02GF74
|
| posted on 23/3/07 at 11:37 AM |
|
|
DIY megjolt - spec for megajolt output
going on from this:
quote:
The Megajolt box just looks at the engine speed and load (via a MAP sensor or throttle pot) and determines what advance is required from the
ignition map programmed into it.
It then sends this information to the EDIS module encoded as a pulse of variable width over a simple interface.
anyone have info about the pulse data and the interface?
How difficult would be to make a simple/cheap mega jolt using a PIC?
All this will do is use crank speed to advance the timeing, much like a vacuum less distributor that one fits when one has twin side draiught carbs.
|
|
|
|
|
Doofus
|
| posted on 23/3/07 at 11:43 AM |
|
|
Have a look here
I'm sure I've seen an EDIS/PIC controller somewhere before just can't remember where.
Paul.
|
|
|
MikeRJ
|
| posted on 23/3/07 at 12:51 PM |
|
|
quote:
Originally posted by 02GF74
How difficult would be to make a simple/cheap mega jolt using a PIC?
Quite easy, you just need at least one hardware timer, idealy 16bit and preferably with at least one capture/compare unit.
The micro has to capture the period of a waveform called PIP which is output from the EDIS, and use this to derive RPM. The pulse sent back to the
EDIS is called the SAW (spark Advance Word), and the length of this pulse determines the spark advance. The timing of the start of SAW pulse is
important though.
quote:
All this will do is use crank speed to advance the timeing, much like a vacuum less distributor that one fits when one has twin side draiught carbs.
You mean exactly what Megajolt does if you use it with a 2D map. Sounds like you want to clone the design but using a PIC instead of the 68HC908
which is perfectly doable, but I have to wonder what the point of it would be. Why re-invent a perfectly good, and well proven wheel?
|
|
|
02GF74
|
| posted on 23/3/07 at 01:29 PM |
|
|
quote:
Originally posted by MikeRJ
Why re-invent a perfectly good, and well proven wheel?
I supposed it is for the same reason we build our own cars instead of buyng ready made caterhams?
From what I am hearing, a £ 1.50 PIC is all it needs as far as hardware goes, the programming will require some head scratching though.
Is the source for mj out there somehwere?
|
|
|
MikeRJ
|
| posted on 23/3/07 at 05:41 PM |
|
|
quote:
Originally posted by 02GF74
quote:
Originally posted by MikeRJ
Why re-invent a perfectly good, and well proven wheel?
I supposed it is for the same reason we build our own cars instead of buyng ready made caterhams?
To make life difficult for ourselves?  You are (presumably) building your car to a published design, but sourcing the materials yourself. You can
build megajolt the same way, and then have lots of people to speak to if it doesn't work.
quote:
From what I am hearing, a £ 1.50 PIC is all it needs as far as hardware goes, the programming will require some head scratching though.
Is the source for mj out there somehwere?
Yes, but the assembly source is for a £1.50 68HC908 instead. 
Have a look at the Megajolt website, it's freely available to download as it's an open project.
http://www.autosportlabs.net/asl_dist/mjlj/mjlJunior_src_3.0.0.zip
Note that you will still need the majority of the rest of the bits that make a megjolt, the RS232 interface, the voltage regulation, the
buffering/protection components for the I/O lines etc. You could cost reduce it a (very) little by removing the RPM activated outputs for shift
lights etc. and you won't need the filtering components for a throttle pot if you only want a 2D system (though it's only a couple of
resistors and a cap IIRC).
As an exercise in learning and playing with PIC's then great, it should be an interesting project. If you are doing it to save money, then
it's a non-starter unless you put a negative value on your time!
I think the time could be better spent improving the existing code, maybe adding multiple spark support (which the EDIS is supposed to support).
|
|
|
PeterW
|
| posted on 23/3/07 at 08:28 PM |
|
|
Isn't the holy grail of this MS/MJ stuff to find a replacement for EDIS..?
Anyone actually worked out what it does..?
I know MSII can run its own ignition drivers, how long will it be before Megajolt can..??
Cheers
Pete
|
|
|
MikeRJ
|
| posted on 24/3/07 at 12:34 AM |
|
|
quote:
Originally posted by PeterW
Isn't the holy grail of this MS/MJ stuff to find a replacement for EDIS..?
Anyone actually worked out what it does..?
EDIS does all the complex parts of driving the ignition: trigger wheel interfacing and decoding, time to angle conversion, dwell calculation, coil
driving etc.
A system that does not rely on the EDIS is a nice idea, but it will be considerably more complex, probably more expensive (depending on how much you
pay for the EDIS unit) and quite possibly less robust depending on design and implementation. The EDIS system used in conjunction with a MJLJ box
also has some fail safe measures built in, which would significantly complicate a discrete design if it were to be implemented.
None of this matters if you have an engine configuration that MJLJ does not support though, e.g. odd number of cylinders or uneven firing angles.
|
|
|
TangoMan
|
| posted on 29/3/07 at 10:04 PM |
|
|
Not sure about Megajolt but Megasquirt can run without Edis.
It can use vb921 direct coil drivers to fire the coils if you want to do away with Edis.
Summer's here!!!!
|
|
|
02GF74
|
| posted on 12/4/07 at 08:46 AM |
|
|
quote:
Originally posted by MikeRJ
Have a look at the Megajolt website, it's freely available to download as it's an open project.
http://www.autosportlabs.net/asl_dist/mjlj/mjlJunior_src_3.0.0.zip
linky no worky - what is the megjolt webiste - when I google for it, I only seem to be able to find megasquirts 
|
|
|
MikeRJ
|
| posted on 13/4/07 at 02:40 PM |
|
|
Link works fine for me?
The megajolt web site is http://www.autosportlabs.net
|
|
|
