UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Other Discussions > Homebrew Equipment

Notices

Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members.

Closed Thread
 
Thread Tools
Old 19th Nov 2016, 10:16 am   #41
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Ah, I mean thank you Red and Black, too!
__________________
Al
Al (astral highway) is offline  
Old 21st Nov 2016, 12:00 pm   #42
retailer
Heptode
 
Join Date: May 2005
Location: Adelaide, South Australia
Posts: 541
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Hello Al
looks like you are having a great deal of fun, I would heed Sean's warning about multi KV supplies - back in the early 80's I worked as a lab assistant in a private school and built a convential spark gap type Tesla using a 1kw 10kv smoke precipitator transformer - good for 24-30 inch sparks. One day I forgot to move our nice new expensive Beckman 3020 digital multimeter well clear and noticed too late that sparks were jumping from a few tools (which them selves were around 3 or 4 feet from the secondary coil) I had on the bench to one of the probes - the Beckman was toast and my name was mud.
I left to finish my uni degree and some years later the school closed up, they called and asked if I would like to take the Tesla to save it from the skip - which I did - sadly I eventually had to sell it, even though it was on castors it was big and I did not have the room to keep it, my wife said it frightened her she was glad to see it go, she said the fact that the ceiling fluro tubes lit up with being turned on must surely mean I'd get radiation poisoning, I couldn't convince her otherwise.

Will be keeping any eye on this post and looking forward to seeing pictures of if operating.
retailer is offline  
Old 23rd Nov 2016, 5:53 pm   #43
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Hi there,

Thanks for your illuminating tale, that's definitely one to remember for life!

Building this big old thing is many things - absorbing, challenging (mainly because I don't have a workshop!! ) , and strangely rewarding. I think it probably will be deeply gratifying when it is all tuned-up and producing nice long plasma , but at this stage of my journey , I wouldn't describe the stages leading up to that as 'fun' - I find it very sobering to work around and there are many frustrating setbacks to deal with first.

I think because I've built three already, the act of making the plasma is less of a big deal, and the satisfaction of building a machine that will run without overheating or catastrophic breakdown of a key component is much more in the foreground for me.

It means I have put a lot of thought into the mechanical side of the construction and various features to protect against surges, flashovers etc.

I'll certainly be posting photos , probably in a couple of weeks' time.
__________________
Al
Al (astral highway) is offline  
Old 25th Nov 2016, 6:03 pm   #44
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Hi folks, quick update:

-- The primary coil is now wound on Litz wire, in two layers (two identical inductors in parallel) to handle an average current of 20A in the tank circuit. I heated the Litz wire with a hair dryer and wound it wearing surgical gloves, as the coating is very moisture-wicking. I then coated it with three coats of varnish, baked in between. Over that I put two layers of Kapton tape for durability, and finally, I used epoxy potting compound to insulate the protruding ends of the coil windings. This is because a flashover can easily occur between here and the grid windings, above or below (below in my case.) NB epoxy potting compound is NASTY. Use in a well-ventilated room, or preferably outside. it is highly irritant to eyes, lungs and skin.

--The grid coil is also wound, below the primary. It is capable of being moved up or down slightly for tuning/ coupling purposes.

--The secondary coil is finished. I reached a depth of over 1mm of cured varnish, in layers, then, starting at the top, I went over the whole coil with epoxy potting compound, paying special attention to the top of the coil where breakouts or flashovers can occur. I also finished the mounting detail.

The base of the coil features half of a threaded brass plumbing fitting, which is brazed onto two sections of brass mini-busbar and a copper strap (pic. 1) This screws onto the other half of the brass plumbing fitting, which is brazed onto the 3mm copper sheet which forms the main ground for the secondary coil. This is a very sturdy arrangement, and allows the whole secondary coil to be removed for storage or transit. Pictured is the whole secondary coil, complete with its topload (pic. 4, detail)

--The topload is also finished (detail, pic 4). I brazed a sturdy beryllium copper bolt onto the stainless steel ball. This is where the breakout will occur when the coil is operating, although it will probably need a wire 'streamer' to extend it for tuning purposes.

--I found a more powerful fan to cool the two EHT transformers. (Pic 3.) This one is an 'Arctic' instead of the ubiquitous Evo Labs that I see everywhere, and comes from a gaming PC that was scrapped in the street and which I raided for the purpose.

The fan draws 250mA, instead of the usual 150 for most CPU fans, and has nine blades instead of the usual seven. Three of the four support spars on one side were broken so I performed some elaborate surgery to mend them without affecting the very tight fan blade clearance, not easy to do.

--I installed a separate SMPS for the 12V needed by the fan, to isolate it from the regulated power supply I built earlier, which will control some logic circuitry.

-- I installed home-made copper sheet heatsinks (one shown in pic 2) for both the 22R resistors which control the huge current inrush from the EHT transformers. These are shunted in about a second by a relay, but they instantaneously deal with a gigantic amount of energy in kJoules and previously failed (the air inside expanded violently, blowing one lug and its ceramic contents out of the side).

Next week I will install the legs that Andy (Dr Wobble) has now finished, I think made of pear wood, and sent down by courier!! I can't wait to see these, due for delivery on Monday!!

I will also start work on the rest of the tank circuit and the anode connections. Critical design, huge current at 15kV knocking about in here. This will occupy the area painted yellow.

I'll also make a switch panel, humdrum but rather essential, since things have to be switched on in a certain order and I'd like some nice clear indicators for filament on, EHT on, etc.

I think we're about a week away from initial testing...

Watch this space!!
Attached Thumbnails
Click image for larger version

Name:	IMG_0882.jpg
Views:	199
Size:	59.9 KB
ID:	133380   Click image for larger version

Name:	IMG_0887.jpg
Views:	200
Size:	73.6 KB
ID:	133381   Click image for larger version

Name:	IMG_0897.jpg
Views:	196
Size:	48.5 KB
ID:	133382   Click image for larger version

Name:	IMG_0902.jpg
Views:	228
Size:	33.3 KB
ID:	133383  
__________________
Al

Last edited by Al (astral highway); 25th Nov 2016 at 6:10 pm. Reason: Uploaded pix correctly, added some details.
Al (astral highway) is offline  
Old 25th Nov 2016, 11:37 pm   #45
Bufo Bill
Heptode
 
Bufo Bill's Avatar
 
Join Date: Jan 2016
Location: Wolverhampton, West Midlands, UK.
Posts: 777
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Very exciting, and I particularly love the use of salvaged items. A man willing to climb into a skip is someone to be greatly admired in our disposable culture with its inbuilt technological crapulence (if I had the money to buy a £600 smartphone, I would expect it to be built to last, not set itself on fire).

Stuff arriving by post is like an extra birthday (having ranted on about rubbish tech, the Internet gets a cautious thumb up for creating the ability to have friends who craft bespoke parts for one's Tesla coil, and indeed friends who know what a Tesla Coil is).
Looking forward to hearing more. Regards from Bill.
Bufo Bill is offline  
Old 26th Nov 2016, 4:37 pm   #46
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Quote:
Originally Posted by Bufo Bill View Post
Very exciting, and I particularly love the use of salvaged items...
Thanks for your geneorous comments, Bill! I do enjoy salvaging, and it keeps the cost down on a project like this -- it would cost almost £100 to build just the EHT power supply from new components (or even second hand ones from EBay)!! There is a lot of 'street-treat'-ing where I live, where it's acceptable to leave working consumer electronics and old furniture in front of houses instead of skipping it. Also, shops frequently dump things like U/s microwave ovens , so it pays to keep the eyes open. Curiously , it's quite hard to buy suitable rated wire for the high voltage end ...

Anyways, thanks again !


NEWSFLASH!!

The legs from Andy (Dr Wobble) are now with me...so I'll be fixing these on next week! Thanks , Andy!
🔙🔜💡💡💡
__________________
Al
Al (astral highway) is offline  
Old 26th Nov 2016, 5:31 pm   #47
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Car ignition cable? The proper copper stuff, not the carbon fibre resistive stuff? I've used it for EHT many times.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 27th Nov 2016, 9:10 pm   #48
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Quote:
Originally Posted by Radio Wrangler View Post
Car ignition cable? I've used it for EHT many times.

Good idea! Only to clarify, I meant it's proved hard to get wire of the same spec as what's in microwave ovens, other than in microwave ovens.

The silicon coated stuff available on Ebay (and I had to buy it from Italy) isn't nearly as good as what appears to be a very high quality version native in microwave ovens, which is high current capacity, maybe 18SWG, silicon coated and then cotton-or similar double-covered. It's of academic interest now as the EHT is all sorted, but in future I'll definitely remember your idea if I need any more. Thank you David.
__________________
Al
Al (astral highway) is offline  
Old 27th Nov 2016, 9:22 pm   #49
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,188
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Farnell sell the 'traditional' EHT cable rated at 25kV. It's not cheap, though...

http://uk.farnell.com/carlisleit/cd1...rfnonsku=false

RS sell a range of wires going up to 60kV. But be sitting down when you see the prices, some of them are over £800 for a 100 foot reel...
TonyDuell is online now  
Old 28th Nov 2016, 7:10 pm   #50
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Ahaha!! i just weighed the chassis without the big main inductor ...I've been struggling to do things that were very easy before my surgery but I was no way prepared for this result from the scales!!

Result? It already weighs 19kg! No wonder it was feeling a bit unwieldy!!. The legs, remaining components and the huge valve will take the final weight to 22Kg.

Good to know if you're intending to replicate this !! Most of the weight is the two microwave oven transformers and the 10A filament transformer, so there isn't a possible lightweight version at 50Hz power in...
__________________
Al
Al (astral highway) is offline  
Old 7th Dec 2016, 2:35 pm   #51
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Update: I seem to be in a phase where aspects of the mechanical construction need sorting before I can finish the RF circuit and get testing. If I carry on building on the upper side of the 'chassis', it will be increasingly difficult to turn the thing upside down to tweak wiring underneath...so it's all about getting the legs on and �� strong! Not something I envisioned being a factor. Right now I am waiting for some internal bracing pieces -kindly being made out of pear wood by Dr Wobble (Andy)- thanks again !!

I am reinforcing the chassis ahead of this .This is an example of the difference between a lash -up and something robust and more permanent . I wondered if it could come across as a bit dull, but when I started his thread, I wanted it to document the journey, ups and downs, so this level of honesty seems appropriate. I know some of you are keen to see spectacular plasma! Please hold on, the time will come soon!!

I have also finished varnishing the cabinet (8 coats) and installed a switch panel from some antique brass that came from the cellar of my old house . A tiny detail on the face of it, but the symbolism of embodying some material from a place I loved means a lot to me. The same brass is in the corner pieces I have fashioned for the legs , pictured.

This may seem trivial to some but with my current facilities , it takes a while just to enlarge holes in 2mm thick brass from 10mm (my largest twist drill bit) to 13mm ...diameter !

Also, in the background , I'm researching and professing with my design for the PLL.

That's it for now...
Attached Thumbnails
Click image for larger version

Name:	IMG_0957.jpg
Views:	159
Size:	99.8 KB
ID:	133994   Click image for larger version

Name:	IMG_0959.jpg
Views:	163
Size:	75.7 KB
ID:	133995   Click image for larger version

Name:	IMG_0961.jpg
Views:	155
Size:	35.6 KB
ID:	133996  
__________________
Al
Al (astral highway) is offline  
Old 7th Dec 2016, 4:39 pm   #52
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,636
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Oh, see you've updated your thread, must have been away with the faeries. good to see you making progress. No worries re legs, twas a pleasure. It was nice to work with my hands again and do something intuitive and tactile that doesn't involve complicated electrons. : )

Will get the bracing done ASAP and sent down for you.

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is offline  
Old 7th Dec 2016, 5:55 pm   #53
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Quote:
Originally Posted by Dr Wobble View Post
Will get the bracing done ASAP and sent down for you.
Thanks again, Andy, it's great to have your artisan woodworking involved in the legs - literally supporting everything above!
__________________
Al
Al (astral highway) is offline  
Old 13th Dec 2016, 6:15 pm   #54
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Update:

I completed construction of the RF circuit and ran what I thought would be an initial test.

A flashover instantaneously occurred between the lead-outs from the tank capacitor (copper strips 3cm apart).

The over-current immediately blew the EHT rectifier in half and its ceramic bypass capacitor to smithereens. Fortunately both those events took place under the chassis.

This kind of thing is much more likely to occur when the tank circuit and secondary coil properly, so I will now try to pin down the exact resonant frequency of the tank circuit before I rebuild the damaged power supply.

I will also use a chain of 3A, 1300V rated diodes instead of a single EHT diode from a microwave oven, given that in a fault event, current here can easily exceed 1A.

I will use a higher value of resistor in the grid-leak circuit - the 40W bulb I am using clearly has a very low cold resistance (positive temperature coefficient, obviously) and the anode likely passes too much current for instantaneously. I will design something that enables the higher value resistor to be bypassed once the bulb has warmed up - perhaps just a second or so after switch on.

The bracing pieces from Dr Wobble (Andy) arrived today. So I can get building the legs and bracing structures while I'm waiting for the new EHT diodes.

I would love to get this going before 21 December, the longest day/ Winter Solstice, as a way of producing an unusual kind of light to celebrate. Let's see.

Cheers folks
__________________
Al
Al (astral highway) is offline  
Old 13th Dec 2016, 6:30 pm   #55
Guest
Guest
 
Posts: n/a
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Bangs, phutts, blowups etc. are all part of high power stuff. It can be high voltage low current or low voltage high current, it's all the same, bang and then expense. I had quite a few bangs making (designing?) motor controllers for an electric tricycle. All good fun, I don't know if high voltage or high current is more dangerous, I once snapped all the spokes on a wheel by making it go backwards all of a sudden (ah the joys of brushless DC motors with ill conceived firmware), rather scary in the kitchen. Sort of a whizz, donk, ping and shrapnel.

Blow ups don't matter as long as you find out why!
 
Old 13th Dec 2016, 10:41 pm   #56
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Quote:
Originally Posted by merlinmaxwell;901265I
don't know if high voltage or high current is
Ah, thank you, that helps me to put it all in context. We have high voltage and high current together here, which is especially destructive when it ends up where it's not supposed to.

The current that destroyed the EHT diode was enough melt the entire body in an instant. There was nothing left, either, of both its lead or those of its bypass capacitor. These were destroyed before the fuse blew. This huge current derived from the tank circuit --

However, I can't quite work out the power in from the power supply. The 4uf capacitor charges to its full voltage in n approx 4uS, assuming it is an ideal components with no internal resistance (which it is not, but I can't see a huge impedance anywhere? ).

So that means the power supply can source about 60J every 4uS at this voltage (0.5 C*V E^2)... or have I missed something? I'd be interested to check in with you guys about what the theoretical instantaneous power is. It's obviously rather a lot.
__________________
Al
Al (astral highway) is offline  
Old 14th Dec 2016, 8:54 am   #57
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,188
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

If you really can charge that capacitor every 4us (repeatedly) and if it is storing 60J, then that implies to me an input power of 60/(4*10^-6) or 15MW. Unless you happen to own a power station, I do not believe that.
TonyDuell is online now  
Old 14th Dec 2016, 10:53 am   #58
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Quote:
Originally Posted by TonyDuell View Post
Unless you happen to own a power station, I do not believe that.
Hahaha! The fact that this is a very silly proposition was implicit in my inquiry

So can you resolve the paradox? I'm getting the 4uS from my understanding of time constants, 4uS for 4 microfarads and a nominal 1R. So actually , yes , 4uS is only one time constant and we need five time constants for full charge, so 20uS. On the other hand, clearly the two transformers have an indeterminate but real impedance and the power in is limited.

The conundrum arises because there is clearly more than the energy available from the energy stored in a capacitor equation I cited...the power supply blew an EHT diode in half, exploded a ceramic capacitor and melted a component lead
and solder joint. That'S more than a few tens of Joules. So, how exactly do I calculate the instantaneous (versus average) power available?
__________________
Al
Al (astral highway) is offline  
Old 14th Dec 2016, 8:22 pm   #59
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,188
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

I don't think I believe that time constant (or rather charging time) either.

Let me explain. You have a 4uF capacitor, and it seems like you are charging it to around 5kV. Now the charge on that that capacitor, when fully charged is (Q=C.V) is 0.02C. Now to move 0.02C is 4uS implies a current (I = Q/t) of the order of 5kA. OK, the current will not be constant but that's a sort of average value. I do not believe you have a 5kV supply capable of supplying 5kA.

Quite simply, I think your charging time is a lot longer than 4us. The supply impedance is a lot more than 1 ohm. It has to be.
TonyDuell is online now  
Old 15th Dec 2016, 1:27 pm   #60
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: Design-and-building a large valve Tesla coil (class C, Armstrong oscillator).

Quote:
Originally Posted by TonyDuell View Post
I do not believe you have a 5kV supply capable of supplying 5kA...

Quite simply, I think your charging time is a lot longer than 4us. The supply impedance is a lot more than 1 ohm. It has to be.
Hi Tony, yes, again, it's an absurd proposition to suggest that my EHT supply can dish out 5kA. Nobody is saying that.

I think we're trying to look at different things here. I'm suggesting that the peak, instantaneous power output is very much higher than the average say 2KW of which it is capable, being 2 x Microwave oven transformers in parallel.

Thinking and sensing this through, I believe that the rig can supply a current in very high tens of amps for an indeterminate (very small) time - this is calculus, of course. I don't have LT spice as I am a MAC user, otherwise I would model the situation to get to the figure I'm sensing.

There are actually two sets of currents that I'm interested in. The second one is in the melt-down I described . This was clearly not energy from the power supply directly, but energy coupled in from the tank circuit.

The flashover occurred across the tank capacitor - i.e. the straps between the capacitor and one of the main inductors.

If we break this down, it's like this.

1) When the arc was made, the power supply continued to supply energy to the arc for a second or so, shorting out the inductor.

When the fuse blew, the arc instantaneously stopped, 'opening' the shorting 'switch' made by the flashover arc.

2) the tank circuit was now intact again with no short across is. Hence it supplied a huge pulse of energy, which began to decay in resonance (the valve was no longer conducting at this point, since the power supply fuse had blown on the primary side). Since there was no RF circuit through the valve to ground, this energy passed straight back into the power supply.

I know from back of the envelope calculations that at least 20A circulates in the tank circuit. It may well be more as the inductor is very high Q. For some reason, I believe a much larger current (RF) was coupled back into the power supply, and it would be interesting to know what that could have been.

What do you make of that?
__________________
Al
Al (astral highway) is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 1:59 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.