Searching \ for '[EE]: Ignition Coil Flyback Circuit (Was MOSFet Tr' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=ignition+coil+flyback
Search entire site for: 'Ignition Coil Flyback Circuit (Was MOSFet Tr'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Ignition Coil Flyback Circuit (Was MOSFet Tr'
2005\04\05@053545 by Milosz Kardasinski

picon face
Well, I took a couple of suggestions and applied them to my circuit,
good news is that my fets aren't dying. Bad news is I've killed two
555 timers with the circuit...they fired the coil for a couple of
seconds then died. Any ideas as to the culprit?

Here is the schematic:
http://www.no-bling.com/image_galleries/galleries/goat_Built/PAC.gif

Cheers,
Milosz K.

2005\04\05@055217 by Thomas Sefranek

face picon face
I see no decoupling on the +12 to the 555.
You WILL get spikes from the coil on that line.

 *
 |  __O    Thomas C. Sefranek   spam_OUTWA1RHPTakeThisOuTspamARRL.NET
 |_-\<,_   Amateur Radio Operator: WA1RHP
 (*)/ (*)  Bicycle mobile on 145.41, PL 74.4

hamradio.cmcorp.com/inventory/Inventory.html
http://www.harvardrepeater.org  

> {Original Message removed}

2005\04\05@060039 by Michael Rigby-Jones

picon face


>-----Original Message-----
>From: Thomas C. Sefranek
>Sent: 05 April 2005 10:50
>To: 'Microcontroller discussion list - Public.'
>Subject: RE: [EE]: Ignition Coil Flyback Circuit (Was MOSFet Troubles)
>
>
>I see no decoupling on the +12 to the 555.
>You WILL get spikes from the coil on that line.


I absolutely agree, supply decoupling is most important for the 555's
(thought the CMOS versions are less critical).  Years ago I had a couple
of 555's literaly blow up (plastic cases had craters in the top of
them!) and the cuprit was lack of decoupling causing some seriously wild
oscillation.

I would be tempted to run the 555 from some kind of regulator, even a
simple zener arrangement (well decoupled) would be good.

Regards

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2005\04\05@063806 by Brent Brown

picon face
> Well, I took a couple of suggestions and applied them to my circuit,
> good news is that my fets aren't dying. Bad news is I've killed two
> 555 timers with the circuit...they fired the coil for a couple of
> seconds then died. Any ideas as to the culprit?
>
> Here is the schematic:
> http://www.no-bling.com/image_galleries/galleries/goat_Built/PAC.gif

Probably what others have said about supply decoupling. Might be a long
shot, but also possible to get capacitive feedback from D-G of MOSFET
(Miller capacitance IIRC). Thinking about it this would probably show up by
killing the MOSFET though. If you are curious, check output pin of dead vs
live 555 with ohmeter, both polarities, and compare values to see if output
stage is what got fried.

--
Brent Brown, Electronic Design Solutions
16 English Street, Hamilton 2001, New Zealand
Ph: +64 7 849 0069
Fax: +64 7 849 0071
Cell/txt: 025 334 069
eMail:  .....brent.brownKILLspamspam@spam@clear.net.nz


2005\04\05@065212 by Russell McMahon

face
flavicon
face
>I see no decoupling on the +12 to the 555.
> You WILL get spikes from the coil on that line.

Yes. 555 and coil must be separated electrically.
At the least a series R to the 555 V+ with a say 100 uF to 1000 uF
electrolytic on both the 555 side AND on the main 12V rail as well.

Also, it would be an excellent idea to add a small series resistor
from 555 to FET fgate (say 22R - 100R range and add a 15v zener from
gate to ground (cathode to gate) mounted electrically near the FET
with shortest possible leads. Miller capacitance will couple drain
spike into gate. This may destroy the FEt and may even be what is
killing the 555.

With the FET you are using you need 10v gate drive at full rated
current BUT at the currents you are probably using 8v is fine. If you
add a say 9v1 or similar zener to ther 555 supply and a suitable
resistor to supply the 555 and the caps as above it should be a good
start. 555 current unknown but say you needed 20 mA then serries R is
about (12-9)/0.020 = 150R. Say 100r.
P resistor is 3^2/100 = 90 mW = minimal
P zener = 30 mA x 9v = 270 mW -> use 1 watt or maybe 500 mW zener.


       RM


2005\04\05@070211 by Alan B. Pearce

face picon face
> Well, I took a couple of suggestions and applied them to my circuit,
> good news is that my fets aren't dying. Bad news is I've killed two
> 555 timers with the circuit...they fired the coil for a couple of
> seconds then died. Any ideas as to the culprit?

As well as spikes on the supply line that others have mentioned, you are
probably getting voltage spikes fed back into the 555 output from the FET
due to capacitance between gate and drain. This will be significant on a
power fet, and with the voltage rise times on the drain will produce
significant spikes back into the 555.

2005\04\05@103536 by Milosz Kardasinski

picon face
I didn't think that decoupling was really necessary in this case, I
guess I need to get into the habit of always decoupling, no matter
what. I will also put a 150ohm resistor between the 555 and fet gate
as well as one in series to the V+ of the 555.

Miller capacitance is new to me, I guess I some reading to do tonight.
Is this a trait all fets posses? Is the miller effect more pronounced
in certian circumstances?

Cheers,
M.

2005\04\05@103603 by Milosz Kardasinski
picon face
I didn't think that decoupling was really necessary in this case, I
guess I need to get into the habit of always decoupling, no matter
what. I will also put a 150ohm resistor between the 555 and fet gate
as well as one in series to the V+ of the 555.

Miller capacitance is new to me, I guess I some reading to do tonight.
Is this a trait all fets posses? Is the miller effect more pronounced
in certian circumstances?

Cheers,
M.

2005\04\05@131445 by olin_piclist

face picon face
Milosz Kardasinski wrote:
> I didn't think that decoupling was really necessary in this case, I
> guess I need to get into the habit of always decoupling, no matter
> what.

That's a good idea, but I don't think lack of decoupling is your primary
problem in this instance.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

2005\04\05@153724 by Danny Decell

picon face
Maybe I will learn something here too. First I will qualify this by saying I
don't know squat about analog circuits but what happens when that magnetic
field collapses from the ignition coil. Does that setup some reverse voltage
across your 555? Is that what the diode D1 is suppose to protect and if so
why is C2 there? Does that form some RC discharge of roughly 1k * .82uf * 5
of like 4ms or something? Why? Can't the diode go straight to ground with R1
there? Is the 555 getting any of that reverse polarity while R1,D1,C2 are
trying to escort T1's backlash to safety? Or am I just all mixed up because
I've been programming all morning?


{Original Message removed}

2005\04\05@181254 by Thomas Sefranek

face picon face
You can get a simple understanding of the Miller effect by this:

You are driving the gate of the FET high. (relatively small change)
The output (Drain) is going LOW (relatively big change!).
The gate to drain capacitance couples the change in voltage of the drain
back to the gate, canceling some of the charge your are driving to the gate.

 *
 |  __O    Thomas C. Sefranek   WA1RHPspamKILLspamARRL.NET
 |_-\<,_   Amateur Radio Operator: WA1RHP
 (*)/ (*)  Bicycle mobile on 145.41, PL 74.4

hamradio.cmcorp.com/inventory/Inventory.html
http://www.harvardrepeater.org  

> {Original Message removed}

2005\04\05@182103 by Milosz Kardasinski

picon face
On Apr 5, 2005 3:37 PM, Danny Decell <.....d.decellKILLspamspam.....comcast.net> wrote:
> Maybe I will learn something here too. First I will qualify this by saying I
> don't know squat about analog circuits

That makes the two of us.

> but what happens when that magnetic
> field collapses from the ignition coil.

The coil has inductance so when the juice is switched off...current
flow wants to remain constant so the voltage rises until the energy
stored in the primary coil is released. If my explaination is
wrong....it's due to statement above.

>Does that setup some reverse voltage
> across your 555?

I hope not...since I'm frying stuff it isn't working like I want it
hence the disscussion.

> Is that what the diode D1 is suppose to protect and if so
> why is C2 there?

R1 & D1 are meant to clamp the flyback voltage around 200V. C2 is a
secondary clamp....now that I think about it I can't find a really
good reason for it being there.


The more I read about the Miller effect the more I see a reason for a
resistor between gate and 555 output, what is a good value here? I
think 1k-10k would be fine because gate sinks very little current so
the voltage drop would be negligible, my reasoning correct?

Cheers,
M.

2005\04\05@182812 by Thomas Sefranek

face picon face
> -----Original Message-----
> From: EraseMEpiclist-bouncesspam_OUTspamTakeThisOuTmit.edu [piclist-bouncesspamspam_OUTmit.edu] On Behalf
> Of Danny Decell
> Sent: Tuesday, April 05, 2005 2:37 PM
> To: Microcontroller discussion list - Public.
> Subject: Re: [EE]: Ignition Coil Flyback Circuit (Was MOSFet Troubles)
>
> Maybe I will learn something here too. First I will qualify this by saying
> I
> don't know squat about analog circuits but what happens when that magnetic
> field collapses from the ignition coil.

Think of the coil as a current flywheel.
You cause a current to flow, then try to shut it off, the coil keeps the
current flowing until it exhausts it's magnetic field.

It does this by REVERSING the voltage across itself!

> Does that setup some reverse voltage across your 555?

Well, yes and no.
The coil can generate ANY voltage necessary to keep the current flowing.
This means BIG voltages in many circumstances.  (Added to the supply
voltage!)

> Is that what the diode D1 is suppose to protect and if so why is C2 there?

> Does that form some RC discharge of roughly 1k * .82uf * 5 of like 4ms
> or something? Why? Can't the diode go straight to ground with R1 there?
> Is the 555 getting any of that reverse polarity while R1,D1,C2
> are trying to escort T1's backlash to safety?

I don't have the circuit memorized, I can't answer specifics.

Normally, there will be a flyback diode across the coil to allow a
controlled current discharge path.  Without it, other paths are possible,
Including charge coupled back thru the drain to gate, and your 555s output
now see's reverse current from what it is driving.

The real problem is the 12 volt line gets NOISY every time the coil reacts.
You do not have a noise decoupler so the 555 could be seeing all kinds of
voltages.  And if the output is driven one way and the power pin another...
Magic smoke released...

As you have already been advised, limit the current from/to the gate with a
resistor, provide an RC decoupling to the power pin from the +12 volt line.

> Or am I just all mixed up because I've been programming all morning?

Being mixed up while programming is (sometimes) normal.
(In MY case it is a prerequisite.)

Tom

 *
 |  __O    Thomas C. Sefranek   @spam@WA1RHPKILLspamspamARRL.NET
 |_-\<,_   Amateur Radio Operator: WA1RHP
 (*)/ (*)  Bicycle mobile on 145.41, PL 74.4

hamradio.cmcorp.com/inventory/Inventory.html
http://www.harvardrepeater.org


2005\04\05@183228 by Thomas Sefranek

face picon face



> -----Original Message-----
> From: KILLspampiclist-bouncesKILLspamspammit.edu [RemoveMEpiclist-bouncesTakeThisOuTspammit.edu] On Behalf
> Of Milosz Kardasinski
> Sent: Tuesday, April 05, 2005 5:21 PM
> To: Microcontroller discussion list - Public.
> Subject: Re: [EE]: Ignition Coil Flyback Circuit (Was MOSFet Troubles)
>
> On Apr 5, 2005 3:37 PM, Danny Decell <spamBeGoned.decellspamBeGonespamcomcast.net> wrote:
> > Maybe I will learn something here too. First I will qualify this by
> saying I
> > don't know squat about analog circuits
>
> That makes the two of us.
>
>
> The more I read about the Miller effect the more I see a reason for a
> resistor between gate and 555 output, what is a good value here? I
> think 1k-10k would be fine because gate sinks very little current so
> the voltage drop would be negligible, my reasoning correct?

You would have a SLOW response, the FET would be LONGER in it's linear
(HOT) mode.  T=RC.  C is the input capacitance (Yes, Miller and all...)
So I'd keep it to 100 ohms or less.

>
> Cheers,
> M.
> --
Tom

 *
 |  __O    Thomas C. Sefranek   TakeThisOuTWA1RHPEraseMEspamspam_OUTARRL.NET
 |_-\<,_   Amateur Radio Operator: WA1RHP
 (*)/ (*)  Bicycle mobile on 145.41, PL 74.4

hamradio.cmcorp.com/inventory/Inventory.html
http://www.harvardrepeater.org  

2005\04\05@225328 by Russell McMahon

face
flavicon
face
>> The more I read about the Miller effect the more I see a reason for
>> a
>> resistor between gate and 555 output, what is a good value here? I
>> think 1k-10k would be fine because gate sinks very little current
>> so
>> the voltage drop would be negligible, my reasoning correct?

> You would have a SLOW response, the FET would be LONGER in it's
> linear
> (HOT) mode.  T=RC.  C is the input capacitance (Yes, Miller and
> all...)
> So I'd keep it to 100 ohms or less.

Yes.

Gate capacitance is usually in the low nF range and this must be
charged and discharged by the gate drive cct.
As I suggested a few posts back - something in the 10R - 100R range
AND a clamp zener from gate to ground mounted near the FET. This
prevents both gate high overshoot due to miller capacitance and
undershoot ringing. The latter can be significant and damaging.
Purists frown on the FET and use a Schottky diode from gate to ground
to prevent negative going ringing - but I have had excellent success
with a zener. I have a design where the lifetime without the zener is
minutes at best.

The gate drive CURRENT usually wants to be in the 100's of mA to low
amps range (!) - hence the low gate resistor. The current only flows
at turn on & off. At 10's of kHz you get a few mA mean drive current.
With 10R its about 1 Amp peak with 10V drive. With 100R it's 100 mA.



       Russell McMahon




2005\04\06@044956 by ISO-8859-1?Q?Ruben_J=F6nsson?=

flavicon
face
Hi,

You may also try a CDI (capacitor discharge ignition) circuit. Here you first charge a capacitor across the inductor and discharge fet. The discharge fet is  switched off during this time. When the capacitor is fully charged it is discharged through the primary coil winding and the discharge fet. Put a (fast schottky) diode across the discharge fet and keep the discharge fet switched on until the capacitor is fully discharged. The capacitor and the coil wil form an oscillation circuit and will oscillate and fade out untill the capacitor is fully discharged. The initial pulse and the oscillation will produce the high voltage on the secondary winding of the transformer.

To charge the capacitor, you need a high side switch (charge switch) such as a p channel fet or a combination of an n channel fet/igbt with a charge pump to produce the higher voltage needed to switch it on (a high side driver chip will do that). Also, if you put an inductor after the charge switch, before the capacitor, the capacitor will be charged smoothly and it will be charged to a voltage roughly twice as high as your powersupply, which will give you even more energy for the spark. Because of this you will need another (fast schottky) diode in series with the coil. The smooth charging will also let you charge the capacitor only partly in order to reduce the high voltage output (most useful if you build a variable high voltage powersupply).

Instead of the 555 you can control the pulse timings for the charge and discharge switch with a small PIC, perhaps even the 10F series. Turn charge switch on. Turn it off and wait for the coil to finish charging the capacitor. Turn discharge switch on until the capacitor is fully discharged. Repeat.

The beauty of it all is that the charge switch (and the diode) isolates the discharge circuit from the powersupply (and the rest of the electronics) when the discharge fet is discharging the capacitor through the coil.

You also, as has been discussed here, need the resistor in series with the gate for the discharge switch plus a zener between ground (source) and gate and perhaps even a zener between gate and drain.

I have also successfully used a triac instead of a fet as a discharge switch (with higher voltages though).

I have primarily used this circuit to produce high AC and DC voltages and never tried it for ignition.

I have found that using a capacitor with a low value for the loss factor (tan fi?) such as a polypropylene type will be most efficient. The fading oscillation will last longer which is good especially if there is a diode/capacitor multiplier on the secondary side of the transformer (DC power supply).
Good luck and watch out for those sparks / Ruben


> Well, I took a couple of suggestions and applied them to my circuit,
> good news is that my fets aren't dying. Bad news is I've killed two
> 555 timers with the circuit...they fired the coil for a couple of
> seconds then died. Any ideas as to the culprit?
>
> Here is the schematic:
> www.no-bling.com/image_galleries/galleries/goat_Built/PAC.gif
>
> Cheers,
> Milosz K.
> -

More... (looser matching)
- Last day of these posts
- In 2005 , 2006 only
- Today
- New search...