15th Dec 2016, 2:36 pm | #61 |
Tetrode
Join Date: Aug 2016
Location: London, UK.
Posts: 58
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
I don't know if this is of any use, but this site has a lot of info about how Tesla coils work and the maths involved:
http://www.richieburnett.co.uk/operation.html http://www.richieburnett.co.uk/operatn2.html#quenching
__________________
Lucia |
15th Dec 2016, 3:36 pm | #62 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Thank you Lucia, yes , Richie had been around for years and was something of a pioneer, very much focussed on DRSTCC's (Double Resonant Solid State Tesla Coils. ) I very much relied on his experience when I built my first one of these operating in this mode. The manner of operation is very different to a valve Tesla coil as the primary is hard switched by a MOSFET half bridge , typically . Very different challenges and obstacles arise , but flashover in the tank is much less likely.
__________________
Al |
15th Dec 2016, 11:03 pm | #63 |
Heptode
Join Date: Jan 2016
Location: Wolverhampton, West Midlands, UK.
Posts: 777
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Thanks for the links Lucia, some interesting bedtime reading for sure!
Chin up Al, this is far beyond my capabilities I'm afraid, but if all else fails, go back to the textbooks to refresh your memory on the theory and the actual calculations you have already gone through, I know I've misread my own notes more than once, I have also felt like a prize cabbage having been informed by those better versed than me that I had missed a stage of calculation! What I mean to say is go over your notes with a fine toothed comb or start again and compare each stage, I am sure you can find out what's up, and judging from those links above, it could be one little discrepancy and no one could blame you! Also, try not to put extra pressure on yourself with a deadline of New Year's Day, that rarely if ever "concentrates the mind" as some teachers and Bond Villains would have it. That's my advice Old Bean, take a breath, have a nice meal with the Missus, and go back to the notes fresh tomorrow. |
16th Dec 2016, 3:52 pm | #64 | |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Quote:
It'S much more a question of rethinking the actual physical placement of components in relation to eachother, plus maybe adding insulation or an alternative conduction path for violent oveevoltage. I was hoping to do this by the solstice but likely my replacement EHT diodes won't be in time. We'll just have to see. Thank you for your supportive comments, though! Update as soon as I have one ....
__________________
Al |
|
16th Dec 2016, 5:12 pm | #65 |
Dekatron
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,195
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Hi Al, strategically placed spark gaps may work.
I may have some bits from an old diathermy machine is you are interested. Ed |
16th Dec 2016, 6:26 pm | #66 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,518
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Al,
I think your power supply failure has little do do with the back EMF of your Tesla coil, and more to do with correct selection of components. Your microwave oven transformers paired up will generate more power than the rectifier diode (from a microwave oven) will reliably cope with rectifying - it is quite likely that any arc generated simply acted as a short circuit, which then dumped all of the charge from your 4uf of reservior, and then allowed the rectifier to work into a short circuit, probably overloading the rectifier by an order of magnitude. Because the diode junction was not expecting to pass the S/C current, and as such got a bit hot, the resulting expansion blew the diode apart, with the resulting internal diode parts drawing an arc in the process, thus exaggerating the destruction. This is why for big HV supplies Microwave oven diodes are avoided like the plague, they really are not rated for the sort of abuse that we amateur experimenters are throwing at them - I tend to use lots of 1N5408 diodes in strings of 5 as a minimum giving a 5KV PIV - this is for any EHT supply up to 2KV DC - anything bigger and I use some more exotic stuff - used to get some from Surplus Sales of Nebraska, but they do work out flipping expensive! I think you would be best to invest a decent sum on some heavier rectifiers, to increase reliability, sadly, this will mean the next thing in the chain will fail - HV fusing is not a trivial subject, but probably well worth reading up on. Cheers Sean
__________________
Engineers make things work and have spare bits when finished |
16th Dec 2016, 6:52 pm | #67 | |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Quote:
Thank you! That sounds like a reasonable and actually quite comforting explanation. I have now got hold of 15 X BY255, 1300 PIV rated each and 3A. I hope 3A is enough, although again, in a S/C it won't be. That destracting speculation about what happened in a flashover was partly to see just how much current those bit MOTs can sink. I think it is a frightening amount in a short circuit. I am also now keen to test the big old triode-strapped pentode before I rebuild the power supply. I wasn't able to pass a single mA of cathode (centre-tap) current earlier, when I was testing the grid-leak circuit again. What's the easiest way to test a big transmitting triode like this?
__________________
Al |
|
16th Dec 2016, 7:03 pm | #68 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,518
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
I normally just do DC tests - in the case of a Triode, pop around 50% rated Anode voltage on, and tie the cathode to ground (via a low value resistor to measure voltage across), apply a suitable variable -ve voltage to the grid to stop the whole thing melting down, then adjust grid volts while monitoring the voltage across the Cathode resistor - a bit of ohms law will then let you see how much current the valve passes, and gives an indication of function.
Quite what it does at RF is still a guess, but it is a quantifable test in it's own right.
__________________
Engineers make things work and have spare bits when finished |
16th Dec 2016, 7:20 pm | #69 | |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Quote:
Thank you, Sean, that's easy enough. Cheers! BTW, the first flashover occurred in the tank circuit - I had two copper tape straps coming off the main tank capacitor. They were 3cm apart at the point of the flashover, and it also passed through the outer conductor of the capacitor and through a 3mm lexan casing! The flashover thinned out one of the copper foil strips to the point of near destruction. That's where I get more than 30kV from. The EHT diode exploded shortly afterwards. (A few moments) - your reasoning still stands precisely...
__________________
Al |
|
18th Dec 2016, 10:51 pm | #70 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
As I'm now waiting for new components to upgrade the EHT power supply, I will be stopping work on this project until the New Year.
Here is the beauty in her Christmas finery. My wonderful partner trained as a research scientist and so she is not only tolerant of this sight but accepts it as totally acceptable part of our Christmas 🎄 surroundings. Back on this thread next year when I have more to add! Thanks for following!
__________________
Al |
18th Dec 2016, 11:13 pm | #71 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,902
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Most other people do have slightly less dramatic Christmas trees...
David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
18th Dec 2016, 11:36 pm | #72 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
As I'm now waiting for new components to upgrade the EHT power supply, I will be stopping work on this project until the New Year.
Here is the beauty in her Christmas finery. My wonderful partner trained as a research scientist and so she is not only tolerant of this sight but accepts it as totally acceptable part of our Christmas 🎄 surroundings. Back on this thread next year when I have more to add! Thanks for following!
__________________
Al |
19th Dec 2016, 10:38 am | #73 |
Dekatron
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,195
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Hi Al, looks good, I see you got the meters repaired.
Ed |
19th Dec 2016, 11:23 am | #74 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Hi Ed, thank you!
Yes, I feel very happy about having restored the meters again. They are rare and wonderful things with a lot of history. It took a lot of TLC but clearly worth it. You can see that I have replaced the front with a kind of deep lacquer finish instead of the copper foil there before. I may well put a Lexan overlay on the front panel as I think it will be safer. I knew this wasn't going to be a rush project. It's definitely an order of magnitude harder to build a thing like this with a 2kW power supply compared to the last valve one I built with 'only' around 400W power out. And that's entirely to do with the critical placement of components, RF and power supply design etc.
__________________
Al |
19th Dec 2016, 10:29 pm | #75 |
Heptode
Join Date: Jan 2016
Location: Wolverhampton, West Midlands, UK.
Posts: 777
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
That is a beautifully hung piece of tinsel, but what's underneath looks very exciting too. Merry Christmas Al (was going to say "don't get too frazzled and get it fired up over the festivities", but I fear misrepresentation over use of the word "frazzled" in relation to the whole 2KWatt side of things)!
All the best from Bill. |
19th Dec 2016, 10:55 pm | #76 | |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Quote:
I hope you have a great Christmas and New Year, too!
__________________
Al |
|
12th Jan 2017, 5:51 pm | #77 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Quick update: I stopped work on this over the holiday season.
In the last few days I have now rebuilt the EHT power supply using 12 diodes each with a PIV rating of 1,300V and a peak current capacity of 3A. I have also re-configured the voltage doubler so that a single 0.85uF capacitor is in the doubler itself and then approx 4uF appears as a reservoir on the doubled supply. I have also increased the size of some earthing braid , now that I have an idea of the huge size of fault currents (thanks to Ed Dinning for the supply) and of the main conductor to safety earth internally. Ancillary but important - I have now added a flasher at about 1Hz which operates an incandescent bulb through a sizeable red warning indicator. This fires up when the EHT is powered, whether under load or not, and acts as a constant reminder to be vigilant. I may augment this with a buzzer or similar for a few seconds at standby. I need to do a few more tests on the main inductors , including measuring and adjusting coupling , tweaking the grid leak circuit and and anti-flashover measures before applying power to the tank circuit again. Cheers, back soon, I hope!
__________________
Al |
12th Jan 2017, 7:29 pm | #78 |
Heptode
Join Date: Jan 2016
Location: Wolverhampton, West Midlands, UK.
Posts: 777
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Hi Al, happy new year. I have no background in electronics unlike many here, so a thread like this really helps my understanding of the subject. I have a bit of a newbie question, if you don't mind helping out the less educated? What is the earthing braiding made of (material construction, size, etc.? I don't wish to bother you too much, but a quick picture may be worth a thousand words in this case? Sorry if it's all a bit "Ladybird Book Of . . ." but I am rather a greenhorn as mentioned.
Many thanks from Bill. Last edited by Bufo Bill; 12th Jan 2017 at 7:29 pm. Reason: Spelling. |
12th Jan 2017, 7:58 pm | #79 | |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Quote:
Any questions are always welcome here! Earthing braiding is made from (relatively) fine copper wire and tinned copper wires that are interwoven to a kind of mesh, very strong but flexible. The surface area is large and so is the sum of the cross-sections of all the wires; accordingly it is both good for conducting high RF as well as large DC fault currents. The 5A rated braiding I was using before completely vapourised in a flashover in one part of the power supply. I am now using 100A rated, which may seem like overkill, but was kindly supplied by Ed and also a handy RF conductor. I will post a photo tomorrow so you can see. Cheers!
__________________
Al |
|
12th Jan 2017, 8:34 pm | #80 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).
Clarification : it may be tinned or not tinned. So some types are tinned and some not, I don't mean both tinned and untinned in the same sample!
__________________
Al |