Need help with this point saver circuit

Thread Starter

sdowney717

Joined Jul 18, 2012
711
http://home.comcast.net/~loudgpz/GPZweb/Ignition/GPZheiModForPoints.html
Simply that I cant get it to work.
Trying to improve my Onan generator ignition.
I have two 4 pin GM ignition modules, they both work, I tested them in a GM distributer, they make a spark.

I dont have the 2N2222 NPN transistor, I do have c3205 NPN transistors

Any ideas of why, is it the transistors are not 2N2222?
They test ok in my meter transistor tester.
My transistors look like these


I wired it as Emitter on top, collector in middle, Base on bottom, is this correct?
 

Reloadron

Joined Jan 15, 2015
7,501
Lying on the curved side looking down at the flat of 2N2222 the pin out is EBC (Emitter Base Collector) the C3205 has a different pin out of ECB (Emitter Collector Base) have you taken that into consideration?

Ron
 

ian field

Joined Oct 27, 2012
6,536
http://home.comcast.net/~loudgpz/GPZweb/Ignition/GPZheiModForPoints.html
Simply that I cant get it to work.
Trying to improve my Onan generator ignition.
I have two 4 pin GM ignition modules, they both work, I tested them in a GM distributer, they make a spark.

I dont have the 2N2222 NPN transistor, I do have c3205 NPN transistors

?
One problem you can run into with a TAC module is insufficient "wetting" current through the points, the point face get oil spray, tarnish etc and insulate. The bike I was designing for had a weedy alternator and couldn't spare any extra current for points wetting.

The solution I came up with was to use a high voltage power MOSFET in grounded gate configuration (gate tied to +12V) - drain to the coil LT terminal and source to the points. The source current is equal to the drain (and coil) current, so the requirement for wetting current was satisfied, but the voltage when the points opened could never exceed 12V - so no arcing/pitting.

The grounded gate configuration makes a seriously fast switch, and there is virtually no energy dissipated when the drain switches off as there would have been as the points opened.

The finished TAC cost me an engine - which threw a rod before I realised that I shouldn't red-line in every gear just because I could!

The downside is RDSon not being so good on high voltage MOSFETs, there was no improvement at low RPM and winter starting wasn't neccessarily helped by the low cranking voltage heading down towards VGSthr.
 

Thread Starter

sdowney717

Joined Jul 18, 2012
711
Ian, would your config be using a P channel or N channel mosfet?
I dont have any P channel, I have many N channel here.

Well, I gave up on the circuit in that comcast link as it seems to be defective?
I notice in his, W is on Bat+, while this one W is on ground.

This other configuration does work, the module fires the coil.
Although it is firing on points close, the opposite of what it should do, which is fire on points open.

At least something is working.
Is there a way to invert the on and off to ground?
So that when the points open, a circuit conducts which allows the module to conduct between pin G and ground and the spark fires?
Basically like when a circuit goes open from being grounded (points) , it allows a current to flow from terminal G to ground and the module will fire off the coil.
I think that should be possible, to flip flop it.

This is a working schematic that functions the ignition module so that it fires.



This is my creation for the generator, but as it fires on points closing, the timing will be way advanced, I dont know if I can adjust the tower (it slides) to compensate.
HPIM1287.JPG
 

Thread Starter

sdowney717

Joined Jul 18, 2012
711
I wonder if a PNP transitor could do this switching?
I was looking at this diagram


Imagine If the module is connected as G terminal is +Vcc
The points are connected to the base -Vin (negative input)

I assume that if base goes negative, then transistor connects between Emitter and collector?
So when the points close the circuit with ground, would it be like terminal G going to ground which fires the module off?

Ok, that wont work as it will just fire like it does now when points close. So no good.

I did find this which supposedly will work, It uses a big PNP power transistor, and is simple enough.
http://www.5bears.com/tim4.htm
 
Last edited:

Thread Starter

sdowney717

Joined Jul 18, 2012
711
I got it to work!!!

The comcast weblink schematic is wrong!
Terminal W does NOT got to Battery positive!
All those schematics are wrong. I have been looking at that for couple years and it is wrong.

So after I saw the module working, it got me to think why cant the NPN transistor switch the circuit, so current can flow between E and C.
So I setup the circuit so that I manually shorted emitter to collector and it sparked, then I knew it could work.
So to make this work, set it up as the working diagram with the pullup resistor 1K between terminal G and terminal B and W as ground.

So then to make it work with the added little NPN transistor
Terminal W goes to ground
Terminal G goes to collector
Battery + goes to resistor
resistor goes to base
base goes to points

And when you open the points, it sparks.
 

ian field

Joined Oct 27, 2012
6,536
Ian, would your config be using a P channel or N channel mosfet?
I dont have any P channel, I have many N channel here.

]

It was N-channel, salvaged from the PSU of a 19" PC monitor, rated 9A, 900V.

6A 600V MOSFETs are far more common in smaller monitors and some CRT type TVs. They're easy to connect in parallel for adequate current rating - the more the merrier for a lower collective RDSon. The 600V rating would be tempting fate a bit, but you can protect the MOSFETs with zener diodes - the highest voltage zener I've actually held in my hand was 180V, but I believe you can get higher voltage ones specially for ignition circuits.
 

Thread Starter

sdowney717

Joined Jul 18, 2012
711
It was N-channel, salvaged from the PSU of a 19" PC monitor, rated 9A, 900V.

6A 600V MOSFETs are far more common in smaller monitors and some CRT type TVs. They're easy to connect in parallel for adequate current rating - the more the merrier for a lower collective RDSon. The 600V rating would be tempting fate a bit, but you can protect the MOSFETs with zener diodes - the highest voltage zener I've actually held in my hand was 180V, but I believe you can get higher voltage ones specially for ignition circuits.
Thanks for that, I do have lots of N channel power mosfets. They are rated from 40 to 60 vdc and 20 to 120 amps.
I figure one could work, and would like to test this idea.
Can you share a hand drawn circuit or something, I would like to visualize your circuit.

Here is my module and little board installed in the Onan. It is to the left of the ammeter in the center.
I bypassed the ballast resistor, because with it in circuit and cranking, the module and coil was only seeing 5.5vdc, and that is not enough to work. I ran it for 30 minutes, and the coil stayed cool, and the module got slightly warm. It is running very well.

For the little circuit PCB, I simply cut a piece from a failed computer speaker amp, matching up some traces to fit a transistor and resistor and some wires in, then used some hot glue to embed it, worked out fine.

 
Last edited:

ian field

Joined Oct 27, 2012
6,536
Thanks for that, I do have lots of N channel power mosfets. They are rated from 40 to 60 vdc and 20 to 120 amps.
I figure one could work, and would like to test this idea.
Can you share a hand drawn circuit or something, I would like to visualize your circuit.
Its so simple you don't need a schematic.

Just connect the drain to the coil LT terminal, the gate to +12V and the source to the points.

Just for peace of mind, you could clamp the gate/source with a 20V zener - obviously oriented so its not forward conducting when the gate is positive with respect to source.

When I built mine, I just bolted the MOSFET to an offcut of aluminium sheet, most of the space was taken up by terminal posts for securing leads with connectors that fit the wiring loom.

The 9A MOSFET I had to hand didn't run hot, so I just immersed the assembly in melted candle wax to seal it.
 

Thread Starter

sdowney717

Joined Jul 18, 2012
711
Its so simple you don't need a schematic.

Just connect the drain to the coil LT terminal, the gate to +12V and the source to the points.

Just for peace of mind, you could clamp the gate/source with a 20V zener - obviously oriented so its not forward conducting when the gate is positive with respect to source.

When I built mine, I just bolted the MOSFET to an offcut of aluminium sheet, most of the space was taken up by terminal posts for securing leads with connectors that fit the wiring loom.

The 9A MOSFET I had to hand didn't run hot, so I just immersed the assembly in melted candle wax to seal it.
Thanks, I will try this out and see how it does. I have higher powered ones to use, up to 120 amps. Some IRF 1010E and IRF 3205

I thought about this and was wondering, where is the current flow going to be heaviest?

Mosfets conduct source to drain, or drain to source with the gate controlling the on and off, seems the points will see a lot of current? So how will the current flow in the circuit?

I uploaded what I think your saying, is this right looking to you?
0428150747.jpg
 
Last edited:

ian field

Joined Oct 27, 2012
6,536
Thanks, I will try this out and see how it does. I have higher powered ones to use, up to 120 amps. Some IRF 1010E and IRF 3205

I thought about this and was wondering, where is the current flow going to be heaviest?

Mosfets conduct source to drain, or drain to source with the gate controlling the on and off, seems the points will see a lot of current? So how will the current flow in the circuit?

I uploaded what I think your saying, is this right looking to you?
View attachment 84802
The LT+ of the coil to +12V, drain to LT-, the winding polarities are arranged to put the negative spark potential on the plug center electrode, that bit gets hotter so electrons jump off it easier.

I used a 0.22uF capacitor across the LT winding - never got around to trying it without. On the points, the capacitor slows the dv/dt so the parting points break the arc earlier - but I believe it can also resonate with the LT winding, ultimately ringing delivers a little more spark energy.
 

KG8NS

Joined Jun 17, 2015
1
I got it to work!!!

The comcast weblink schematic is wrong!
Terminal W does NOT got to Battery positive!
All those schematics are wrong. I have been looking at that for couple years and it is wrong.

So after I saw the module working, it got me to think why cant the NPN transistor switch the circuit, so current can flow between E and C.
So I setup the circuit so that I manually shorted emitter to collector and it sparked, then I knew it could work.
So to make this work, set it up as the working diagram with the pullup resistor 1K between terminal G and terminal B and W as ground.

So then to make it work with the added little NPN transistor
Terminal W goes to ground
Terminal G goes to collector
Battery + goes to resistor
resistor goes to base
base goes to points


And when you open the points, it sparks.
Hi SD, I am looking to add a point saver module to my old 16hp Briggs. I converted it to 12v coil ignition and want to add circuit like yours. I have the HEI module. Reading your thread, I understand the signal transistor is to invert the pulse so the timing is correct. Could you post your complete updated schematic using the NPN transistor and resistor? Your wiring description above is confusing to me.

I also found that a Ford TFI fires at the proper timing as is and the inverter in not needed. http://www.advrider.com/forums/showthread.php?t=417099

Thanks
Chuck KG8NS
 
Top