Home made Megajolt type ignition.

[markypug106]

Hi guys, new to the forum and thought I'd say hi and share my lil project I'm going to start.

I've had a Discovery 1 for about 4 years now I believe, and 2 years of that was spent rebuilding it, which included the Rover 3.5 v8 on hotwire.
Soon after owning it, she started to suffer a misfire around 2-3k rpm, which has remained after the rebuild and I cannot for the life of me resolve it, even with posting for help on a couple forum's (not this one as I had big issues getting signed up for some reason!)

So recently I decided it must be down to the archaic HT design due to the reason above and down to this being a 4x4 as it's well known distributors are fickle when damp is involved.

I'm going to build a controller utilizing the 8 tooth trigger wheel on the original Lucus distributor to trigger the spark as it does the mechanical and vac advance for me (I may modify this to be electronically controlled at a later date depending on how it goes, but for now it will do what I need!).
I'll be using Ford (wasted spark) coil packs as I know they work well and are readily available and also some electronic wizardry!

Out of curiosity, has anyone done one like this and if so how did it work?

Cheers Mark

[DEVONMAN]

Initial quick comment,

You will need 4 separate timed pulses to drive the coils.
Are you going to have 4 pick ups in the dizzy??

If so the 8 tooth wheel in the dizzy will not work.

[DaveEFI]

My initial reaction is - why bother? There are so many aftermarket systems that do what you want.

Using the dizzy as a trigger leaves one of the major problems - the slop in the drive to it. So most would convert to a crank trigger. Wasted spark also needs some way of knowing TDC - the Lucas dizzy trigger doesn't give this. So you'd need to remove two opposite teeth from the reluctor making in effect a 4-1 trigger. The most usual type on a crank is a 36-1, giving a far better/quicker resolution of position.

Cheapest option would be EDIS8. With a bit of luck you could get all the bits for that for about £150. That is a standalone system which does everything apart from advance curve. Connect it to any Megasquirt and you have a fully programmable system, with plenty of support for free. A used MS1 should cost about £150 too. Or you could get one (or modify one) to four coil drivers and not need the EDIS8 module. There are other makes which work with EDIS like MegaJolt - but the numbers of MS around make it cheaper secondhand, and easy to get.

However, you obviously have a fault on your current setup. Dizzys were used for many many years without misfires. There have been lots of reports of poor quality aftermarket parts for this dizzy.

[markypug106]

Hi guys, thanks for the comments.

Initial quick comment,
You will need 4 separate timed pulses to drive the coils.
Are you going to have 4 pick ups in the dizzy??
If so the 8 tooth wheel in the dizzy will not work.

Yes and no, It doesn't make to much difference, if I have 4 pulse's, each one will will drive 1 coil, but if I have 8 pules's the output from the cmos will will be bridged (1/6 8/5 4/7 3/2)

My initial reaction is - why bother? There are so many aftermarket systems that do what you want.

Using the dizzy as a trigger leaves one of the major problems - the slop in the drive to it. So most would convert to a crank trigger. Wasted spark also needs some way of knowing TDC - the Lucas dizzy trigger doesn't give this. So you'd need to remove two opposite teeth from the reluctor making in effect a 4-1 trigger. The most usual type on a crank is a 36-1, giving a far better/quicker resolution of position.

Cheapest option would be EDIS8. With a bit of luck you could get all the bits for that for about £150. That is a standalone system which does everything apart from advance curve. Connect it to any Megasquirt and you have a fully programmable system, with plenty of support for free. A used MS1 should cost about £150 too. Or you could get one (or modify one) to four coil drivers and not need the EDIS8 module. There are other makes which work with EDIS like MegaJolt - but the numbers of MS around make it cheaper secondhand, and easy to get.

However, you obviously have a fault on your current setup. Dizzys were used for many many years without misfires. There have been lots of reports of poor quality aftermarket parts for this dizzy.

There's a few contributing factors that come in to why I wish to make it, rather than buy an off the self product.
I enjoy creating bits for myself, as its more rewarding when it works and you tend to learn more on the journey.
Budget. I'm not wealthy, if I can save on one part, it'l means I get to spend on other bits.

I have to agree that the drive slop isn't the greatest characteristics with the distributor. but I'm not that clue'd up on the mechanics of advance, I understand how it works and why, but not to an extent where I can comfortably replicate it digitally.
But as this starts coming together, I may change to a crank trigger, rather than the dizy.

So the EDIS8, can you use this without an after market/standalone engine management system (like MS) and if so how would you control the advance? Is that what Megajolt is, and advance controller?

Since posting this on my regular forum a possible solution may have popped up for my misfire, but even if I do sort it, I'll still go ahead with this project!

Thanks again, Mark

[DaveEFI]

Thing is, all you'd really be doing would be to get rid of the rotor arm and cap.
The ability to create your own map (easily) to a far greater accuracy than springs and weights can achieve is very worthwhile. Especially with an accurate trigger from the crank.

EDIS is a stand alone ignition system which will produce sparks at a fixed 10 degrees advance with no external control. For use with an external controller it sends a tach signal to it (PIP) and receives back spark advance word (SAW) The external controller would normally also have a vacuum sensor to allow MAP to be brought into the equation.

I too love building things - but in this case I doubt you'll save much money, if any. Decent coil drivers cost about the same as a used EDIS module, which not only has them built in, but the VR sensor electronics too, as well as much else, like always sending the correct dwell to the coils.

Jaycar do a digital programmable ignition kit (or did - I've not looked recently) That can be used with a locked dizzy and single coil. It *might* be possible to adapt it to drive EDIS. The main snag with it is it uses a dedicated programmer, rather than software on a computer. But the cost is low.

[ChrisJC]

I agree that making use of the existing kit out there is much easier. If you want to 'brew something yourself', I would still use EDIS8, and buy a trigger wheel. Then you can muck about creating something to handle the advance / retard (electronically), but it will still work if your kit falls over for some reason.

MegaJolt Lite is just that, a mapping unit for EDIS8. I use it on my Landie, and it's great!

Chris.

[DaveEFI]

I agree that making use of the existing kit out there is much easier. If you want to 'brew something yourself', I would still use EDIS8, and buy a trigger wheel. Then you can muck about creating something to handle the advance / retard (electronically), but it will still work if your kit falls over for some reason.

MegaJolt Lite is just that, a mapping unit for EDIS8. I use it on my Landie, and it's great!

Chris.

It would probably be possible to adapt a Raspberry Pi to do this - assuming you could write the software. Which I couldn't.

[sidecar]

Thing is, all you'd really be doing would be to get rid of the rotor arm and cap.

I agree with what has been said with regards to the pit falls of a dizzy system, having said that I still run one with my programmable MSD 6530 unit, my dizzy is locked out and the timing curve is programed in to the MSD. (I've done loads of other stuff to my dizzy which should help my setup).

Really the main pit fall of a dizzy is when it is used to trigger the ignition system, that is where the errors in timing can occur. If you can trigger the system some other way then it is perfectly OK to use a dizzy to distribute the HT voltage to the spark plugs, a bit of slop in the drive gears won't make any odds at all. So like DaveEFI said above, basically you are keeping the bad bit of a dizzy and loosing the good bit which does not make much sense.

[ChrisJC]

I agree that making use of the existing kit out there is much easier. If you want to 'brew something yourself', I would still use EDIS8, and buy a trigger wheel. Then you can muck about creating something to handle the advance / retard (electronically), but it will still work if your kit falls over for some reason.

MegaJolt Lite is just that, a mapping unit for EDIS8. I use it on my Landie, and it's great!

Chris.

It would probably be possible to adapt a Raspberry Pi to do this - assuming you could write the software. Which I couldn't.

I wouldn't use a Raspberry Pi for that, I would use an Arduino instead. Actually, I would roll my own PCB, but that's just because I can!

Chris.

[markypug106]

Thing is, all you'd really be doing would be to get rid of the rotor arm and cap.
The ability to create your own map (easily) to a far greater accuracy than springs and weights can achieve is very worthwhile. Especially with an accurate trigger from the crank.

EDIS is a stand alone ignition system which will produce sparks at a fixed 10 degrees advance with no external control. For use with an external controller it sends a tach signal to it (PIP) and receives back spark advance word (SAW) The external controller would normally also have a vacuum sensor to allow MAP to be brought into the equation.

I too love building things - but in this case I doubt you'll save much money, if any. Decent coil drivers cost about the same as a used EDIS module, which not only has them built in, but the VR sensor electronics too, as well as much else, like always sending the correct dwell to the coils.

Jaycar do a digital programmable ignition kit (or did - I've not looked recently) That can be used with a locked dizzy and single coil. It *might* be possible to adapt it to drive EDIS. The main snag with it is it uses a dedicated programmer, rather than software on a computer. But the cost is low.

Well, if I can sort this issue with the misfire I'll have a bash at replicating the advance also, as I'll be in less of a hurry to get this together.

I think I've got most of the parts for the coil driver side of things, its only cost about £15 as I had a few Cmos chips floating about (but they dont cost much anywho), I'm pretty sure I've got the coil driver worked out well, but I must admit running it from a crank trigger over distributor would be a better solution if I can sort the advance control.

[markypug106]

Thing is, all you'd really be doing would be to get rid of the rotor arm and cap.

I agree with what has been said with regards to the pit falls of a dizzy system, having said that I still run one with my programmable MSD 6530 unit, my dizzy is locked out and the timing curve is programed in to the MSD. (I've done loads of other stuff to my dizzy which should help my setup).

Really the main pit fall of a dizzy is when it is used to trigger the ignition system, that is where the errors in timing can occur. If you can trigger the system some other way then it is perfectly OK to use a dizzy to distribute the HT voltage to the spark plugs, a bit of slop in the drive gears won't make any odds at all. So like DaveEFI said above, basically you are keeping the bad bit of a dizzy and loosing the good bit which does not make much sense.

I agree with what your saying but my initial idea was to overcome a couple of issue, which would have been covered by running it from the dizy drive.
I have also to disagree with what you say about the HT side, for me isn't good.
I want to be able to run without cutting out when water comes into the bay.

[DaveEFI]

The expensive part of the coil driver is the output. MS uses the Bosch BIP373 which is a pretty bomb proof driver with internal protection. But costs something like £12 each. One per coil.

[markypug106]

The expensive part of the coil driver is the output. MS uses the Bosch BIP373 which is a pretty bomb proof driver with internal protection. But costs something like £12 each. One per coil.

I'm also making that from scratch, much cheaper

[DaveEFI]

The expensive part of the coil driver is the output. MS uses the Bosch BIP373 which is a pretty bomb proof driver with internal protection. But costs something like £12 each. One per coil.

I'm also making that from scratch, much cheaper

Cheaper doesn't always mean better.

[ChrisJC]

The expensive part of the coil driver is the output. MS uses the Bosch BIP373 which is a pretty bomb proof driver with internal protection. But costs something like £12 each. One per coil.

I'm also making that from scratch, much cheaper

You have to remember that when you turn on the coil (with your transistor / FET / whatever), the current ramps up, and when the coil saturates, it's effectively a short circuit (but you should turn off before then). However, when you do turn off, the back EMF from the coil will give you at least 400 volts, which is likely to blow up the transistor.
A transistor that will handle all of this stuff is quite expensive......

Chris.