Re: GU5B VTTC project
Dr. Drone, Sat Jun 27 2009, 11:00PM

Re: GU5B VTTC project
Proud Mary, Sun Jun 28 2009, 12:32AM

12.6V @ 25A for each filament!

Given the relatively short run times, you might consider using float-charged lead-acid batteries for the filaments to save time, trouble and money.

The averaged filament characteristic curve in the data sheet suggests that this valve was intended to have a mobile or remote operation capability by means of auxiliary 12V lead-acid batteries, as the curve tracks the whole range from 12.6V at full charge down to 3V full discharge - 0,5V per cell.

Re: GU5B VTTC project
vasil, Sun Jun 28 2009, 05:34AM

Secondary characteristics:

-11 cm diameter grey PVC pipe
-total lenght: 50 cm
-wound lenght: 48 cm
-magnet wire 0.4 mm
-1200 turns
-2 varnish layers
L = 32247.595 microH
C self = 7.7806 pF
F0 = 317.896 kHz

with a 11.77 pF toroid, the frequency drops to 200.544 kHz.

So it would make probably branched sparks according to the frequency theory (though, I think it depends also by the output electrode shape and position, I will see).

The fillament transformer was made on command here:

http://www.petra-toroid.ro/cere-oferta.html

No load voltage = 13.4 V (it will drop under load at 12.6 V)
Max curent = 25 A
Rated power = 350 W
Toroidal transformer type
Breakdown interwinding voltage tested at 4 kV
It will has some big power resistors on the primary side for pre-heating fillament.



I have to set the tube in position and put the fillament to work. I will slowly crank the voltage on the fillament transformer with a variac to see what is the output voltage under load (maibe I will need a few more turns). Pics soon.

Re: GU5B VTTC project
Proud Mary, Sun Jun 28 2009, 06:04AM


That's a very convenient frequency, Vasil.

Re: GU5B VTTC project
vasil, Mon Jun 29 2009, 06:16PM

Yesterday I was very busy....hmm...hmm



But today I've found a hour for the coil. I put the cooler on the base. The screws from the cooler will allow air input to the fan and will decrease the socket vibrations.



Preparing the tube:







Finally at its place:



The grid and fillament connections:

Re: GU5B VTTC project
vasil, Sat Jul 11 2009, 10:28AM

Made a few things today:

-put the fillament transformer in place



-made the fillament connections



-had to wait a long time to get a big resistor for pre-heating circuit (that's 56 ohms at 160 W). It is set up in the primary circuit of the fillament transformer.



-started formating the tube. I would have 12.4 V at 230 V, but in the middle of the day the mains voltage drop to 213 V. I will not modify the fillament transformer (adding extra turns), because if the mains voltage turns to normally I will overvolt the fillament.
The sound is from the cooler placed under the tube. The anode heatsink is warm after 30 minutes of run.

FORMATING THE TUBE

Hopping to put it together in a few days.

Re: GU5B VTTC project
vasil, Wed Jul 15 2009, 04:58PM

Put all the things together. It is huge, but I need some space for cooling down some components and the wood plates I am using are so small....and I dont know where to put it in my room....





Have to add some more coolers and make the electrical wiring.

Re: GU5B VTTC project
vasil, Tue Jul 21 2009, 06:32PM

Almost ready. Pre-heating, wiring, coolers. Not sparks yet, sorry.

http://www.youtube.com/watch?v=syRoYHKmwfo&feature=channel_page

A few tips about preparing the tube:
How much time do I need to heat the fillament to seal the tube, before applying HV? A few hours? 24 h? A few days?
And how much time is necessary to run at low power level?

The little cracks you are hearing are not from my old bones..hehe... they are from the wooden floor.

Re: GU5B VTTC project
StevenCaton, Wed Jul 22 2009, 12:25AM

Looks good Vasil.

I can't tell you how long you need to burn in that specific tube, but a tube such as the 833C for example, should have at least 10 hours bake time. When I built my 833C powered VTTC, I was gone one whole weekend so I just let the filament cook the whole time.

You are aware of the fact that a low Fo secondary resonator tends to cause the sparks to "branch out" right? Your spark length may suffer a little bit from a low Fo setup. Chris Hooper had this problem on Quadzilla. But hey, you already made a secondary so give it a try for sure.

Also, are you going to be building a pulser circuit with a triac, scr, etc?

Re: GU5B VTTC project
vasil, Wed Jul 22 2009, 05:50AM

Yes, I keep a little free space on the wood plate with the RC network for a staccato circuit, so it can be added later.

Re: GU5B VTTC project
vasil, Sat Jul 25 2009, 06:49PM

FIRST LIGHT

-40 primary turns
-21 feedback turns
-1.46 nF @ 18 kV primary cap
-1.36 nF grid cap
-11 nF filer cap
-5 microF doubler cap

The final setup:



Primary cap



Coils:



First light: 60 cm sparks to a grounded copper rod.

http://www.youtube.com/watch?v=wUbfZzKTpiw&feature=channel_page

I am using 2200 V output MOTs in my doubler. I will try to find some 2500 V output ones, so maibe I will have a few inches more in spark lenght.

Re: GU5B VTTC project
brtaman, Sat Jul 25 2009, 07:27PM

Amazing coil, vasil!

Those are some very nice sparks. Do you plan on adding staccato? How much more do you think you can squeeze out of that monster tube?! Gonna start searching the Internets for a GU5B. :O

Again great work, your project is really coming together.

Re: GU5B VTTC project
vasil, Sat Jul 25 2009, 07:32PM

No staccato yet, have to see how much can get out without it.
I have already a plan to modify the power source to get more juice.

Re: GU5B VTTC project
Dr. H., Sat Jul 25 2009, 09:57PM

Good day Vasil

nice looking coil. Congrads on first light. :)

Looking forward to breaking the 1m barier. With that tube you shoud be able to pull double the size of those streamers. Start tinkering with the feedback coil and the primary circuit.

Keep up the good work sir.
Cheers.

Re: GU5B VTTC project
vasil, Sun Jul 26 2009, 06:53AM

Are you sure?

For, let say 3600 W on plate, the Freau formula predicts only 0.5 * sqrt (3600) = 30 inches sparks for a well balanced system.
Maibe I can break 1 m but it would be a risk for the tube.

Re: GU5B VTTC project
Steve Conner, Sun Jul 26 2009, 10:01AM

Nice work Vasil! :)

Spark length is determined by peak power, but plate dissipation is determined by average power. So you can get longer sparks without damaging the tube, by increasing the peak power without increasing the average.

This is what the staccato circuit is for. If you set your staccato to every other cycle (25Hz PRF) you can now retune your tank circuit for twice the RF power. Or increase the HT voltage by a factor of 1.4, or whatever.

Given the size of the tube you're using, I think the power supply would burn out before the tube does! :O Just watch out for overheating of the control grid. You can't easily see it glowing because of the metal envelope and bright filament nearby, so a grid current meter would be a really good idea.

Re: GU5B VTTC project
Dr. H., Sun Jul 26 2009, 05:26PM

Good day Vasil

YES - I am absolutely sure - if you look at my thread about the double GMI-7 coil you will see that I am using two MOTs in series, and the plate dissipation rating for each of the tubes is like 180W and I am hiting the 1m (40inch) ceiling without any problems (and the system is still not tuned because have problems with flashovers.) And also have in mind - I do not use staccato (for now that is).

Cheers Dr. H.

Re: GU5B VTTC project
vasil, Sun Jul 26 2009, 05:55PM

Thanx

From datasheet I can not know a few things:

http://www.tubes.ru/techinfo/Broadcast/gu-5b.html

What is the max power dissiaption on the grid? The max grid current?
How much is the peak voltage supported by this tube? If I increase slowly the plate voltage are the eventually breakdowens fatal for the tube?
Where do I have to put the A-meter (on the grid or between RC network and negative pole? Have to use a DC A-meter or AC one? (didnt use anything like that before).

This is the actual power source schema (4 MOTs in phase and doubled):



I will modify this to get twice voltage:



I would not go above 7 kV (at 7.2 kV an 833 suffered breakdown in Freau VTTC magnifier) though for safety reasons for the tube. Thats under 140 V on variac from my calculations.

Re: GU5B VTTC project
Steve Conner, Sun Jul 26 2009, 08:30PM

Dr. H: If I remember right GMI-7 tubes are designed for extremely high pulse currents in radar modulators. They have huge oxide-coated cathodes that can deliver several amps of current. So it's not surprising that they can generate high peak RF power and colossal sparks. The peak power capability of Vasil's tube might well be less, even if the average power capability is greater.

As for all those questions, we'll need to hope that someone who reads Russian comes along. :( I do know that the grid current meter should be DC, and you insert it in series with the grid leak resistor at the grounded end.

Re: GU5B VTTC project
Wavetuner, Tue Jul 28 2009, 07:49AM

Wow, Nice tube coil!! I'll have to look up that tube's specs. My question, which is what I am strugling with, is creating a stiff power supply. What are you using?I am to the point of rewinding with added turns on he two outer coils and rewind the center core as the primary on a former 3p transformer. Usually this scenario the outer coil are wound in reverse. Will I need to cut one off and rewind one or can I wire the secondaries in reverse series. Does anyone know if this will work? Also all cores are the same size which is wierd Do I need to machine the outer cores thinner? The core is reversed E core, oh goody.

There was areally nice plate transformer on ebay awhile ago but I was too tight on cash at the time. I am not sold on MOV's though I have 4 identical 1100watt ones but am not happy with them. They just weren't designed for CW operation. We tend to forget that when we use doublers for voltage we half the current I want a 4500v or so transformer, even if I have to wind it. I would like to run it raw or CW with stacatto. I have a great core for the project. Does anyone know if the HV windings have to be wound in oposite or can they be reverse series connected?

Again, a really nice coil! Very motivating:-D

Jim Mora

Re: GU5B VTTC project
Wavetuner, Tue Jul 28 2009, 08:13AM

wow, that tube rocks! 5kv, 3.5kw!! I wonder what it will do with a stacato reduced duty cycle!

Re: GU5B VTTC project
vasil, Tue Jul 28 2009, 01:26PM

I am using a strange configurations of 4 MOTs, with different outputs, in a level shifter, here is its construction thread:

http://4hv.org/e107_plugins/forum/forum_viewtopic.php?60496

I had to use what I have in my hands. Probably, not the best solution, but this is it. Now, I will re-wire it to get twice output voltage, at the same power level (I hope).

I will build a staccato before trying higher voltages. It would pity to distroy such nice tube.

Wind your secondaries in reverse and attach the inner output to the core (ground the core). That would reduce the electric stress.

Re: GU5B VTTC project
vasil, Sun Aug 02 2009, 05:58PM

Got the A-meter (0-600 mA, I hope that is enough). Working on staccato.

Re: GU5B VTTC project
vasil, Sat Aug 22 2009, 08:09PM

Well, some good news.

Added a staccato, that's the low voltage side:



I made myself a lighter heatsink from an aluminium sheet:



I am very clumsy when soldering (had burned a lot of transistors along the time), so I put them in chip sockets:



Added 2 new MOTs to my power source (that's 6 MOTs in serie-parallel).

Results:

MOVIE 1

Retuning for a lower frequency on primary tank gives up to 80 cm:

MOVIE 2

I think I reached my power limit. Even at this low rate, the power cord is hot after 1 min of run and I have epuised my MOT stock. But I think that Dr H has right, the tube can take more abuse.

Enjoy

Re: GU5B VTTC project
Dr. H., Sat Aug 22 2009, 10:06PM

Good day Vasil

I am glad you are still working on the coil. The results are looking great ... and I still think you will be able to squize a bit more out of it. Without a staccato I was drawing around 28-30Amps (atleast that is what the analog A meter showed) at 240V to produce 100+ sm of sparks. Have in mind that I was using only TWO MOTs ... this makes me think that you have to tinker with the tunning/tickler coil, try different secondaries and toploads. Also play with the grid feedback circuit. I've noticed a dramatic increase/decrease in spark lenght only by moving the tickler coil a couple of sm up or down. Another important factor - the level shifter capacitor - placing to big of a cap on mine decreased the spark lenght dramatically. I've had a 3.5uF (or something like that) in the beginning and I've ended with using only 1uF for best results. Sooo keeep experimenting.

Cheers

Re: GU5B VTTC project
vasil, Sun Aug 23 2009, 05:34AM

OK, I will do it

I liked the idea to have the feedback tapable, thats why I made it with heavy wire and taps. I will make a new one with magnet wire (20-21 turns) and place it above the primary. Still afraid to think what can happen in a breakdown.

The doubler cap is now 2.6 microF. Not too much..but I can experiment with different values.

Do you think that a different secondary would work? Making secondaries for 300 kHz, 400 kHz (increments of 100 kHz) would be a good idea? I am too lazy to winding though.

I will rebuild the source. Indeed with more MOTs the sparks is hoter not longer.

I tryed a bigger toroid but didnt notice semnificative better results (only more fiercely and hoter sparks).

In the precedent VTTC I got a few cm of sparks just lowering the primary cap at only 700 pF from 1.1 nF.

I will experiment a bit.

Re: GU5B VTTC project
Dr. H., Sun Aug 23 2009, 06:22AM

Good day Vasil

I am not shure what frequency the tube "preffer" to be at - I've got the best results at around 360khz. From experiments - tubes seem to like the 300-400khz range. Also noticed that my coil didn't like big torroids - it ran much happyer with smaller ones. I've had tube arcing and other problems with big torroids.

My secondary is about 70sm tall wound with some very thick wire. (don't ask about size don't remember )

I think you shoud up the Cprimary and lower the turns. I would go with 1.1nF or even a bit bigger and tune it with the primary (watch the tube ! ) Decreasing the cap value and with soo much and getting bigger sparks suggests you are out of tune ! A lot.

On the feedback coil - make it "movable" tunning this coil is critical ! But take your time making it properly because it will be a great "ground" target for the sparks (this is where I have problems). In my experiments changing the number of turns didn't changed a lot but playing with the feedback circuit *(the C and the R ) made a GREAT difference.

Cheers

Re: GU5B VTTC project
teravolt, Sun Aug 23 2009, 06:43AM

DR. Spark got better results in the end by using about 10 turns of 2 awg wire on the primary. Lower inductance higher capacitance makes for larger circulating current. Remeber that 3.5kw is the power disapation of the tube in heat. If the tesla is 75% ephishant as in class C then the other 25% is heat mainly in the tube. With about 5 mots you should put out as much as Quadzila. Please feel free to correct me if I am wrong

Re: GU5B VTTC project
Wavetuner, Sun Aug 23 2009, 09:21AM

Wow is #2awg a typo? I plan to use .1865" soft copper tubing seperated by equal size heavy fishing line. Maybe I should scale that up? This is also a quad 833C.

Re: GU5B VTTC project
vasil, Sun Aug 23 2009, 07:29PM

I think it is a missunderstanding. I got the best results on my anterior VTTC (with GU81 tube) by reducing the primary cap from 2 nF at 700 pF.

The GU5B VTTC is in tune and is using a 1.35 nF cap.

I added 500 pF more on the primary cap in my experiments and didn't notice longer sparks, but hoter, with important warming of the primary windings.

Have to figure a feedback design to protect it from the arcs. Maibe a strike ring above the feedback would protect it a bit.

Re: GU5B VTTC project
Dr. H., Sun Aug 23 2009, 09:29PM

Hi there

on the cap - YES you have to experiment but changing the cap value will mean that you have to retune the primary. And YES - there are systems in wich smaller capacitors work better than big ones.

I don't think putting a strike ring is a particularly a good idea because grounding it will create a perfect target for the sparks (will "attract them"). I've withnessed how a spark missed two strike rings and went directly for the primary (on a DRSSTC) 3-4sm bellow the strike ring, so I woudn't use one on a tickler coil. A better idea are the "Hooper" rings. Look at Dr. Spark's work - the re are some great ideas on it how to make it properly. One of the most importan things - use insulated wire ( not Enamelled Magent Wire). And after that cover it with epoxy and put the rings.

Cheers

Re: GU5B VTTC project
teravolt, Mon Aug 24 2009, 02:16AM

I have seen a youtube video "big vttc" and he has the tickler on the bottom

Re: GU5B VTTC project
vasil, Mon Aug 24 2009, 05:09AM

Mine also is on the bottom.
I was thinking about the things to do next time:

-rebuilding the source
-increasing the primary cap to 2 nF
-winding a new primary with lower AWG wire (decrease primary resistence)
-a new feedback

It will take some time to do it, so have patience.

Re: GU5B VTTC project
vasil, Sat Oct 24 2009, 04:46PM

YYupii, a new little vacance started.

Shortly:

-redone the primary with heavier solid copper wire, less coupling:



-increased the primary cap at 2.96 nF (I used all the caps found):



-rebuilt the HV source with new 4 identical output MOTs and a new 50 nF filter cap:



-the system is ready to fire. Added later a bigger toroid:



At the first try I had some RF interference problems with staccato. Though the sparks are longer 94-95 cm with rare hits at 102-103 cm:

http://www.youtube.com/user/teslina#p/a/u/1/jPZDqgVUosM

Resolved the staccato problems (alu foil and ferrite beds on the input and output wires), so now GU5B-zilla can make its big sparks:

http://www.youtube.com/user/teslina#p/a/u/0/jv6Syp-BlEg

Unfortunately I didnt found higher voltage MOTs. I am confident that with some new modifications the sparks can be even longer but I stop here. These russian tubes are really wonderful.












My next project will be a classical spark gap coil. I bought from the internet these nice NSTs (7 kV@25mA each):



Believe or not, they are the first NSTs I have in my hands since I begun coiling. The coil will have a spherical topload in an attempt to aquire spontanously formation of a swirling streamer, like in one of my old movies:

http://www.youtube.com/user/teslina#p/u/56/nj0ARi5JTgc

...but at higher power.

Enjoy

vasil

Re: GU5B VTTC project
Mads Barnkob, Mon Oct 26 2009, 09:20PM

Good looking sparks and wonderful that you are hitting 100 cm+

This makes me want to get going with my next VTTC sooner than planned :)

Got a nice 5kV 3kW plate dissipation tetrode tube at hand, filament transformer, ceramic caps for tank and lots of mots, but would like a power supply that is less power hungry and smaller :)

Re: GU5B VTTC project
darkspeed, Sat Jul 03 2010, 02:46AM

Vasil do you have a schematic for the GU5B circuit? I would really like to build a GU5B VTTC

Thank you!
darkspeed

Re: GU5B VTTC project
vasil, Sat Jul 03 2010, 09:33AM

Hi
I dont have a schema on paper but I will give you some details:

Power source:

2 strings in parallel of 2 MOTs in serie (for 4200 V @ 1 A)
doubler made with a 44 pieces of 1N5408 diodes (44 kV @ 3 A) and a 2.3 microF cap
filter cap: 50 nF @ 20 kV

fillament transformer: a big 350 W toroidal xfrm (I added a few turns to get the right voltage under load)
5.4 nF cap filter on fillament connections
big 56 ohms @ 160 W resistor in the primary circuit of the fillament xfrm to feed the fillament gradually, in two steps.
Rgrid: 150 W light bulb
Cgrid: 1.3 nF @ 10 kV polietilene cap
Cprimary: 2.96 nF @ 18 kV (made with different size ceramic russian caps)
20 cm diameter PVC pipe for primary, 34 turns
grid coil 20 turns, under the primary coil
11 cm diameter PVC pipe for secondary, 1200 turns with 0.4 mm wire
little toroid on top made with a wheelbarrow tire
Steve Ward staccato circuit

The schema is just usual VTTC schema.

Happy coiling!

vasil


-------------------------------- -------------------------------------------------- ------------------

Oct 16 2020

Hi forum

I made a few grid network modifications preparing the setup for full CW (I collecting parts for a new power source). I am runing now with a 0.56 Kohm high power grid resistor (made with different size high watage resistor).

Grid setup:



Whole system ready to fire:



Over 80 cm hot streamers:



and a MOVIE

I hope to finish the new source maibe in a week (just currious how it behaves in CW).

Enjoy

vasil
>
>

Oct 18 2010

The new source is ready.

I was lucky enough to get these nice HV diodes (75 kV @ 100 mA):



The new PSU is a Greinacher cascade. The voltage from the 4 MOTs is doubled. The cascade capacitors and the filter caps reduce ripple. I used ferite beads as RF suppressors on the output HV cables:



The pics of the PSU:









and the final results: HOT discharges

YOUTUBE LINK

After playing with the new toy, my little variac was a bit toasted:



....but I will find another one.


So, an important thing for the VTTCs is the HV power source. They need a dedicated HV source. More power = better spark. With a weak power source (1-2 MOTs), the coil would need higher current than provided by the HV setup, so the MOTs will run hot. Just give enough current to the tube, it will take what it needs. My experiments just illustrate the old tesla concept that the lenght of the spark is proportional with the power input of the tesla coil. And this principle is very importanty, especially for VTTCs with big toobs and high catode current.

Enjoy

vasil

Re: GU5B VTTC project
darkspeed, Sat Jul 03 2010, 08:00PM

Thank you vasil!!