Maker Pro
Maker Pro

Old tractor ignition upgrade problem

NMNeil

Oct 3, 2014
111
Joined
Oct 3, 2014
Messages
111
So here's my project
I have a 1949 Ford 8n tractor, 4 cylinder side valve with 6volt electrics, which the previous owner converted to 12volts and used a regular GM alternator instead of the generator. This is a common conversion but leads to problems with the ignition.
The distributor is on the front of the engine and has the 6volt coil integrated, like this:


frntmnt1.jpg


That's the coil balanced on top.

When it's converted to 12 volt a different ballast resistor is put in the circuit to keep the current down to
about 6 amps through the coil. As it wasn't the most reliable system to begin with I have decided to bring the
ignition system into the 21st century.

Simple enough idea, replace the points with 2 digital hall sensors placed 90 degrees apart and activate them
with 2 magnets on the old points cam placed 180 degrees apart. This means hall sensor 1 will switch at 0 and
180 crank degrees and hall sensor 2 at 90 and 270 crank degrees.
These signals will be used to activate 2, N channel IGBT's, such as the Fairchild ISL9V3040D3S which is
purpose built as an ignition coil driver and needs no external components. From there to either a Ford Ranger
or Dodge Neon coil pack which both use the wasted spark principal.
My problem is, what if the magnet stops opposite one of the hall sensors and keeps the IGBT 'open' sinking a
high amperage and cooking the coil until the engine is cranked?
Can anyone suggest a circuit that would turn off the IGBT after say, 20 milliseconds of inactivity (maximum
dwell will be around 5 milliseconds) but re-active it when it received a new pulse from the hall sensor?

Sorry, still working on posting circuit diagrams.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Sounds interesting. I'm looking forward to seeing the diagrams.
 

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
I built a similar system here for use on model engines.
Problem was as you outlined.
Another aspect was I needed to install several magnets directly behind eachother to get enough dwell time.
Went to CDI after that.
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
You could use a monostable multivibrator (monoflop) built e.g. around an NE555 timer circuit. This monoflop serves a dual purpose:
  1. It will create one single pulse for every pass of the magnet by the hall sensor. If the magnet stops opposite the hall sensor, the monoflop will still issue only a pulse, not a constant signal (make sure you do not use any of the retriggerable modifications).
  2. It will ensure a consistent pulse length for driving the IGBT regardless of the rpm of the motor ( a fast running motor will make the magnet pass faster by the sensor and the sensor's output will be a shorter pulse than for a slow running motor) because the pulse length is now defined by the monoflop's time constant, not the rpm of the motor.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
FYI a "monoflop" is normally called a "monostable" or "one-shot".
 

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
Problem with the monostable is the dwell is the same regardless of what the revs are.
 

NMNeil

Oct 3, 2014
111
Joined
Oct 3, 2014
Messages
111
After reviewing the potential problems, such as the precision required for the magnet placement and the different Gauss from one magnet to another, which will affect the point of triggering for the Hall sensors, and the problem with the dwell time I decided to use an Arduino. This should allow me to set an exact dwell time, adjust for magnet position etc.
Trouble is, I have an Arduino Uno in front of me, but no idea how to use it.
Going to have to dust off the learning side of the old brain:eek:
 

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
As they say, if you think you have problems now , wait until................ :)

There is a post here with a reference to a microcontroller system that works fine.
Runs via a hall effect pickup to a 12F683 with an excel spreadsheet for timing table.
Small drawback is you have to re-program if you want to change the timing curve but easy to do.
It comprises a "timer board section" and a "high voltage section" for CDI but the former might get you started.
Shouldn't be too much work to convert over to operate a standard kettering ignition coil via say an MJ10012 darlington.

https://www.electronicspoint.com/th...or-a-good-project-or-kit.268300/#post-1626305
 

NMNeil

Oct 3, 2014
111
Joined
Oct 3, 2014
Messages
111
As they say, if you think you have problems now , wait until................ :)

There is a post here with a reference to a microcontroller system that works fine.
Runs via a hall effect pickup to a 12F683 with an excel spreadsheet for timing table.
Small drawback is you have to re-program if you want to change the timing curve but easy to do.
It comprises a "timer board section" and a "high voltage section" for CDI but the former might get you started.
Shouldn't be too much work to convert over to operate a standard kettering ignition coil via say an MJ10012 darlington.

https://www.electronicspoint.com/th...or-a-good-project-or-kit.268300/#post-1626305

Thanks for that.
Timing isn't much of an issue as this thing only hits 2000 RPM flat out and it still has the original centrifugal weights in the distributor.
 
Top