Johnson Thunderbolt restoration

[ke7trp]

Ok. Its time to start the Tbolt project.  I had a nice Tbolt.  But someone plugged it into 220 volt when the unit was wired for 110 volt and now its chokes and plate trans is history.

I had a fellow Friend w7MD Donate a second Tbolt for free!  That Tbolt had a bad oil cap and other "mods" done to it around the Regulator section.  There are banks of resitors lined up and some repair work. 

The plan is to take two Tbolts and come out with One working Tbolt.  I will of course need help so I started this thread  

I will now start the process of moving parts around. 

1. The First question is about the three caps on the Regulator filter board.  C102, C103 and C104.  They are 30MFD at 450volt..  Can I use 100 MFD here?   I have a bunch of 100 MFD 450 volt caps on hand.  I dont know much about the tube regulators and my gut was that it might load them down having tripple the capacitance.  If I cant use them, I will order some 47MFD at 450 and use those.

2. The HV cap in the one Tbolt is missing, The second Tbolt has an 8MFD "Pyronol GE" cap rated at 2000 volt.  This scares me as the Tbolt has a design voltage of 2350 according to the manual and that would be on 220 volt.  I have 246 volts here. That might make 2500+ unloaded.   What would you do? Run it?  Find another that will fit rated at higher voltage?  Get a board and use computer caps in a bank?   How about one of the new style caps that look like hockey pucks.. The will fit easy but are costly.

Remember that I am fully aware of the shortcommings of this amplifier, I do not wish to modify the hell out of it.  Stock performance is fine.  I am going to use it with a ranger and HRO60 on a vintage Steel case desk.  Probably on 20 meter AM once in a while.  If I get 200 watts, thats great.  Its more about having it and using it then it is about it putting out alot of power ect...

C

[Pete, WA2CWA]

2. The HV cap in the one Tbolt is missing, The second Tbolt has an 8MFD "Pyronol GE" cap rated at 2000 volt.  This scares me as the Tbolt has a design voltage of 2350 according to the manual and that would be on 220 volt.  I have 246 volts here. That might make 2500+ unloaded.   What would you do? Run it?  Find another that will fit rated at higher voltage?  Get a board and use computer caps in a bank?   How about one of the new style caps that look like hockey pucks.. The will fit easy but are costly.

Later versions of the Thunderbolt got rid of the 8 mfd HV Cap. and used 6, 80 mfd/450V Caps in series. Each cap has a 120K, 2 watt resistor across it.

[ke7trp]

Good to know...  I will download the later manual and have a look.  If I can find some fiberglass, I can Drill out and make a board with the 6 caps and add the resistors.   I can stud mount it inside the amp. I can use 6 x 100 MFD x 450 volt. 

Thanks pete.

How about the Regulator caps?

[Opcom]

The one I had ran 2500V idling and about 2200V at 600W CW output on 10M. It was 240V there. The cap never did fail, but I don't recall it being a 2KV unit or what. Why not plan for the higher voltage in your filter? Also consider if the dropping resistance for the regulator needs increased slightly to keep the regulator current in spec.

[ke7trp]

Yes, I planned on a 3000 volt filter spec to be safe.  I made a typo above. My wall voltage is 256 volts. Not 246.. So volts are high on PS supplys.

We will check the HV and Screen HV once its up and running on the bench. One thing is that you can swap the regulator tubes around to OD3/OC3 ect ect to get back to the design voltage.

C

[ke7trp]

I think I will order the CDE 944U caps tomorrow for the HV.  They make a 33UF 1400volt DC cap.  The cap is 2.8 inches by 1.5 tall roughly.  I can mount two of them easily on the Tbolt chassis. Then wire them up.  I will have 16.5 UF with 2800 volt working.  I see no downside to this other then the $80. 

Thoughts?

http://www.cde.com/catalogs/944U.pdf

[ke7trp]

Robert confirmed that the 944U is the way to go now.  So I ordered them up from TTI. They had the best price.

Today, I will rebuild and transplant the Regulator filter board into the good Tbolt, Then wire the rig for 220 volt.   

When the caps get here, i will mount them up and wire them in.  Lots of cleaning to do...
C

[W2PFY]

Remember that I am fully aware of the short comings of this amplifier, I do not wish to modify the hell out of it.  Stock performance is fine.  I am going to use it with a ranger and HRO60 on a vintage Steel case desk.  Probably on 20 meter AM once in a while.  If I get 200 watts, that's great.  Its more about having it and using it then it is about it putting out a lot of power ect...

Good move, I get a little tired of people wanting  to make something out of something that is not after the modifications, anything like the something that they had at one time

[ke7trp]

still no answer on replacing the 30MFD caps with 100s for the regulator filter board.  I guess I will go ahead and use the 100's.  I hope it does not throw the regulators off and cause them to motorboat.

C

[N0WEK]

still no answer on replacing the 30MFD caps with 100s for the regulator filter board.  I guess I will go ahead and use the 100's.  I hope it does not throw the regulators off and cause them to motorboat.

C

If that thing doesn't have soft start, you may want to add it with that capacity increase.

[ke7trp]

Its only a few more mfd....   

[ke7trp]

Got the cde 944 caps.  Bit larger then i thought.  They will fit though. I need to cut down the studs.  How much clearance from the end of the stud to the chassis do i need for 3kv?

[ke7trp]

Ok.  Found a few hours to work on the Tbolts.  I got The new caps on the LV supply board. I got the three large resistors installed back into the Tbolt. Someone removed them all and used a handfull of small resistors.  Not sure why.  I put it all back factory with factory hardware from the blown up Tbolt and soldered them in.

I replaced the broken fuse holders.  The GUY that burned her down broke them off.

I oiled and cleaned the contacts, bushings and switches.  Everything moves free now.

I am going to take the 944 caps to work tomorrow and cut the Studs down. They are to long and I am worried there might be an arc.    I only need enough thread for a ring terminal.

I will stop at hardware store and get mounting hardware for the caps.

Another hour of work and the Tbolt will be fired up.  If that passes, I will load her up. I have a busy 2 weeks ahead of me so I hope to find the time.

C

[DMOD]

Good luck with the Tbolt Clark.


BTW, I had a pleasant visit with Clark while out west visiting my son.

Needless to say, Clark has an impressive collection of Heavy Iron.

And I also got to see his Inverted-L and the "famous" loading coil up close.

Phil - AC0OB

[ke7trp]

It was nice meeting you Phil.  Glad you could stop by. You are welcome anytime man.

Here is a pic of the Tbolt on my little work bench.  You can see the resistors mounted and the new filter board. 

Pictured is the 944U High current Filter cap.  33mfd at 1400 working DC.  You can go 150% on the B+ at 75 amps for 10 minutes.  The two will be 2800 working and 16.5 MFD.  Only a few more then the original but twice the value of the older style Tbolt that came with an 8MFD at 2000.

Cant wait to fire this ole girl up and see what she lays down.

[Opcom]

Do you mean the little round black item is the 33/1400V CDE 944U cap? What did that little jewel cost if I may ask?

I've also looked at these similarly rated caps, from Illinois Capacitor.
http://www.illinoiscapacitor.com/products/power-film.aspx

Can anyone say what differences there may be between a film cap rated for ripple filter current and for "high dv/dt" use? There are a lot of special caps out there.

[ke7trp]

The Ill caps look the same.  They are used for Electric motor inverters and in some electric bikes and cars.

I got them all mounted up and wired up today.  Now I have to go through the amp to make sure its wired for 220 volts. Not easy since there is little information in the manual. The entire terminal strip does not even show up on the schematic. I have to go through wire but wire.


C

[ke7trp]

Ok. I found some interesting things about these two tbolts.  They where run on 120 volt.  I know that because they have the jumpers.  What was NOT done was to jumper the back of the connector where the power enters the Tbolt chassis.  If you dont jump the two HOT's, the other wont get power.  I wonder if this is why some people say the amp does 600 and some say it does 1200?   

Maybe the HAM's that wired them put the jumper inside the twist lock connector on the cord side?  Maybe they thought that was easier then following the instructions.

My good friend is comming over tomorrow to double check my work before we put power to the old girl.  It will be on 256 volts where the manual says 230 to 250 volts. So its only going to be 6 volts over the Johson spec.  The power line is 20 ft long and heavy gauge to the panel. It wont have sag.

The caps are installed and mounted.  I have 17 MFD as measured.  The original was 13.

Fans are oiled, Contacts are cleaned,  Covers are in place. I only need to find a 6by5 tube. Its missing. I think I have one.. Not sure yet.  The 866's are gone and diodes installed in LV and HV supplys. 

Wish me luck.  It might just fire up tomorrow..

C

[ke7trp]

Ok.  I had some time off work and we finished up the Tbolt.  I need some help badly.  We are stuck.

I turned the Tbolt on with NO 4-400s. 

I get 510 volts screen...
I get -159 Bias
I get 2600 plate volts

No screen current
No Plate current

Fans on, Filiments on, regulators glowing.

All by spec and normal except the high plate volts which is from my 256 volts input.

I installed the 4-400s next.

I flipped the filiment on, PLATE OFF and the tubes lit up, I saw the screen current hit 150 MA!  I flipped the Tbolt off right away. 

I have double checked everything by the schematic.  I dont see anything out of the ordinary.

According to the manual when the plate is OFF and there is NO HV, there should be NO screen current.  It says 45 MA on CW fully loaded.   up to 5 ma at idle.

I swapped the tubes out and flipped the fil switch on and when I saw the screen current needle come off the peg, I turned it back OFF.  It is NOT the tubes.

The Screen supply is ALWAYS on with the Tbolt. It is NOT switched.  It runs off the filiment trans.  There is a single pole switch on the Tbolt so there is no way to have the screen and HV come on at the same time. 

The later Tbolt shows a 2 POLE switch and runs the screen and HV on  and OFF at the same time.

Is it normal to have high screen current with the plate off?   The bias is there...  I dont understand why this is happening.

Any thing I should check?

C

[WD5JKO]

  Clark,

   I believe Johnson got it right the second time around.

With a Tetrode, if the plate is off, and the screen is on, then the screen becomes the plate and the tube is now a triode. The safety bias might not be enough to keep the screen (now plate) current low.

Adding that double pole switch seems to be the best route, and it follows the Johnson circuit evolution.

I just looked at the Globe King 500C schematic, and there they key the screen voltage to the 4-400 along with the T/R relay, and the T/R relay is keyed off the Plate AC voltage going to the plate transformer.

I'd resist digging in with the existing configuration, as at best the tubes are at risk of failure. With no screen voltage, and no plate voltage, there is no risk taken.

Oh, with that high AC line voltage, your 4-400 filaments are likely at 5.5 VAC or higher. You might want to ponder the consequences of that, and what to do about it.

Jim
WD5JKO

[ke7trp]

Thanks for the post. 

I am only 6 volts higher then the manual lists for the input voltage.  It says 230 to 250 is required.  I measured the voltage at 5.7 volts. I think thats ok.  They prob ran a little low years ago and now run a little high.  The little use this thing will get I am sure its ok.

The screen supply thing is idiotic in this thing.  Take a look at the schematic if you get a chance. Its on BAMA.   Choose the Tbold.pdf and not the tboltlate.pdf. 

I think MORE - bias might bring the screen current down. I am going to try that tonight. If not, Then I think I will add a 110volt relay,  Wire one side of coil to the fil switch on terminal, One side to plate switch on front panel and then other to ground.

Then one set of contacts will switch the Plate primary supply.  The other will switch the Screen voltage 510 volts at a max of 45 MA should be ok on a 20 amp relay I think??? 

I cant switch the primary of the screen supply as that does the bias and the filiments also..

Very very odd....  I bet people cooked there tubes sitting there with all that screen current flowing.   

I am all set to modify but really wanted it to work like a factory tbolt.  I think you confirmed this is normal...

C

[ke7trp]

OK. I got it running.  The Pegged Screen current turned out to be normal.  When I saw this, I turned the rig off instantly.  I ran through every wire in the circuit and realized there was nothing wrong.  Poor design? Hell yes.

I flipped the power switch on and after the tubes warmed up, the screen current came down to 0.   I tried this because of what you said Jim.  The screen is the plate when there is no plate volts.. I figured the tube needs to be warm for the bias to work. Now that the new Caps in the screen supply are charged, This happens much much faster now...  Goes to 80 MA then falls as the fills come up.


So I guess the Bias cant do its job until the 4-400s come up to temp.  Scary but thats the way this ole girl runs.  Screen on all the time.

I followed the instructions and loaded up the ranger into the RES postion until I had 4 MA of grid.  Then I switched on the attenuator. I am using a 6 DB model from Z communications. Then into the tuned input mode. 

On CW, I get 800 watts out on the bird.  Plate voltage is 2500 and rock solid. NO sag.   Those big caps and the stiff 256 line helped here.

In linear mode with the tuned input and 6 DB attenuator, I loaded up the Tbolt to the 360MA the chart said to.  I got 600 watts AM carrier. Way to much.

I tried to detune the ranger, That did not seem to work so well.  I then tried to detune the grid tune on the Tbolt. 

What is the correct way???

If I detune the Grid on the tbolt, I can get the Carrier down to 200 where I want it. But then I cant get more then 300 watts pep and almost no mod on the scope.  After playing with this for an hour, I found I can get 200 carrier and 350 pep.  I can never get more ratio..  I can have a huge 500 watt carrier and then only go to 600 pep.  Or I can have a 100 watt carrier and still go only 200 pep.

I tried tuning the amp with modulation. No help. 

The ranger is full power and has a 3 to 4 times ratio of carrier to mod on the scope. Thats not possible with the Tbolt.

I might have soft 4-400 tubes... But they made 800 CW which is all I would expect.

I dont care how much the amp makes.  200 is fine.  But I need it to be linear signal.

I think I will check Screen volts under mod next.  The Reg tubes go dark when I load up hard.  Not sure if thats normal.  If I back off loading, then the regulators go bright like normal.

Any help would be great Jim.   

C

[Pete, WA2CWA]

You said: "I tried to detune the ranger, That did not seem to work so well.  I then tried to detune the grid tune on the Tbolt."

Neither way is good. You don't want to tune either one out of resonance. If you have too much drive, add more attenuation, or run the 6146 plate off the low voltage supply (the downside of this is you lose the Ranger plate meter reading). When my Ranger drives the Thunderbolt, I use the low voltage supply for the 6146 plate voltage. Also remember that the RES position is 350 ohms. Glen's 6 DB model comes in either a 50 ohm or 350 ohm model.

[ke7trp]

I have the 50 ohm version I think.. Not really sure. I got it 2 years ago and he changed the design and model number.

I misled you when I said detune the ranger. I meant load the ranger down.  I would never run it off resonance.  Sorry pete.

I did mean to detune the Grid on the Tbolt.  I was running that off peak.  I am NOT doing that now and I am using the Tuned input with the 6DB attenuator as the manual says.

I hooked up a 1K audio gen and retuned the ranger and Tbolt for max output modulated. 

The eureka moment happened when I turned the course loading up on the Tbolt.  Power jumped to 600PEP with a 300 watt carrier. Still not the correct ratio but alot better.  I then loaded the ranger down a step and tuned for modulated again.

What I found that at 175watts, I can get 100 percent modulation and a linear pattern that follows the ranger.  About 500 watts PEP.

If I go to 200 carrier, then it starts to go non linear.   Is this the power your Tbolt makes Pete?

I am going to now measure screen volts after work to see if its dropping.  I am also going to get some VR150s and put them in all four locations.  I am told if I do this, I will get 600 volts screen instead of 500.  Two of them regs are VR105s now and they are all run in series.

That might help..   Frank made a comment on his post about getting screen volts up.

Past that, I think the old girl is working.  Its disapointing that it does the power of a valiant  

It does do a solid 800 CW key down.  I guess in 1954 if you where a CW guy and had a little ranger, navigator or other PW rig and got a Tbolt you where in HOG HEAVEN


C

[Pete, WA2CWA]

I seem to remember that last time (several years ago) I measured the power output for a good, clean, and well-modulated carrier, it was somewhere between 180 and 195 watts give or take some line voltage variations.