Need ultra-reliable "indestructable" 555 ignition coil/flyback driver / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Need ultra-reliable "indestructable" 555 ignition coil/flyback driver

1 2 next

Move Thread

LAN_403

V3rge

Fri Oct 15 2010, 10:48AM

Registered Member #3253 Joined: Wed Sept 29 2010, 12:21AM
Location: New Zealand
Posts: 13

I'm not new to ignition coil drivers, been making them for years now, very unsuccessfully. In fact, the longest runtime I ever had on one of my drivers was barely 4 minutes. I have always made them loaded with protection, reverse parallel diodes across across coil and FET, snubber circuits etc... the fact is none of them ever worked properly, even though everything was overkilled considering what should be needed and what other people use. Eventually I found out that I had killed my ignition coil, thats right the secondary insulation had broken down and was feeding straight into my timer and FET. I had never run the coil over 12v, always used low frequencies, varied duty cycles etc but in some builds I had quite often left the secondary wide open for long periods of time.
It really annoys me seeing others having such reliable circuits using things like 2N3055's and "low" voltage FETs or transistors. Some of my drivers that performed worst used 500V 20A FETs with heavily fortified circuits, and others where 55V FET's and no protection outlasting them dramatically.

Past that, I plan on buying another ignition coil soon and this time i'd like to make the driver nearly indestructible. I want it to be able to run for hours at a time (ignoring heat) and be able to handle a continuous open secondary with no damage, something most of my other circuits could have easily done. My experience with the coil I have is just rediculous, there is no other way it could cause as much damage as its done without an internal breakdown which has given me absolute hell for years.

Does anyone have an idea of what voltages are reached on the primary, when the secondary of the ignition coil is open circuited with or without snubbing/protection components?

Has anyone got an 555 ignition coil/flyback driver as stable and robust as what im looking for? Tell me about it.

Steve Conner

Fri Oct 15 2010, 11:22AM

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

Do other people really have reliable circuits, or do they just keep quiet about the failures?

All of the 555-based circuits are flawed, in so far as they don't include a snubber capacitor to tune the energy kicked back into the primary. So it just ends up avalanching the switching transistor and cooking it.

The kickback reaches hundreds of volts whether or not the secondary is loaded, because the coupling between primary and secondary is less than 1, and energy is stored in that leakage inductance. Besides, the kickback IS the output voltage, transformed down according to the turns ratio. If you don't let it kick back a couple of hundred volts, you won't get much output voltage.

ScotchTapeLord

Fri Oct 15 2010, 11:36AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

Certain companies like "Ford" and "Honda" have come up with fairly robust ignition coil drivers, as far as I can tell. Maybe you could learn a lesson from them?

A lot depends on the output. When you introduce a light load to an ignition coil, you are running it "improperly" (but that's what it's all about on this forum!) and cannot expect reliability.

1n5ane

Fri Oct 15 2010, 05:52PM

Registered Member #3283 Joined: Wed Oct 06 2010, 11:03PM
Location: Toronto, Canada
Posts: 29

Have you tried placing an avalanche or TVS diode across the ignition coil?

Adam Munich

Fri Oct 15 2010, 09:13PM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

I've had good luck with varistors (MOV type). They are much faster than avalanche diodes and can handle more current.

ScotchTapeLord

Fri Oct 15 2010, 10:24PM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

I've paralleled MOVs and RC snubbers across the ignition coil input terminals to some degree of success. You might want to use IGBTs instead of FETs. The MOSFET's intrinsic body diode is probably one of the flaws in your setup. If you use an IGBT with a diode built in, it will be faster and better suited for the application. Also, the frequencies aren't high enough to justify using a MOSFET over an IGBT.

What's also important is plenty of decoupling capacitors! Across the primary+semiconductor and also across your timing circuit. Also, a metal enclosure for your circuit could solve some issues. Those 555s are susceptible to interference, and if you've ever driven an ignition coil anywhere near a computer mouse with a laser "eye" then you know how much interference those things can give off.

Also, ignition coils aren't meant to be run unloaded... That is what destroys the secondary insulation. The voltage simply rises too much. If you want to run it unloaded then you'll need a circuit to sense when it is sufficiently loaded.
Hmm...
Maybe tie pin 5 of the 555 timer to a low value resistor between the switching device and the ground? I think I will look into this.

V3rge

Fri Oct 15 2010, 11:30PM

Registered Member #3253 Joined: Wed Sept 29 2010, 12:21AM
Location: New Zealand
Posts: 13

Thanks for your replies. I will definitely be using MOV next time, I completely forgot they existed until you reminded me. A couple of my worst build failures used IGBT's (HGTG12N60C3D 600V 24A w/ hyperfast A-P diode) which were totally outlasted by my very first unprotected ciruits with 55v FET's. My dead coil absolutely nuked everything I attached to it :S
I know that there is still a relatively high voltage generated on the primary, even when the secondary is conducting but that should be handled no problem by something rated over 500v. I had a couple caps in the snubber break down and short a couple times, but they were 20/30 year old caps :)

I would have expected the insulation on the secondary to handle an open circuit, it would be interesting to see what sort of voltage it actually reaches when left open... (but I can't see it going a whole lot higher than the nominal output voltage)
Before I go and buy myself some more coils, i'll start this all over again with a flyback (if I can ever find the pinout!) since i've been wanting to get a flyback to work for years...

I was also thinking about using a sensing/feedback cricuit for detecting wether or not its loaded, something like this could significantly more robust. No one can tell me they've never left a secondary open before right?

ScotchTapeLord

Sat Oct 16 2010, 01:41AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

Have you ever seen the spacing in a spark plug? It's a tiny, tiny gap that limits the output to maybe a few kV to start the spark, then only a few volts across it. They simply aren't meant to be run with an open secondary. As for the voltage on the primary... it can certainly get VERY high. In fact, overvoltage is pretty much the only way to kill an IGBT in an application like this where current is limited!

Adam Munich

Sat Oct 16 2010, 01:56AM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

If you want an ultra-reliable one use a relay. Their abrupt switching will cause quite a high voltage on the secondary. Plus it would be incredibly hard to burn them out. In fact the contactors on old spark coils could be thought of as a relay. The max frequency will be only about 40hz, but some impressive sparks can be made.

The 555 could control the relay, but I've simply hooked it up to a magnetic switch taped to the ignition coil and had it oscillate. All you need is to hook the thing up to the NO terminals on the relay and put the magswitch in series with the coil. Bring it close to the coil and you'll find a sweet spot where it'll start oscillating. Then use popsicle sticks and a hot glue gun to position the magswitch. The drivers don't get any more unbreakable.

GhostNull

Sat Oct 16 2010, 05:52AM

Registered Member #2648 Joined: Sun Jan 24 2010, 12:45PM
Location: Australia
Posts: 291

You'll need to remember a flyback diode across the relay coils or the kick back from the coil might do some damage.

1 2 next

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