 125 watts input?

--Shane
KD6VXI
Logged
W2PFY
Contributing
Member
*
Offline Offline

Posts: 13312



« Reply #12 on: December 17, 2010, 11:56:25 PM »

Quote
The YC-156 I've only run monoband

How much power will a YC-156 run in AM linear?
Logged

The secrecy of my job prevents me from knowing what I am doing.
W2VW
Contributing
Member
*
Offline Offline

Posts: 3489


WWW
« Reply #13 on: December 18, 2010, 07:14:15 AM »

2 KW carrier with plenty of headroom. Need more use a larger valve.

Grid current should be treated just like a tetrode. Monitor it carefully.



Quote
The YC-156 I've only run monoband

How much power will a YC-156 run in AM linear?
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #14 on: December 18, 2010, 09:49:33 AM »

An option is cathode drive it but add some screen volts  to increase the resting current and gain. I agree with Dave find someone who really built this type of linear that works on multiple bands. Fairly easy to get it to play on one band but getting the layout right for multiple bands is a tall order. The 4X5 grids can take some power unlike the 4CX250. A valiant as an exciter HMMM.
Logged
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #15 on: December 18, 2010, 10:42:31 AM »

Quote
2 KW carrier with plenty of headroom. Need more use a larger valve.

Grid current should be treated just like a tetrode. Monitor it carefully.


Or add a second tube they are cheap enough at $300-350 for a full output medical pull.  Grid current is the ideal tuning indicator just as screen current is for tetrodes and pentodes. Peak both. At least with the YC-156 you dont have to worry about blowing the grid with a mistune. The specs say 6 minutes or more for filament warmup; most go with 8 minutes.

Carl
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #16 on: December 18, 2010, 10:59:15 AM »

Nice thread revival from 2004.


Carl said: "Run 2500-8000V"


Wow... has anyone actually run a YC-156 over 6KV for a long period of time without "barnacle" arcing?  


Yes, it is a VERY robust tube. There is a story on the web of a company doing a test. They ran one all night at 40KW out. The thing was black in the morning but was still working. After all, it's really a 3CX-15000A7 inside - just has a smaller anode heatsink.

They are starting to get rare due to the CBers burning them up and hams buying them for projects and letting them sit in the closet. But the biggest problem is the MRI machines have switched to solid state finals and the pipeline supply is now zip.

If interested, grab one while you can, as they will go the way of 750TL's eventually. Unobtainium.  Ebay has them used sometimes, but ususally for $300+ and some real nasty looking. NOS ones can cost $600++ these days. Brand new ones are what, $1500?

Best tube ever made for linear operation at reasonable fil power consumption. Like four 8877's in parallel without the trouble and cost.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
W2VW
Contributing
Member
*
Offline Offline

Posts: 3489


WWW
« Reply #17 on: December 19, 2010, 09:54:21 AM »



Difference in grid dissipation is a moot point, if you bolt the grid flange on the YC tube to a LARGE spreader...  It's the grid, think heat sink.


I doubt the grid will actually be able to dissipate even one more watt safely.

The construction of the grid itself does not appear to be able to conduct very much heat toward the flange. If this were true Eimac may have had an inclination to put it on the specification sheet.

Certain other radio services seem to have the mindset to always run amplifiers and components way past their ratings. I'm all for kicking the crap out of components in the interest of learning but not with a device like a YC-156. Failures are not pretty.

There is plenty of reliable data out there on use of these tubes in amateur service.
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #18 on: December 19, 2010, 10:22:37 AM »



Difference in grid dissipation is a moot point, if you bolt the grid flange on the YC tube to a LARGE spreader...  It's the grid, think heat sink.


I doubt the grid will actually be able to dissipate even one more watt safely.

The construction of the grid itself does not appear to be able to conduct very much heat toward the flange. If this were true Eimac may have had an inclination to put it on the specification sheet.

Certain other radio services seem to have the mindset to always run amplifiers and components way past their ratings. I'm all for kicking the crap out of components in the interest of learning but not with a device like a YC-156. Failures are not pretty.

There is plenty of reliable data out there on use of these tubes in amateur service.


My info came from sharing info with VE7RF and his use of the direct grid mounted tubes. 

He builds them for amateur bands.

From what I remember, the data he had came from Reid.  Senor Brandon, at Eimac.

Bolting a device to a heat spreader and heat sink doesn't increase Diss of the device?

Having built MULTIPLE amps with the tube, I'll take your bet.


--Shane
KD6VXI
Logged
KM1H
Contributing
Member
*
Offline Offline

Posts: 3519



« Reply #19 on: December 19, 2010, 10:40:37 AM »

Quote
Carl said: "Run 2500-8000V"


Wow... has anyone actually run over 6KV for a long period of time without "barnacle" arcing?

If it doesnt hi-pot to 8KV then it needs to be de-barnacled.  Considering they are rated to 6500V in industrial pulse service, and SSB for the 3CX15000B7, Id say there is room for 8KV in CW/SSB service anyway.

They are specifically not rated for Class C which says that the ultimate voltage breakdown is not sufficient. Id guesstimate that 4KV DC is close to pushing it for AM if you want it to survive and not become a CBers paperweight.

OTOH the 3CX10000A7 is rated to 6500V as a plate modulated Class C tube. Why the difference between them? Marketing or is it real?

One of these days I'll find out, Im one of those guys who got a full output pull almost 20 years ago ($200) and its still in the carton. The hard part is over, I have a transformer and huge oil cap to do it justice and a Variac that I can barely lift.

Carl
Logged
W2VW
Contributing
Member
*
Offline Offline

Posts: 3489


WWW
« Reply #20 on: December 19, 2010, 11:18:57 AM »

A wise man referred to the learning process as "the price of admission."




Logged
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4484



« Reply #21 on: December 19, 2010, 06:45:46 PM »

 "   Nice thread revival from 2004.  "

Sometimes its better to not rush in with a comment.


klc

Logged

What? Me worry?
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #22 on: December 19, 2010, 10:31:46 PM »

Glad to see the thread come back...I'm one of those guys with a couple 4cx5000's,a socket,fil trans, 7kv plate xfmr and some vac caps that are in storage.......waiting for me to get enough understanding to  build something ...I should print out this thread and hang it on the shack wall.....as inspiration is a fleeting thing..
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #23 on: December 19, 2010, 10:46:19 PM »

http://shop.ebay.com/?_from=R40&_trksid=p5197.m570.l1313&_nkw=yc-156&_sacat=See-All-Categories

Looks like a clean pull "buy it now" on eBay for $550.  He has five.


The second one looks like a misprint at $6K.. Wink

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
W2VW
Contributing
Member
*
Offline Offline

Posts: 3489


WWW
« Reply #24 on: December 20, 2010, 07:57:36 PM »

Here's a pic of YC-156 guts. The grid flange is the plated ring in the bottom of the rightmost picture. The grid is the screen looking structure in the center picture.
It's pretty obvious that the grid dissipation cannot be doubled by cooling the grid flange.

http://www.wv7u.com/yc156amp/pictures/tube.jpg
Logged
Pages: [1] 2   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 0.078 seconds with 18 queries.