First tesla coil thread - Dual Mot / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

First tesla coil thread - Dual Mot

first previous 3 4 5 6 next last

Move Thread

LAN_403

MrFlatox

Sat Oct 19 2013, 04:51PM

Registered Member #9349 Joined: Mon Jan 07 2013, 08:50AM
Location: France
Posts: 102

Yes, homemade capacitors are rude, but I want to at least run the coil with them, even if i do not exclude the possibility of going with mmc in the future. I saw a tc with homemade rolled capacitor that works great.

At the moment, i am trying small rolled caps with overhead projector transparent plastic sheets. I tested with only one sheet (0.1mm) between the two rolled plates and it worked well across my power supply with or whithout the rotary spark gap. I am planning to make 5 of them, and parallel them to reach my goal capacity of 130nF. I also tried your technique about the overlaping plates at each ends of the capacitor and it works just fine.

OHP sheets are not cheap but i happen to have 50 of those laying around.

@ Hydron : thank you :)

MrFlatox

Mon Oct 21 2013, 04:57PM

Registered Member #9349 Joined: Mon Jan 07 2013, 08:50AM
Location: France
Posts: 102

Hi everyone !

So I built five rolled capacitors, stacked them in parallel to achieve 115nF, and I had a chance to hook them up on the tesla coil.

I got no sparks at all on the toroid, so I assume my coil is severly out of tune. I already measured (physically) my secondary resonance frequency which is 164 kHz. So I ran javatc to help me find the right tapping point on my primary coil. Javatc gave me 4.85 turns, but my coil ends at 4.75 turns, and when I set the point at the end (4.75 turns) I get nothing on the toroid.

I tried moving my wire along the coil to try 4.5 turns, still nothing. I even tried 3.75 turns and as a result I got some tiny 5cm violet streamers on the toroid, almost invisible even at night.
After 30 seconds of cumulated tests, one of my capacitor literally burnt out...

So, I was wondering, how accurate is the "sweet spot" on the primary coil to make things work ? a cm ? a mm ?

When javatc said 4.85 turns and I set my coil to 4.75 turns, should I atleast get something on the output ? I must have a greater primary capacitance to lower my primary frequency.

I expected the rotary spark gap to be really loud and "dazzling", but instead, the sparks are not that bright and loud. Should I suspect a problem with my capacitor setup ? Or the power supply ?

As this is my first TC, I am kinda lost.

Thanks in advance.

RateReducer

Mon Oct 21 2013, 07:01PM

Registered Member #4603 Joined: Wed Apr 25 2012, 07:33PM
Location: Austria
Posts: 159

The sparc of the spark gap should be really loud an powerfull. When i tuned my SGTC i tried to vary the primary tap, 1/4 winding lenth - i did not see a noticeable change of the spark lenght. mm, cm does not change anything...
You should nearly always get some streamers - even when your coil is far out of tune.
It looks like your caps are dead. Maybe one cap has an internal shortcircuit?

MrFlatox

Mon Oct 21 2013, 08:14PM

Registered Member #9349 Joined: Mon Jan 07 2013, 08:50AM
Location: France
Posts: 102

It's what I thought, my caps were not functioning properly. One of them failed short, but others are good on a DMM. I think my caps are not well design to properly work on a tesla coil, I must have missed some mandatory features for TC applications...

On every youtube video of spark gap tesla coil, the spark gap sparks seem very bright and loud, which is definitely not the case with mine. As the capacitor are not providing enough power to the coil, that would explain why i get nothing on the ouput.

My power supply seems to work, as it cans draw large arcs alone, and draw 20+ amps on the 240V line.

I think somewhat is limiting the current flow on the primary side. I understodd that a tesla coil on operation is putting several RMS Amps of high frequency current on the primary side when the capacitor is discharging and ringing through the coil. All my cables are heavy gauge multistrand wire, connections between elements are made with aluminium blocks and copper cable ends. The weaker point of the whole thing might be the aluminium foil of the capacitor. When I tested the capacitor with AC current, they got pretty warm after about 20 seconds of operation. The dissipated heat might come from current (R*IÂ²) or dielectric losses. This might explain the difficulties I am encountering.

While investigating those possibilities, I will look for MMC. What is the nominal voltage my MMC bank should withstand if my power supply is 4200Vac ? Is 10kv a reasonable figure ? Does a well designed MMC with balance resistor can be reliable and last a long time ?

Thanks

Hydron

Tue Oct 22 2013, 12:40AM

Registered Member #30656 Joined: Tue Jul 30 2013, 02:40AM
Location: UK
Posts: 208

Assuming you have good parallel contact to the edges of all the aluminium foil "plates" then the RMS current in them is unlikely to be the issue. The MMC caps everyone use have extremely thin films of evaporated aluminium for plates, though many (e.g. the CD 942C) also use foil for more reliable electrode contacts.
These caps use a much lower loss dielectric though (polypropylene), and also don't have the corona issues you'll have with a dry DIY cap. You can heat up these caps too though if you're not careful, and they don't like it!
I suspect it is the dielectric loss that is causing the heating, though I only have limited experience with this, so others can probably give a more definitive answer.

Edit: A quick check of the dissipation factor for cellulose acetate (which is likely what your OHP film is) suggests it is very lossy. The numbers I found suggested it is ~100x lossier than polypropylene, which is a bad sign for tesla coil use.

Uspring

Tue Oct 22 2013, 11:06AM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

MrFlatox wrote:

I understodd that a tesla coil on operation is putting several RMS Amps of high frequency current on the primary side when the capacitor is discharging and ringing through the coil.

That is somewhat of an understatement. Assuming a primary C of 100nF and a primary L of 10uH, the primary tank Z is about 10 ohms. If you have your primary C charged to 6kV when the gap fires, the peak current will be 600A.

MrFlatox

Tue Oct 22 2013, 03:26PM

Registered Member #9349 Joined: Mon Jan 07 2013, 08:50AM
Location: France
Posts: 102

100x lossier than PP ! damn ! that must be the cause of the very poor efficiency, and high heat dissipation I got.

I started looking at MMC, expecially the 942C serie, and it reminds me the major reason why I tried to build homemade caps : they're are really too expensive !

Ash Small

Tue Oct 22 2013, 03:35PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

MrFlatox wrote ...

100x lossier than PP ! damn ! that must be the cause of the very poor efficiency, and high heat dissipation I got.

I started looking at MMC, expecially the 942C serie, and it reminds me the major reason why I tried to build homemade caps : they're are really too expensive !

If you can't get PP try PE, if one layer of dielectric isn't enough, try two. Look for good quality polythene, which feels smooth to the touch.

EDIT: Have you taken 'skin effect' into account?

It looks like only the outer 160 microns or your conductors are conducting at 164kHz, and you apparently have 600A.

MrFlatox

Tue Oct 22 2013, 04:45PM

Registered Member #9349 Joined: Mon Jan 07 2013, 08:50AM
Location: France
Posts: 102

I do not really see how could I fight against skin effect.

I do have PP plastic rolls, but it's kind of a textured plastic, and it is very thin (35 microns) so I do not see that as a viable option for making a capacitor.

Ash Small

Tue Oct 22 2013, 05:03PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

MrFlatox wrote ...

I do not really see how could I fight against skin effect.

I do have PP plastic rolls, but it's kind of a textured plastic, and it is very thin (35 microns) so I do not see that as a viable option for making a capacitor.

Either use litzt wire or foil/plate conductors.

You could try sevearal layers of the PP, depending on how much you have. You may well be better off looking round for some PE, though.

first previous 3 4 5 6 next last

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