Voltage Spike/Capacitor

Would the best way to prevent transient voltage spikes when powering on/off a circuit be to add a capacitor between the DC supply and ground? If there's a better solution, I'd appreciate the advice!

I have a 555 timer circuit feeding some 2n3055's which I'm using to make high voltage sparks through ignition coils. I keep frying the 555's and 2n3055's when I turn the circuit on or off when it's tuned to resonance. For my project I'd like to be able to keep it tuned to resonance when I power it on or off.

Actually, adding capacitors will make the turn on worse because of the initial current pulse that charges up the capacitor.

To supress resonant effects from disconnecting current through coils (ie. turn off) you need flyback diodes or RC-snubbers (which in a sense is adding a very small capacitor...but you need the resistor to tune the damper circuit and stop the capacitor from burning out).

However, you are being more vague than you might think when you say that the voltage spike occurs when powering on/off the circuit. Which is causing the volage spikes? Turning on? TUrning off? or both? It is much more likely that it is just the turning off part.

I built one of those when I was a kid. Had the exact same problem. The 2n3055 (RadioShack?) tops out at around 60 volts colletor to emiter. Thats way to low for an ignition coil switching circuit. the voltage pulse blows through die and often goes back up the base and into the 555 output distroying one or both.
I found the transistor they used in many chrysler solid state ignition sytems from the earlier days was up in the 200 volt range. More modern stuff uses mosfets or IGBT's now. Voltage controled switching devices. They are easier to control and filter the 555 output to the gate line too. Plus IGBT's and industrial Mosfets are often 600 volt and high amps rated but still in a small package. Plus very cheep.

Thanks for the replies, and sorry for the delay. I had to build a new one to test whether it was getting fried on on or off, and it's off. Since that's the case would the cap help?

What transistors would you recommend to power an ignition coil? The first time I tried it I used a TIP3055 and that went up in smoke after about three seconds. Every schematic I've seen suggests the 2n3055's(from radio shack), so that's what I've been using.

could you post a schematic?
I have played with high voltage circuits for as long as I can remember.
I love to make sparks too!
I burned out a lot of stuff along the way though! A good schematic will help alot. Plus how many amps are you switching? is it an internal resistor (self curent limiting) type coil or not?
most newer ignition coils are built with a very low ohms primary. They are specificaly for the modern CD and multi spark ignition systems. At .25 ohms or less they would blow the 2N3055's dies right off the inside of the transistors. 12/.25 = 48 amps.

2N3055 is a transistor for audio amplifiers and linear power supplies. It is actually a terrible choice for an ignition coil driver (trust me, I have LOTS of experience with them).

Look for a power MOSFET in a TO-220 or TO-247 case. Get a high voltage rating - 200V or more should be the minimum. You don't want to snub or limit the di/dt spike on the primary, because it's the inductive kick that is giving you the HV spike on the secondary. Choose a high voltage mosfet that will withstand the inductive kick that will be seen on the primary (500V is a safe choice). I have scoped the drain voltage of a MOSFET used as an ignition coil switch and it frequently is in the 200V range. There are several good choices on Digikey and Mouser; if you want me to pick out a specific one for you I can. Just give me a max price, but remember it is better to buy one or two $10 transistors and never have them blow than to buy a dozen $3 transistors and have to keep replacing them.

For keeping the 555 from frying, I would use an RC snubber across the power/ground rails. You could also use a transient voltage suppressor zener (TVS zener) or a metal oxide varistor (MOV), but a proper RC snubber is more robust and likely cheaper.

EDITED TO ADD:

Here is a good reasonably priced part from Digikey.

Digi-Key - IRFP460PBF-ND (Vishay IR - IRFP460PBF)

**broken link removed**

This is the circuit I'm using. It's a common schematic I've seen a few places online. The coil I'm using is a low impedance coil (external resistor required). Should I place the resistor between my transistor and the coil or the coil and the power, and if so, how many ohms? For the snubber, would a 100 ohm resistor and a .01uF cap work ok? I'd prefer the snubber to the varistor since I have tons of caps and resistors lying around and I live in the middle of nowhere and it's an hour drive to a radio shack from here! Thanks again to everyone!

I've put an ammeter on the power rail (12v) and I'm pulling 4 amps, plus or minus a fraction of an amp depending on the frequency.

marfire said:

Thanks for the replies, and sorry for the delay. I had to build a new one to test whether it was getting fried on on or off, and it's off. Since that's the case would the cap help?

What transistors would you recommend to power an ignition coil? The first time I tried it I used a TIP3055 and that went up in smoke after about three seconds. Every schematic I've seen suggests the 2n3055's(from radio shack), so that's what I've been using.

Click to expand...

If it's just on off, the cap would not help. You need a snubber. FOr example, an RC snubber plaed across the ignition coil (though I suspect it may also hamper performance which is what Speakerguy is saying). SO I guess the only solution is to get a more rugged transistor.

I built another one tonight, and I manged to fry it during operation so I'm going to take everyone's advice and add the snubber to the power/ground rail and get the more rugged transistors. Thanks!

To be more specific, I fried the transistor and not the 555 during operation.

marfire said:

I built another one tonight, and I manged to fry it during operation so I'm going to take everyone's advice and add the snubber to the power/ground rail and get the more rugged transistors. Thanks!

Click to expand...

THe snubber might limit performance since it's trying to supress the very thing that is producing the spark. Althought, one across the power-ground rail will not affect it nearly as much as one in parallel with the ignition coil (if it does anything at all).

oh.. That P.O.S. schematic! The dumbass that published it should be hooked up to the correct version of it and fried for a few minutes! I have seen this one countless times!
Or one so close to it I cant tell the difference from memory.

I have done several dozen 6 volt to 12 electrical conversions on tractors, trucks, old viehicals and assorted farm machinery over the years. I often put a 4 ohm 10 watt resistor between the 12 volt source and the coil if I dont have a internal resistor type coil available. To greatly help out the "inductive snap" put the capacitor parallel to the switching transistor.

This is how the old points and condensor ignition systems are set up. Most of the condensors I have checked over the years run anywhere from around 2 uf to as much as 20 uf. most are in the 2 to 6 uf range.

Use an old condensor from a lawn mower or some old points type ignition system if you have one available. Mylar non polars work great too. just make sure they have 200 volt or higher rating. Playing with the capacitor value will help you tune it to get the best sparks.
I would put a diode in the line feeding all the 555 control stuff with a capacitor on the 555 side.
This blocks reverse voltage spikes and helps isolate the controls from the 12 volt power. I would also put a diode and capacitor across the whole switching end too from the input of the coil to the output of the switching transistor. This gives a bypass path for the inductance snap to follow and lets the coil and condensor power have a good path to follow when they resonate.

And definitely toss the 2N3055 transistor and use a Mosfet or IGBT or at least a high gain darlington with a 200 volt 10 amp or higher rating!

um... turn the coil around. If your using the schematic shown your firing it the wrong way! wierd as it may sound this helps alot on this circuit.
The primary of the coil is in series with the secondary not parallel as most seem to think.

When properly done this circiut can blow sparks right out the top of some coils outputs and back down to the side terminals!

Just like in the picture I posted earlier! Thats one of my spark toys. I am running back to back coils one is forward firing the other is reverse firing. (180 degrees out of phase) I have 6 volt coils with 24 volt sources and no resistors. the switches are 600 volt 50 amp IGBT's and the circuit is set up for adjustable dwell (Switch ON/OFF time) and frequency! Very tunable and can vary the spark size and intensity greatly!

Once you have your sparker set up and working properly feed an audio signal into pin 5 on the 555. If done properly your sparker will sing for you!

Here is a fixed schematic of what IMO you should try. I never found that the coils I was playing with needed a resistor, so you might try it without one. In the picture I've changed the 2n3055 to an N-channel FET like I linked to above, corrected the polarity of the ignition coil (that was a HUGE mistake in that picture), and drawn in the snubber. The arrow points to where you will see a large (>100V and maybe a lot more) spike when you turn off the MOSFET. Don't snub this. Snubbing it allows current to flow in the primary and slows the collapse of the magnetic field inside the transformer. The faster the field in the transformer collapses, the higher the voltage you get on the secondary.

Also, tcmtech made a good point. Ignition coils are autotransformers, not your regular type transformers. The primary and secondary coils are the SAME COIL, it's just that the primary is taken from a tap after a small amount of the total turns. They share a common (-) terminal.

Thanks everyone for the help! I just ordered all the parts I need so I'll post an update and hopefully pictures in a week or so of a lot of sparks.

Speakerguy79 is right on the snubbing capacitor part but yet when you take that capacitor and put it across the switch you get a much more powerful spark! IF its the right value. Too small and it hurts it too big and it hurts it.
Check out the points and condensor setup on any old viehical or farm tractor. That condensor capacitor is always in parallel with points. Try taking it out once and see if your engine even starts with out it! Thats how weak your spark will get with out it.
That capacitor is not owrking as a snubber in a ignition circuit its working as the C in a LC tank. It doesnt snub, it "rings" the inductive collapse.
In this circuit you are replacing mechanical cam driven points with an electronic switch and triggering timer.

That capacitor is not owrking as a snubber in a ignition circuit its working as the C in a LC tank. It doesnt snub, it "rings" the inductive collapse.
In this circuit you are replacing mechanical cam driven points with an electronic switch and triggering timer.

Click to expand...

Interesting, I'll have to try this. It makes sense, but finding the right value might be tricky. I'm guessing a film type is the best for this?

Yea the mylars or ac line snubbers work good but every coil I have ever played with seems to have a different ideal value. One that worked super on one old ignition coil was a way off on the next one. If you ever did or checked out the replacement coils for old equipment years ago they always supplied you with a matching condensor for that coil. now they dont! they just sell you coil and let you sink or swim. Most never look at the spark from the new coil to see if its any good or not.
I tuned a badly done 6 to 12 volt conversion for a good friend of mine. He had the new 12 volt coil but still had the old 6 volt condensor in the distributor. His tractor was hard to start and ran poor. The new coil could hardly make an 1/8 inch spark between the plug wire and engine block. I brought along several different size condensors to try and tune it for him. Now it jumps a spark right out the coil top and down to the coil terminal lugs! Hurts like hell now too! But it starts every time and runs super!

That makes perfect sense about the cap being the C in an LC tank, but it seems like finding the right size could be a real tedious process. Is there any ballpark the caps tend to be in as far as farads, or are they just all over the place? It's not hard to calculate the size cap you'd need if you know the inductance of your coil, so I guess maybe it's time to order an LC meter.

Most are within the 2 to 6 uf. Although being that you dumping some of that resonate energy making the sparks I think that would toss a variable R into the equation.
Trial and error. Start low and add is what I do.
Most of the old condensors I have tested seem to hold a .5 uf jump from one size to the next.
2, 2.5, 3, 3.5, 4 with an actual slight plus or minus variance of course!
But dont be surprized if you are using an old cheepie mystery coil you could hit well over 10 uf before it works better.
I pulled one off an old 4 cylinder mystery engine at the scrap yard many years ago. The condensor was mounted external of the distributor and was the size of a D battery.
My digital meter put it at 24uf!