My Î“K-71 (GK-71) HFVTTC / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

My Î“K-71 (GK-71) HFVTTC

1 2 next

Move Thread

LAN_403

PhilGood

Sat Jul 21 2012, 09:12PM

Registered Member #3806 Joined: Sat Apr 02 2011, 09:20PM
Location: France
Posts: 259

GK-71 HFVTTC Project

Hey all,

This is my project thread for a High Frequency Vacuum Tube Tesla Coil using the russian GK-71 power pentode. It is based on RateReducer's Coil. I will try to build the same circuit but in a neater manner for a better look and longer runtimes.

Description :

Frequency is around 20-30MHz
Anode, feedback and ground connection are made of 10mm x 1mm section flat copper foil for higher current
Resonator and choke coils are wound on the same coilform with Ã˜2mm enameled copper wire
Coilform is a Ã˜50 ceramic tube to withstand the heat
I found nice chinese ceramic sockets for the GK-71 on Ebay, they look way better than russian bakelite ones.
HV power supply is an arrangement of two MOT's with voltage doublers connected in parallel, and primaries in series 180Â° out of phase for full wave rectification and 230V operation. Thanks Steve Conner for the suggestion
EMI/RFI filter + Variac on mains side
I expect a 30cm high silent flame output
GK-71 datasheet

I'm awaiting for the following parts :

Ceramic tube socket - received on 27/07
1mm copper foil - received on 24/07
Filament transformer (20V-100VA)
500W Wirewound Power Resistor for the ceramic coilform - received on 01/08

Question, comments and suggestions are welcomed

Schematic :

GK-71 Tube :

HV Power Supply parts :

Changelog

2012/23/07 For the HV Power Supply, I might use a single MOT with full wave rectification if I manage to properly insulate secondary, discussed in this thread :
2012/24/07 "glitch resistor" added in series with HV Power supply as suggested by Steve here

Related threads :

Weston

Sun Jul 22 2012, 01:13AM

Registered Member #1316 Joined: Thu Feb 14 2008, 03:35AM
Location: Cambridge, MA
Posts: 365

Will the DC blocking capacitor on the grid connection marked "optional" to be used? Without it any flashover on the feedback electrode has a high chance of ruining the tube.

Also, a DC blocking capacitor between the anode of the tube and the resonator. Otherwise, any ground strike will also have a chance of shorting out the power supply to ground and poses an even larger electrocution risk.

Aside from that, great documentation. Keep it up. Such coils are fun to play with.

PhilGood

Sun Jul 22 2012, 04:07AM

Registered Member #3806 Joined: Sat Apr 02 2011, 09:20PM
Location: France
Posts: 259

Thanks Weston for your kind help and encouragement

Well yes I will put the "optional" capacitor in the feedback circuit, doesn't hurts :)

About the DC blocking cap between plate and resonator, I understand it would be safer. But wouldn't that reduce the coil performance ? I mean I haven't seen any of those HF coils designed for big flame output have such a capacitor. Wouldn't it require a special capacitor with high power at high frequency capabilities ? Also it seems these coils do not tend to arc to ground, probly because of the plasma heat that keeps the flame growing upwards.

Hmm... I have two big 300pF 15kV ceramic capacitors rated for 30 kVA (kvar ?) Would it fit for this ?

Weston

Sun Jul 22 2012, 06:09AM

Registered Member #1316 Joined: Thu Feb 14 2008, 03:35AM
Location: Cambridge, MA
Posts: 365

The DC blocking capacitor should have a minor impact on performance. The capacitance should have a very low reactance at operational frequency, and at that point it is only presenting parasitics. Both the ESL and ESR of doorknob caps are very low and you should not even need ones with a KVAR rating.

Just on a side note, I would trust surplus Russian doorknob caps more than the Chinese ones that you have, and they are usually cheaper too. However, the ones you have should do fine on in your application.

Do you plan on trying to add capacitance between anode and ground to optimize the coil? It forms part of the impedance matching circuit and can help achieve proper loading on the tube.

PhilGood

Sun Jul 22 2012, 07:46AM

Registered Member #3806 Joined: Sat Apr 02 2011, 09:20PM
Location: France
Posts: 259

Ok lol, I thought the chinese cap would be better because it's a lot bigger (maybe because it looks nice too)

Well I also have some of these russian surplus caps, KVI-3 and K15Y-1 series, 470pF, 680pF and 1nF. (The K15Y-1's are indeed rated 40 KVAR!)

How do you calculate the optimal value of this cap you suggest to add between anode and ground ?

Sorry if this seems noob question, but tbh I'm new to this great hobby (this will be my 2nd coil), and for now I'm rather trying to reproduce others great coil as close as possible and get same results.

So I will first try to get the same (or close) output as RateReducer (40cm high silent flame) with the same build. (I will just add the feedback circuit capacitor).

Once I get this done, I will try to add all suggested improvments, one by one, in order to see how they each affect the output.

I found a russian schematic for a GU-50 that shows two additional caps :

- the left one is probly just an RF bypass cap, but is it necessary as we have the choke to block RF ?
- a 100nF between G2 and ground, what is it's purpose ?

Oh btw, I have a weird doubt about my HV Power Supply circuit, but I posted it as a separate thread here:

Thanks again for your kind help :)

PS: I could easily find all the needed documentations and detailed explanations about SGTC's, VTTC's, SSTC's, DRSSTC's operation. But I couldn't find any complete and detailed explanation of this base fed HF coil circuit... I had to gather little parts from here and there, mostly from russians and germans sites, and try to put them all together. If anyone has some good link explaining in details this circuit theory and operation it would be greatly appreciated ! Thanks

zrg

Sun Jul 22 2012, 04:41PM

Registered Member #4762 Joined: Sun May 06 2012, 05:59PM
Location: Russia
Posts: 93

The KBÐ˜ cap is 100pf. Protects the grid if the flame hits the feedback capacitor. I would also recommend using several small power ceramic caps like K15Y of value 3-10 pF as a feedback variable capacitor instead of copper plates that are usually used. With them the 100 pF ÐšÐ’Ð˜ can be omitted.
That schematic works really well, but don't expect much more than 15-20 cm flame from GK-71. That is a good 1000+W already while the tube is 200-400W rated only, and 30-40 cm are 4+kW.

PhilGood

Sun Jul 22 2012, 10:42PM

Registered Member #3806 Joined: Sat Apr 02 2011, 09:20PM
Location: France
Posts: 259

I ordered 1mm copper sheet for this project, so there should be no problem.
But this string of small caps as feedback connection might be interesting on an aesthetic point of view !

About flame height, have a look at this video :

zrg

Thu Jul 26 2012, 02:15AM

Registered Member #4762 Joined: Sun May 06 2012, 05:59PM
Location: Russia
Posts: 93

That is much more than a tube can carry on long runs. Also the flame is enhanced with a salt which gives +50-100% of length ;) but surely you can try to tune it better, it you believe that it is possible.

radhoo

Thu Jul 26 2012, 09:59AM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 699

GK71 has some great potential, so good look with this project, looking forward to your first results!

PhilGood

Wed Aug 01 2012, 08:32AM

Registered Member #3806 Joined: Sat Apr 02 2011, 09:20PM
Location: France
Posts: 259

@zrg: I don't think salt increases that much flame height, at least it didn't at all on my 572B VTTC, but it was horizontal sparks and not vertical flame, so I'm not sure about this.
I know I won't be able to run the tube at full power for long times, I forgot to say in my project description it will be powered thru a variac, allowing longer runs at lower power.

@radhoo: Thanks

____________

I just received my ceramic coilform

(I noticed there are two resistive wires wound in opposite direction, probly to cancel inductance)

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob