question re: stepping up ignition coils / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

question re: stepping up ignition coils

1 2 3 4 next last

Move Thread

LAN_403

thrival

Sat Sept 22 2007, 03:17PM

Registered Member #1019 Joined: Sat Sept 22 2007, 02:39AM
Location:
Posts: 29

I've a friend who has done some experiments running ignition coils into each other
in series stages. (Of course this voids the manufacturers' warranties --LOL). I tried this
myself with two coils and initially got a bit longer and fatter sparks but now whenever I
hook up a second coil, it just hums & buzzes (presume is arcing to the core internally);
as single coils they still work fine.

Which got me to thinking, wouldn't it be great, instead of building Tesla coils that use miles
of wire, to just step up an ignition coil? What say, we took a toroid and wrapped an auto-
transformer winding, similar to tapped RF chokes, stepping up 2 to 10 times?! My question
is, what would the parameters of such a device be? Obviously insulation is critical. A small
toroid could be potted in epoxy/mica dust as a filler. (Mica has a k of 10,000/mil.) But my
real question narrows down to the turns ratio. Do I need a lot or only a few turns? (Fewer is
easier/better in my view.) Note I don't plan on driving the ignition coil beyond 1): what an
automobile ignition module can deliver (or 2): a few tens of kHz.

...

Sat Sept 22 2007, 04:06PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

a. your image is way to big
b. You would need a huge number of turns to make such a device work, or else you will end up saturating the ferrite. We are talking on the order of 10v/turn at the absolute max for a huge core.
c. there also is no point in running ignition coils in series, as when you run them in 'tesla coil mode' (ie, dumping a cap into them), you can easily produce enough voltage to cause internal arcing. IIRC, the record for such a system was about spark about a foot long, which the system produced twice before it had internal breakdown and died. Even if you did want to run one off of a stepup transfromer, another ignition coil would be a horrible match, because it has way to high of a stepup ratio.

thrival

Sun Sept 23 2007, 02:06AM

Registered Member #1019 Joined: Sat Sept 22 2007, 02:39AM
Location:
Posts: 29

The friend who ran the ignition coils in series is not much older than you. Sure it's crazy
but that's how some of us learn. If you've spent enough time studying, experimenting
and thinking about a subject to provide an accurate answer, that's good enough for me.
Age has nothing to do with it.

Yes, I read on Sam Barro's PowerLabs.org page about the spark record for running an
ignition coil in Tesla mode. Obviously running such a coil to destruction is NOT my intent.
We know what DOESN'T work here.

You mention needing more/many turns or risk "saturating the ferrite" but my concern was
(given the current output of an ignition coil is so weak) for nothing happening at all. In RF
power supplies/HAM radios, ferrite cores of but a few turns are common-place. The difference
is they're running at much higher frequencies. Obviously the turns thing is where my
ignorance lies and why I asked; (have no books on HV coil-winding handy.) The best I
could find online are references to winding flybacks. The ARRL 2007 handbook says leave
.10" per kV for an air gap between wires but that's just a general recommendation, not
specific to coils. Epoxy/mica has such good k, one can squeeze more kV in a smaller
package.

Edited pic attached. Sorry, I tried to keep it small with .gif format, didn't realize there was
so much white space/would irritate.
Mod edit- image still too big- wouldn't voom if you put a million volts through it

Sulaiman

Sun Sept 23 2007, 02:43AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

There are some limitations that come to my mind

1) volt.seconds per turn, volts x time per turn is what determines core saturation
so you may think - if the time is incredibly short then the volts/turn can be incredibly large, true but
2) winding capacitance will linit the maximum volts per second hence ultimately, due to #1, volts.

3) inter-turn insulation is one concern but consider the path (first turn) to core to (last turn)

4) when I was 'playing' with ignition coils I found that insulating parts that worked really well at low/audio frequencies
completely failed at just tens of kilohertz. (e.g. the plastic top part of the coil)
If you multiply (volts per second) by (inter-electrode capacitance) you will see that large currents flow via insulators!

thrival

Sun Sept 23 2007, 02:57AM

Registered Member #1019 Joined: Sat Sept 22 2007, 02:39AM
Location:
Posts: 29

Sulaiman:

Many thanks for your reply. Volts X time-- ...hmmm. OK, So will a saturated core fail to work?
Again, I just want to step the V of an ignition coil up a few more times, from say 35kV to
100kV - 250kV. (Still trying to get my head around this coil-winding stuff, as pertains to HV.)
"Large currents flow via insulators" ...doesn't scare me, (in fact I kinda like that) so long as
the insulation doesn't fail! The ignition coil can do it's thing at one frequency while the
torroid can do another!

Sulaiman

Sun Sept 23 2007, 08:59AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I think your project is doomed to failure
but I learn most from my failures so I'm not saying that you should stop.

You could consider instead a voltage multiplier
have a look here

and here

thrival

Sun Sept 23 2007, 10:29AM

Registered Member #1019 Joined: Sat Sept 22 2007, 02:39AM
Location:
Posts: 29

Sulaiman:

I already know about Cockroft-Walton Voltage Multipliers, but thank you. My seeking
another way concerns the cycle-time required to charge all the caps in such a device.
Also, my understanding is they require AC input and an ignition coil employs DC, no?
Yes, learning from mistakes is good, but I've already made enough to want to save
myself time. For example I've already made a ferrite core variac using MOT
primary wire, about 130 turns, and a variable choke on threaded rod. When I measure
R, L & C across the contacts, I get readings, but when I connect to a sig-gen good for
200 kHz, get nada, like a dead short, across all the windings. I also hooked them into
an ignition coil circuit and get nothing out. So I'd really like to stop wasting time
(and materials) with devices that don't work and learn why such is the case.

The best I can figure re: the latter two devices is either my frequency isn't high enough
to create sufficient XL, or I really do have internal shorts, despite the care I took to avoid
them. Anyway with respect to the image/idea I posted, it would be a big help if you could
refer me to the relevant formula, (with insights gleaned) because my ultimate goal is to
design and build working inductors without baby-sitting or hand-holding, and learning 'how'
is why I'm here.

Steve Conner

Sun Sept 23 2007, 11:15AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I'm not sure I quite understand the question here. You're asking why we don't use ferrite cores in Tesla coils, or similar transformers where the object is to get an extremely high voltage output for making big sparks?

Well the answer is insulation. Ferrite is an electrical conductor, a very poor one, but a conductor nonetheless. If we took the secondary of a Tesla coil and stuffed it full of ferrite, the high voltage output would just arc to the ferrite and set the whole thing on fire.

Tesla coils are constructed the way they are in order to get adequate clearances and insulation for the high voltage output, and the electrical parameters come second. Of course we would love the higher coupling of a ferrite core, but the challenges of insulating it stop us from doing so.

Exactly the same objection applies to your toroid thingy, only slightly worse because it's physically smaller. You could pack it in mica powder all you liked and the wire would just arc to the ferrite core at a few kV.

Marko

Sun Sept 23 2007, 11:20AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Even the air cored secondaries fail from overvoltage as you see, there's really not much use of ferrite. It would just saturate at high peak powers in GSTC's.

The coils are built the way they don't require high coupling, and too high coupling would actually make a static gap harder to quench.

thrival

Sun Sept 23 2007, 02:37PM

Registered Member #1019 Joined: Sat Sept 22 2007, 02:39AM
Location:
Posts: 29

Steve:

I'm not asking why you don't use ferrite in Tesla coils, they do use it in ignition coils.
(Actually my torroid is powdered iron-resin.) I'm trying to step up the V of an ignition coil
without having to build a Tesla coil (or use a comparative number of turns or large size.)
Also I don't plan on doing anything in the upper 100 kHz range. Also solid-state Tesla coils
don't use spark gaps, and in my case, hard-to-quench sparks are a good thing, so is
high-coupling! Big, fat, looong HV sparks is my goal.

Mica has a dielectric 'k of 10,000V/mil. A mil is one-thousandth of an inch. A coat of paint
is thicker than that. I can insulate the core with enough coats that arcing to it shouldn't be
a problem. Again, my real question concerns the turns ratio, if you think it's possible to
soup up the voltage with only 3-5 turn primary and 20 or 30 turns total, (if there's enough
inductance there, assuming I can prevent internal arcing), in the ELF to upper audio range.
thanks.

1 2 3 4 next last

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