Thinking Of Mega Jolt

[tony perks 1]

I would like to fit Mega jolt System to my rally car, my question is how reliable are they is the componentry up to competion treatment ?, by the way its for a tarmac/ concrete rally car.I already have the Mi16 flywheel fitted i ave the ford coil pack and a set of leads just got to source the edis bit and Fab a loom.

[welshpug 1,621]

should be more reliable than a dizzy for sure, There's a fair bit of torque and drivability to be had with mapped ignition I think, my 205 seems to be a lot smoother now it runs ECU controlled ignition (8v 205 on moronic MP3.1 from a ZX)

[tony perks 1]

should be more reliable than a dizzy for sure, There's a fair bit of torque and drivability to be had with mapped ignition I think, my 205 seems to be a lot smoother now it runs ECU controlled ignition (8v 205 on moronic MP3.1 from a ZX)

Got torque a plenty, i run 45's and a modded dizzy, with tailered curve but a proper full ignition map has to be the way to go.

[bren_1.3 1]

well its all ford o.e equipment, so maybe look into asking some of the rally boys who run zetec engines if they have problems with coil packs and such like...........

[Tom Fenton 1,524]

Note that MegaJOLT will not work with the Mi16 flywheel trigger pattern (60-2).

 

It requires a 36-1 trigger and is not sophisticated enough to work with anything else.

[Rippthrough 98]

It ran superbly on our buggy for 6 months and I don't think you'll get much worse conditions for it. Great little system, just a pity about only 36-1 support as Tom says.

[davemar 16]

Surely the software can be modified to make it work with 60-2? What processor does it use, and is the source/assembler code available?

 

I'm having a go at designing my own ignition system at the moment, which will use the 60-2 flywheel, a MAP sensor and the Peugeot wasted spark twin coil setup. Watch this space....well not this one, another one somewhere else probably.

[Rippthrough 98]

Surely the software can be modified to make it work with 60-2? What processor does it use, and is the source/assembler code available?

 

The source and assembler is available, it's an open source project like Megasquirt, MartinM can probably provide you with more info if you PM him.

[MartinM 0]

Surely the software can be modified to make it work with 60-2? What processor does it use, and is the source/assembler code available?

Unfortunately not - the software that processes the trigger wheel inputs is in the EDIS, not in the MJLJ, so that's why you're limited to the 36-1 trigger wheel.

 

All the MJLJ does is get a signal from the EDIS that indicates the rpm (plus it gets a load measurement either from a TPS or from Manifold Air Pressure), works out the advance required from the programmed map and sends a signal back to the EDIS telling it what advance to use...

 

...plus the soft rev limiting (advance goes to zero at limit), programmable outputs (shift lights etc), ability to contain two maps and switch from one to t'other on-the-fly, data logging to attached PC etc etc

 

Yes, they're reliable as well - in fact I've never known of one that's stopped working after the initial install has been proven. Proven under lots of different rally conditions. Would be nice if it was environmentally sealed so that you could put it in the engine bay, but putting it in the cabin isn't a big deal....

 

Happy to accept PMs for more details....

 

Martin

[DrSarty 90 1 Cars]

Wow, what a great answer Martin.

 

If you want to get hands on and build what you need, you could always buy Megasquirt 2; where I know for definite it will work with 60-2 flywheels and the Bosch VR sensor. It will run wasted spark by adding a second coil driver too. All unassembled for less than £100. MS2v3.57 even has 4 coil outputs but costs a little more.

 

Have a nose at DIYAutotune.com. Excellent site, products, service and support, but in the US.

 

Maybe Martin's a MS UK dealer and could help you. Certainly knows his stuff.

Edited March 13, 2008 by DrSarty

[boldy205 74]

Cant rate it enough on my twin 45 Mi16. Added 10Ft Lbs of torque and a few BHP too. Cheap to get the parts (check out wwwtrigger-wheels.com) And a fraction of the price of a Omex Etc setup!! Plus MartinM is a very clever and helpfull chap.

So my answer: Do it!!

[davemar 16]

Unfortunately not - the software that processes the trigger wheel inputs is in the EDIS, not in the MJLJ, so that's why you're limited to the 36-1 trigger wheel.

 

All the MJLJ does is get a signal from the EDIS that indicates the rpm (plus it gets a load measurement either from a TPS or from Manifold Air Pressure), works out the advance required from the programmed map and sends a signal back to the EDIS telling it what advance to use...

 

...plus the soft rev limiting (advance goes to zero at limit), programmable outputs (shift lights etc), ability to contain two maps and switch from one to t'other on-the-fly, data logging to attached PC etc etc

 

Yes, they're reliable as well - in fact I've never known of one that's stopped working after the initial install has been proven. Proven under lots of different rally conditions. Would be nice if it was environmentally sealed so that you could put it in the engine bay, but putting it in the cabin isn't a big deal....

 

Happy to accept PMs for more details....

 

Martin

Thanks for clearing that up Martin. So could something be designed to replace the electronics within the EDIS unit (using another coil system) which could then be connected to MJLJ? Is it the EDIS unit that tells the MJLJ where TDC is?

 

I'm also curious to how 'advance' is actually determined, seeing as you can't actually turn back time! Instead of (for example) 10deg advance from TDC, does it actually do 350deg retard from TDC minus one cycle, or 170deg retard from BDC?

[Tom Fenton 1,524]

The trigger point is 90 degrees before TDC, this gives the MJ enough time to lookup in the advance table and signal to the EDIS to fire the coil in time before TDC.

[MartinM 0]

Well, almost........

 

The MJLJ has no idea where TDC is, and doesn't need to know. All it does is signal to the EDIS what advance is to be used, based on its incoming signals of rpm and TPS/MAP.

 

The EDIS knows where TDC is - well, it actually knows where 90BTDC is, as Tom says, because that is where you set up the "-1 tooth" to be

 

My understanding (not guaranteed to be absolutely correct as the exact inner workings of an EDIS are not well published) is that the EDIS starts a timer when it see 90BTDC that runs for a time derived from the advance signal that the MJLJ sends it. The timer has to run for different times for the same advance at different rpm since to get, for example, a spark at 30BTDC the timer has to "run" for 60 degrees - and 60 degrees takes a different amount of time depending on rpm.

 

When the timer runs out the EDIS cuts the coil primary, which makes the spark.

 

To create your own EDIS to deal with a 60-2 wheel and interface to a MJLJ in such a way that the MJLJ didn't know it wasn't a real EDIS would be a real challenge. If you were clever enough to do it, you would probably put the MJLJ functionality into your new EDIS so that you wouldn't need a separate unit (*).

 

I actually like the EDIS setup - it's very reliable, the signalling design between EDIS and MJLJ can take quite a lot of electrical interference and still work and if the MJLJ packs up then it will carry on in limp-home mode, which gives a fixed 10BTDC spark. You won't win a rally on 10BTDC, but you will get back to the service area to fix it.

 

Finally, you could build a Megasquirt with the ignition option and use that. That would give you 60-2 capability but MS is a whole load more complex than EDIS/MJLJ and is fundamentally a fuel injection controller that has had an ignition capability bolted on. To successfully build and install a MS, and understand exactly what you are doing and how to tune it is fairly complex and you have to be a bit of a computer whiz at the same time.

 

Putting a www.trigger-wheels.com 36-1 wheel on the crank pulley so that you can use EDIS/MJLJ is a whole load easier!

 

(*) I have actually done this and it works very well on the bench, but I've not the time or PCB design skills to build it into a unit to go into a car. It just takes a VR sensor input, from just about any wheel type (36-1, 60-2 etc) and powers a wasted spark coilpack. And that's all you need. Mine runs in a slightly different way to the way I understand the EDIS to work (maybe it's the same way and I just don't realise!), gives all the facilities of the MJLJ and a whole lot more as well. Maybe it will see the light of day some time....

[davemar 16]

OK, that all makes sense. So the gap in the 36-1 wheel is at 90BTDC. With the Mi16 60-2 flywheel, where does the gap appear with respect to TDC? I could look at mine, but its always dark and wet when I'm at home these days!

 

I'm wondering whether a 60-2 to 36-1 conversion device would be possible. You take the normal signal from the crank sensor on the 60-2 flywheel and that generates a 36-1 waveform from that. I've made a 60-2 signal generator which can be speed adjusted with a pot. I made it using a PIC processor (the code is dead simple) and a small audio transformer to simulate the crank sensor's characteristics. Maybe I could extend that to read in a 36-1 signal and convert that....

[MartinM 0]

So the gap in the 36-1 wheel is at 90BTDC.

Yep

 

With the Mi16 60-2 flywheel, where does the gap appear with respect to TDC?

Good question. No idea

 

I'm wondering whether a 60-2 to 36-1 conversion device would be possible. You take the normal signal from the crank sensor on the 60-2 flywheel and that generates a 36-1 waveform from that. I've made a 60-2 signal generator which can be speed adjusted with a pot. I made it using a PIC processor (the code is dead simple) and a small audio transformer to simulate the crank sensor's characteristics. Maybe I could extend that to read in a 36-1 signal and convert that....

Good idea if you can get you're head round the maths to do the conversion and program it. Hint: the EDIS can work out 1 degree of resolution from its 36-1 input, so to keep the integrity of the system you need that accuracy as well. Also depending how good you're electronics skills are, you could synthesise enough of the normal 36-1 VR sensor waveform electronically in your converter and feed it straight into the EDIS, avoiding your audio transformer -> VR sensor interface (which is a great idea I've never seen mentioned before )

 

A 60-2 to 36-1 converter creating a VR sensor like electrical output would be a pretty useful Thing To Have :thumbsup:

 

(...still think its a whole load easier to put a 36-1 wheel on the crank pulley)