EEVblog Electronics Community Forum
Electronics => RF, Microwave, Ham Radio => Topic started by: NiaDebesis on May 21, 2020, 05:45:40 pm
-
Hi guys,
at my company were dismissing an old rf amplifier, the instruments during the years has bene cannibalized, missing componente etc, but one of the fewer things that were still inside the instruments were a TWT and some Transformers.
So I took out the transformers and the TWT.
The TWT is a CPI VTW 6195R1 with dual coaxial connectors.
I'd like to test if It works, but I would like to know if there are some preferencies on the supply signals of the helix etc.
And how to figure the pin out of the TWT.
-
They run HV regulated supplies, i noticed high frequency tubes mean higher voltages. None of the TWT amps I took apart have a unregulated supply. Really old ones have tube supplies and the newer ones have rubber cubes that are HV supplies.
So you might want to take more if you can.
They follow a procedure where there is a single HV rail and you use potentiometers to adjust it. I think my 8GHz one was <1200V.
I have a old tube too that I don't really know the pin out for but IIRC the color scheme was similar to the ones in the HP units, so I thought I could at least figure that out and slowly ramp it up, but I don't want to work on that until I get the right adjustable HV supplies so I can properly test it. The adjustment procedure for biasing them is fairly complicated too, in the HP units alot of the pots were chained together so you had to follow the steps in the right order to determine if the tube could reach the point described by the manual with the adjustment ranges provided (which meant the tube was good so long your pots and rails were in spec, I think).
So, the first thing you need, is a high voltage potentiometer adjustment tool for your own safety when you work on it. And HV DMM.
If you had a rack of HV supplies it would be much easier of course. I would be totally fucked on that circuit without a manual. But, the nice thing is, once you work out a old vacuum tube unit you know how the modern one is built. So look in HP manuals
2 out of 3 units I got worked, when I combined all the guts together during trouble shooting (same tubes used between units), and one of them had smoke come out of the transformer (it kept blowing a fuse and I wondered if I could finish the test by raising the fuse a little). It did run for like 1 more hours though. the 1W ones are not that impressive for most things because cheap MMIC can do similar on wider bandwidth actually, so I was rough with it. One of the fails I found was a shorted 980Kohm resistor (imagine that, charred 1 Meg.. not the usual business for me! Insane voltages eh, if you can burn a 1meg resistor thats rated for like 1/2 watt?)
-
I am thinking to just take it work and try It on the HV bench to check first if it's working.
Ill' try to scavange the main board.
Also, It was not from an hp unit, Is from a CPI unit, that can output more or less 28W at 18GHz
-
The only thing Is that I didn't find the HV transformer, I have only found a signal transformer and what I thought to be the HV transformer, but was a EI core with three wires, so I think Is a big inductor
-
thats a beast
is it a wideband?
-
that sounds like a autoformer, the ballasts usually have two wires
-
Yes, Is a wideband.
Also yea, think Is a autotransformer
This TWT Is the same as the One used in the
IFI T184-20 that I also have to work with.
-
http://lea.hamradio.si/~s57uuu/emeconf/eme96.htm (http://lea.hamradio.si/~s57uuu/emeconf/eme96.htm)
this one has loads of useful info from manufacturers guidelines research and book research by the author. from the sounds of things, looks like that guy put alot of thought into it, and that its a big project.
https://www.cpii.com/docs/related/19/OIHPPulsedTWT_Iss.pdf (https://www.cpii.com/docs/related/19/OIHPPulsedTWT_Iss.pdf)
This one has wiring colors, maybe they are the same... you can defiantly probe for heaters easily
I assume RF neg is tied to ground, which is tied to metal tube chassis, so you just have two left, if you have a DC heater and you saw the circuit before you got the tube, you might know which heater is DC - and which one is + and tied to plate
-
http://www.pa0ply.nl/varian_twt.htm (http://www.pa0ply.nl/varian_twt.htm)
I have found the schematic of the amplifier, unfortunally there aren't the schematic for the input 220V , I have hook up the autotransformer, the relay and a signal transformer.
I was looking up to the First link, and, I don't know if it can be adapted to my twt.
Also, my twt has not colored wires, they are all White.
I have already found the filament, but I don't know the polarity.
I have also scavanged all the pcbs, but I don't know how to hook up the input connections.
-
I am looking for information on such a TWT connector pinout.
-
that is a cool looking connector
-
http://www.pa0ply.nl/varian_twt.htm (http://www.pa0ply.nl/varian_twt.htm)
I have found the schematic of the amplifier, unfortunally there aren't the schematic for the input 220V , I have hook up the autotransformer, the relay and a signal transformer.
I was looking up to the First link, and, I don't know if it can be adapted to my twt.
Also, my twt has not colored wires, they are all White.
I have already found the filament, but I don't know the polarity.
I have also scavanged all the pcbs, but I don't know how to hook up the input connections.
you could still follow the advice he gives on sequencing and stuff (guidelines), I think it might be right. I know its a pulsed tube, but there is still advice there on heaters, timing, etc
but I don't know much about the structural differences
-
Do you think that if I simply apply the spec voltages to the tube It will work normally?
-
if you sequence it correctly
-
I have found the pin out of the TWT thanks to the Old pcbs.
The tube says on the label
Ew= 3950Vdc
Eb= 1856Vdc
Ed= 6.3V
If I hook up the HV psus on the tube with this value (with a commong gnd, or same PSU but with a voltages divider in single Raul) and work in its CW range (20w) It should work right?
I Just don't t understand what eb and ew means
-
the first link I posted explains it, you can infer from his writing
Eb = collector
Ew = helix,
It says the collector is approximately 1/2 the helix
If this is accurate IDK, I am afraid you are going to fry it, but I seem to recall something being half of something when I set mine up, and the other thing you adjust and monitor the current to set it correctly. Those amps when they are setup right usually have a dial on the front that tells you the current of something, so you set the voltages to some predefined values then adjust for nominal current. I don't feel like reading the manual right now to remember. the front panel indicator tells you if its over loading some how by measuring some kind of current (forgot which one), but you use it for the 'tuning' process when you adjust potentiometers.
He seems to say the grid is rise time sensitive, and my unit featured a vacuum relay for triggering something, possibly the grid current (its like a mosfet you don't want it in linear region for too long). Vacuum relay gives a very very sharp pulse. You need to verify if the vacuum relay DOES infact give sharper pulse, I am not sure, I think they use oil relays for high power pulse shapers, I think vacuum is next best. ?
Control grid (sometimes called G2 or anode) supply. This voltage determines the beam current, and is usually derived from the helix voltage. It must be adjustable. Current consumption is in the micro ampere range. In most cases, this grid is also used for switching the beam on and off. Because this voltage has a role in beam focusing, it must be switched on and off very fast. While it is rising, the helix current can assume values several times the allowed absolute maximum, and this transient must be kept as short as possible. G2 voltage should not be present in absence of helix voltage, because excessive current would flow and destroy G2. Since in most cases G2 voltage is derived from the helix voltage, this is no problem.
I want to say the final step is adjusting the control grid current, after it switches on, to the 'nominal' value on the dial indicator. I think so long it turns on to some kind of ballpark, you can adjust it so it settles in the nominal place, then play around with the front panel for specifics IIRC, but you need to adjust it so that no matter where your knob is set, it is in a safe spot. IDK how wear is effected by this, I assume there are nominal settings.
-
eB can stand for beam control?
eW is mysterious
eD makes me think of deflection, but its the damn heater
who the hell knows, but the voltages make sense that its heater, helix and collector
And its european, so it could be another language, since they use E instead of V
-
Sorry, It Is Ef, autocorrector.
And it doesn't have a wire for the grid.
Also, from what I understand, since the cables from the tube are 4
One i the commo gnd and the others Collector, helix and filament, and you Say that Is impossible to power the tube directly from an HV PSU?
Because Ill' fry It? Even if I turn on the helix and collector in correct order?
If I give the correct voltages why It will fry?
Controlling also the current.
I thought using this setup
Twta hook up to the three PSU, One for the filament, one for the collector and the other for the helix, if I don't have three hv PSU I use One with a voltages divider and a pot to control the current.
Use a rf generatore in input at -30dbm and booking up the out out to a Power Meter with a security attenuator.
I thought It could function
-
you need to leave it hooked up to the load before you start, some amps actually have a function that prevents them from being powered unless there is a load connected (it uses a imcroswitch to detect a N connector)
No, I think your setup should work if you do it correctly, with multiple supplies. but i am not sure
this diagram shows independent supplies, idont think they need to track, but it could be a feature possibly if you wanted to make it so
https://lh3.googleusercontent.com/proxy/QfEbqLuuCSN12lfmz4eJZ4Zi9TGvdbr8Q-58uLK-e_skMRjRnvFTP0sWn1ewUyd9dxqtAf66czC-MmaVYmxi5djV
-
More or less I was thinking to do this like in the diagram. The only differences are the voltages and the consumptions.
The load will be costantly attached to the output.
Also when not used in the circuit that ill design ill add an rf switch with a dummy load.
And what do you mean with (track) I am not a native speaker so I don't understand, Sorry.
-
Also, Is common that my twt doesn't have a grid / anode wire?
-
The cathode Is in common with the Heather, the anode Is connected to the helix, but no grid.
But I think Is not essentally?
Also, the chassis of the tube, has ti be in common with the gnd of the power supplys?
-
And what do you mean with (track) I am not a native speaker so I don't understand, Sorry.
Multiple power supplies can either be tracking or independent.
When tracking is used, one power supply is the master and is set directly, and the others are set as proportional to the master voltage. So if you vary the master voltage all the others stay the same in proportion to it. This can be useful if you need multiple voltages to the same circuit or component, like a tube or IGBT.
-
I don't know how there is no grid, maybe there is a divider in there to get grid proportional to something ?
-
I have tested with a dmm all the connections, only the filament has resistance, the rest don't show nothing
-
I think the grid is used for gain control and AM modulation.
Look like the T184-20 relies on a preamplifier rather then actual control on the helix tube. More modern I guess.
-
Hi,
Look at this diagram:
http://www.pa0ply.nl/articles/Varian/highvoltage.jpg (http://www.pa0ply.nl/articles/Varian/highvoltage.jpg)
Remember that body is at ground potential.
The cathode and the filament are at -2kV to -4kV with respect to ground.
The power supplies must be sequenced up and sequenced down.
I believe that the Helix supply has to be very low ripple or it will modulate the RF output.
It is not easy to make a power supply for a TWT.
Regards,
Jay_Diddy_B
-
Hi,
I have this exact PCB, since It was scavanged from the instrument.
But i can't figure how to turn It on, since this PCB look the same, but the others (voltages section an Logic) are from a different revision
-
The squence should be
Filament, soft start, 5 Min to warm It up
Collector voltage
Catode voltage
Helix voltage
Gnd Is the body that Is also shorted with the helix 1ohm resistance
This Is what I found on this TWT, using the tp points in the schematic that you sent to me
-
3000 v to gnd by 1ohm??
sounds a little broken. I think that tube had a flash over and died.
How could, 3000V+, be connected to chassis??
please be more clear, this is starting to sound like it can kill you.
Do you mean the helix supply, or the actual helix itself? (as in, the work coil), which I would say RFin to GND and RFout to GND since all the literature calls the helix the plate.
So you are calling the helix voltage the cathode voltage, the helix the helix, ? so its cathode supply now?
so anode, cathode, actual helix assembly. fucking boiler plate
-
It's like in the link that you sent to me
https://lh3.googleusercontent.com/proxy/QfEbqLuuCSN12lfmz4eJZ4Zi9TGvdbr8Q-58uLK-e_skMRjRnvFTP0sWn1ewUyd9dxqtAf66czC-MmaVYmxi5djV
The helix supply Is in common with the chassis
-
https://www.cpii.com/docs/related/19/OIMPCWTWT_Iss.pdf (https://www.cpii.com/docs/related/19/OIMPCWTWT_Iss.pdf)
Here also, the helix supply Is in common with the chassis
-
you mean the cathode supply positive is connected to ground by 1 ohm, since the cathode connection voltage is negative?
is it the depressed or non depressed collector?
-
Yes
-
I am now starting to create the HV transformer.
I have found a massive E E core transformer.
It has 2 x 4 turrets and is 12cm long.
Don't have the datasheet of the core, but I know Is used to make actually hv Transformers.
The doubt Is, Is better to wound the coils on the opposite sides of the transformer, or to make a single coil in the middle with primary secondary and primary configuration, with the primarys in series, a sandwich coil.
-
There was a switch mode power supply in DUBUS by CT1DMK.
IIRC it was presented at the EME meeting in Paris in 1998.
I have not tried it, but CT1DMK things are said to work.
http://www.dubus.org/ (http://www.dubus.org/)
https://www.qsl.net/ct1dmk/tw_twt1.pdf (https://www.qsl.net/ct1dmk/tw_twt1.pdf)
Cheers, Gerhard