I just had this idea.
A stock 2nd gen Mazda CAS has all the required parts for a Lamar
ignition trigger mechanism. Namely, two magnetic sensors and a
sprocket like wheel.
The sensors need to be mounted 90 degrees to each other and half the
teeth removed from the sprocket like wheel.
Some of the floor of the housing will have to be removed to allow
both sensors to be mounted at the sprocket level..
A new steel bracket will be made to hold the magnetic sensors 90
degrees to each other.
More when I get a chance to do some 3D's.
The tricky part comes when it is realized the 2nd gen CAS runs at
half e-shaft speed :)
Does anybody have a suggested solution? There is a hint above:)
Paul Lamar
Paul,
Could you use the sensors as they are? My thought is to use the
home sensor as an interrupt to 'switch' the set of plugs you are
firing. You'd likely need to build another small circuit to do the
switching.
-Josh Jacobs
I'll have to think about that.
Got any circuits in mind?
A data selector could be part of it. See the attached jpg.
Toggling pin 11 would do it.
I think a flip flop would be needed as well.
Sync might be a problem. If it gets out of sync it would spark the
wrong rotor.
Paul Lamar --
Good point about the sync problem. In that case, a logic controller
may be the best bet. You could cut off one of the home switch teeth
so that it is your main reset interrupt. Your logic program could
count the pulses from the trigger wheel. It would flow like this:
Interrupt; at count <6, use output 1; at count 6 use output two;
interrupt... at 10k rpm the frequency of the trigger pulses will
only be 4khz, so finding a fast enough chip shouldn't be too hard.
Josh
Yes a computer would work but the whole idea here is to avoid using
a computer. Keep it simple and easy to understand. One CAS per
rotor.
Here are the latest 3D's. It will fit in the stock housing with a
little clearancing here and there using a die grinder.
Paul Lamar
Paul,
I have been thinking and researching this a good bit. I've come
across a circuit that uses two 4017 decade counters to count higher
than 10. This way, you could count all 12 teeth between interrupts
with no computer needed. I am working on a circuit in LTSpice to
see if I can make it work, but it is slow going since I am teaching
myself as I go.
The problem is that output 1 stays on until output two is triggered
and so on. My thought is to use the outputs from the 4017's to turn
on NPN transistors that can then pass the CAS signal.
In other words, when the count is 1-6, NPN1 remains on and passing
the signals to ignition 1. When the count is 7-12, NPN2 is on and
passes the CAS signal to ignition two. Then you hit the reset from
the home sensor and it all starts over.
It sounds more complicated than it is, I think. The advantage of
this is that you do not have to modify your CAS at all.
-Joshua Jacobs
If you get it to work I will build one and try it.
Paul Lamar
Paul,
I'm getting close on this circuit. I've come across a few ways to do
it, but I'm sticking with the two decade counters for now, mainly
because I can't find LTspice models for the binary counter and decoder
chips that I would like to use.
I need to know if the home sensor align exactly with the other trigger
teeth or if they are offset a bit. It's hard to tell in the photos
you've sent out so far.
-Josh
See the previous message.
I have done away with VR sensors. I am using hall effect sensors.
The advantage is the same circuit in the system can be computer triggered
or triggered by my carb set up unit bolted to the front of the e-shaft.
I only need one circuit design. Patent pending.
Paul Lamar
The Rotary Engine News Letter. Powered by Linux.
ACRE NL web site.
http://www.rotaryeng.net
You Tube
http://tinyurl.com/beqqxas
Copyright 1998-2014 All world wide rights reserved.