Finished install in a Mini

Show off your Megajolt installations, vehicles and action shots here!

Moderators: JeffC, rdoherty, stieg, brentp

charlie
Posts: 16
Joined: Fri Jan 19, 2007 8:18 pm

Finished install in a Mini

Post by charlie »

Right, I finished it a week ago, so here are the build pics.

This is a bit of a how-to, and a bit of a how-not-to. I don't have my own camera, and haven't been able to borrow one since I finished, so all the pics are of the build. Proof it works will follow. Promise.

It all started when brother took my Mini for a drive. Result - a bad knocking whilst cruising. Took it to my local mini people, one of whom said "you've got a loose tappet". "No I haven't" I said. At which point his son came out of the rollers bay munching on a salad, mumbled "big end" and went back in again. At least I know what knackered big ends sound like for next time.

Anyway, this is what was in the bottom of the drain pan. Not a great pic, but the silver bits are what turned out to be no3 big end (mostly).

Image

Long story short, 10 thou undergrind and a few new bits and bobs and it went back together.

Enough about the engine. More about the MJ. Well, trigger-wheels.com did the honours for the EDIS bits and bobs. MK1 coil, 5.125" wheel with 1/2" hole, used plugs and EDIS module. Arrived safe and sound.

First off was making the crank pulley and trigger wheel. The pulley I have is off an MG Metro, so it's of the cast variety. I knocked the old trigger wheel thingy off the inside of it, and put it on the engine (which was conveniently located on the bench - man would I hate to have to do this with the engine in the car). Did some quick measurements, and went off to university to use the CAD software. Made a model of the pulley in situ on the cover and had a look at clearances.

Image

I came up with this design of trigger wheel mount...

Image

...which fitted onto the pulley like this:

Image

Made out of two parts, it would have a boss bolted to the crank pulley, to which would be welded the trigger wheel. Sounds complicated, but it gave me the clearance (I thought - more later), and meant I didn't have to worry about where I put the holes. I could just make the boss and drill & tap the pulley, then weld the t-wheel to the boss in the right place once I had the sensor mounted.

It meant I had to take the hole in the trigger wheel out to 70mm, which I managed to do on the lathe at University (Ta Keith). Also did the boss, but didn't take any pics.
Image
I got it home and offered it up to the engine with the timing cover in place. Surprice surprise it didn't fit. First I needed to grind a bit off the little sensor mount at the rear of the cover, and bent the timing marks flat against the cover - not too hard. Then I found the boss was fouling on the timing cover. Not part of the plan. So after um-ing and ah-ing for half a day, and drinking much tea, I dug this out.
Image
An old Myford we inherited off grandfather which hasn't seen any real use since it arrived. Managed to find an old bit of bar, and turn it down (after studying how to make lathe tools for a few hours, and being busy on the bench grider for a few more). Much tool chattering later, and a bit of filing for the keyway, I had a little 2.5mm thick spacer. This pushed the pulley out more than enough to stop it rubbing the timing cover. Smiles from ear to ear that day, so chuffed was I.

Next step on the Megajolt bit of the build was obviously the sensor mount. A bit of 3mm steel bent into a right angle bracket with a bit of support sorted it. I MIG welded it onto the timing cover so it sat inside the fanbelt.
Image
The sensor bolted on with a single M6 nut and bolt. It was a pretty loose fit in the sensor, which had an 8mm hole, so I had some side to side adjustment. I was a little worried about how floppy the sensor seemed to be. Presumably the sensor itself has a little tube to sit in on the Ford it came off, but it seems to be OK so far.

The pictures are a little misleading as they weren't taken at the right time. At this stage of the build I had the engine on the bench with no head on and no gearbox. I put the timing cover on and fitted the sensor. Then bolted the boss I had made earlier onto the pulley, put the trigger wheel in place and slipped it onto the crankshaft. I then set No1 cyl to TDC with a dial gauge, and rotated the trigger wheel until it was lined up right.

I had previously stamped a helpful "9" into the 9th tooth after the missing tooth. I lined this up with the centre line of the sensor and marked the trigger wheel against the crank pulley with a Tipp-ex pen. Took it all off, and used a centre punch to put a permanent mark, right where the boss joined the trigger wheeel (marked with the arrows labeled "WELD IN HERE" on the pic above). This made the previous sliding fit between the two into a bit of a bodged interference fit, so they stuck together. I took this opportunity to double check it all lined up properly, then have a cup o'tea, then triple check.

To minimise any risk of distorting the trigger wheel when welding it to the boss, I roasted the two at gas mark 6 while I had my lunch. Finished lunch, lined up the centre marks (good welding gauntlets required here) and MIGed the two togeher with 3 small beads in between the mounting slots. Once it had cooled, I bolted it all together and got on with the rest of the engine build.

Before I put the engine back in, I had a go at mounting the coil and EDIS, and making the loom. Put the coil on the bulkhead and the EDIS tucked in quite nicely (read tightly) next to the wiper motor. Coil was mounted on four 14mm dia, 20(ish)mm long posts (made on the Myford again), with an M6 thread down them. Bolts through the bulkhead from the cabin stuck them to the car, and 4 pan head allen bolts through the (drilled out) coil bolted it to them. Easy.... if you happen to have a lathe lying around! I had to remove the wiper motor to put the EDIS in place, but with the engine out, this wasn't awkward. Again, I am very glad the engine wasn't in for any of this.
Image
You can see the EDIS here, and the holes for the coil mount.
Image
This is the coil on its mounts on the bench. I thought I had a pic of it in the car, but it appears not.

Making the loom up was a bit of a hassle. All entirely self inflicted though. I thought it would be fun and look tidier if I took all the old wires away and made up a new loom reusing the terminals. Bit of an error. Not Knowing how these things came apart, I managed to bugger up entirely the sensor plug, and break off one of the tabs on the EDIS plugs. Silly boy. A new sensor plug was cheaply available from Vehicle Wiring Products (off that new fangled tinterweb thingy). I fixed the EDIS plug with a bit of superglue on the appropriate terminal, and it hasn't broken yet.

I should have spent more time on the internet looking for instructions on how to take the multiplugs apart, but I was getting impatient. For those of you like me, the EDIS plug is fairly easy. Take the red bit of plastic out, then slip something small and pokey down the hole it came out of, jiggle it about. Threre is a small plastic tang that needs lifting up into the space vacated by the red plastic locking thingy you just removed. Fiddly but doable. However, I ended up just soldering wires onto the little tails I had removed. Looks OK with a bit of heat shrink over it.

Another mistake I made was to try and make all the other wires end at the EDIS plug. Being the biggest, it was really hard to get them all the same length. I'm guessing Loom Construction lesson one is "Always start from the biggest multiplug".

In the end, it worked out OK. I can heartily recommend using nylon braided sleeving and heatshrink. The braiding is very flexible and easy to push wires down. Much better than convoluted tubing and easier to feed wires down than PVC. Just watch out for frayed ends and the way it changes length depending on how fat it is.

So once the engine was back in and I had oil pressure, I plugged in the coils and tried to fire it up. Nothing, just a load of blowing back up the carb. Double checked I had the trigger in the right place (found rough TDC with a bit of wire down the plug hole, and counted the teeth as I turned the engine manually). Then used this to make sure I hadn't completely buggered up timing the cam. All good. So I swapped 1 & 4 leads with 2 & 3, and off she went. This stretched the plug leads so I had to swap the wires over on the EDIS plug, which was awkward to say the least. I must have misread the wiring diagram when doing it in the first place. Managed to fluke getting the VR sensor plugged in the right way round though so not all bad.

Had a quick check with the timing light, and the spark was somewhere between the 8degree and 12degree marks. In my book, that's 10degrees. All I had to do was wire in the MJ and go mapping.

So that's how I did it. The big question is, 'would I do it like this again?' The answer to that is: 'er maybe'.

The clearance between the timing cover and the boss was always going to be tight. Making the boss took a while - turning it down, milling the slots, counterboring them etc. And in the end, it didn't fit. If I didn't have a lathe handy, and a bit of scrap bar, I would have been in for another wait while someone else made one up, and it would have been just another bill. In short, it would've been a right hassle if I didn't have the benefit of 2 generations worth of amateur engineering facilities at my disposal, as well as a University workshop. If you want an easy install, weld the sensor to the engine mount plate that bolts to the gearbox, and have someone turn a shoulder onto the outside edge of the pulley. Much easier I reckon.

However, it is quite a tidy install. Everything is quite compact, and nothing looks like it really shouldn't be there. Making the mount in two pieces was a good idea I reckon. It really took the pressure off when I was drilling and tapping the crank pulley.

The only really awkward thing is looking at the timing marks now they've been bent out the way and someone has put a sensor right in front of them. That would be a right bugger if I still had a distributor. :D

alexander
Posts: 246
Joined: Sat Jan 19, 2008 4:33 am
Location: sydney, australia.

Post by alexander »

charlie, that is a great essay and the photos are fab.
cheers, alexander.

charlie
Posts: 16
Joined: Fri Jan 19, 2007 8:18 pm

Post by charlie »

Ta. Enjoyed writing it.

It did turn out a bit long though, but I guess more is better than less.

luke6040
Posts: 113
Joined: Wed Dec 05, 2007 12:48 am
Contact:

Differance?

Post by luke6040 »

have you used the car since your install?
How is starting and any imediate improvments?
Got any vids :wink: lol i like vids

charlie
Posts: 16
Joined: Fri Jan 19, 2007 8:18 pm

Re: Difference?

Post by charlie »

luke6040 wrote:have you used the car since your install?
How is starting and any imediate improvments?
Got any vids :wink: lol i like vids
Yep, done arond 450 miles now. 120 miles in the first run. Hasn't missed a beat.

Cruising is much improved, and tickover is a bit smoother. Currently running an unmodified Aldon Yellow map (courtesy of KLAS in the Ign Map Library I think), and it seems OK. Plan is to do 1500 miles or so then take it to the rolling road. I'll let you know what the verdict is when that's done.

No vids - I don't even have my own camera. :(

This is what the car looked like before MJ, and surprisingly enough, it still looks mostly the same. So I suppose this qualifies as a "Powered by MegaJolt" pic.

Image

KLAS
Posts: 93
Joined: Thu Oct 20, 2005 4:01 pm
Location: Germany

Post by KLAS »

would be great to know how much the map changed after the RR sesion

luke6040
Posts: 113
Joined: Wed Dec 05, 2007 12:48 am
Contact:

Post by luke6040 »

looks gr8! cant wait to get mine done, wish i could just take a week off uni and do it lol.
What engine have you got? sorry if stated above?

alexander
Posts: 246
Joined: Sat Jan 19, 2008 4:33 am
Location: sydney, australia.

Post by alexander »

and doesnt that photo just remind you, once again, that the minilite wheel (real or copy, either way) is one of the best looking wheels ever designed? there is something about them that suits smaller english cars perfectly.

charlie
Posts: 16
Joined: Fri Jan 19, 2007 8:18 pm

Post by charlie »

KLAS wrote:would be great to know how much the map changed after the RR sesion
Yeh I'm looking forward to it. It's pretty close to being right as it is.

I made a map based on the numbers given in the Haynes book for my old distributor. It gave various mechanical advances at various engine speeds and the amount of vac advance and the pressures at which it occurred. Made that into a map, and it was rubbish, just like the old dizzy was when set up as standard.

alex and luke:

Yep I love the 10" Minilites.
Engine spec is:

MG Metro 1275
CR 10.5:1
Ported and polished head
11 stud head
ARP head and rod bolts.
Centre main strap
Maniflow exh. and Pipercross filter and ram pipe

KLAS
Posts: 93
Joined: Thu Oct 20, 2005 4:01 pm
Location: Germany

Post by KLAS »

charlie wrote:MG Metro 1275
CR 10.5:1
did you calculated the CR yourself?
i am asking as my next engine will be a MG Metro one i am just rebuilding. will be bone stock and i calculated the CR and it was down to 9.6 due to the piston to deck height, a bit more than 1mm. assuming 0mm piston to deck height the CR was up to factory specs, 10.5:1

mr_bridger
Posts: 52
Joined: Wed May 16, 2007 10:29 pm
Location: Coventry, UK

Post by mr_bridger »

Hey Charlie, nice mini! you definatley went to town with your pulley, i can confirm that mounting it on the outside is a LOT easier!!!

KLAS, checkout the other attachment "CR Master.xls", great for working out your CR
Attachments
IMG3 (Medium).JPG
IMG3 (Medium).JPG (55.75 KiB) Viewed 18822 times
CR Master.xls
(51 KiB) Downloaded 892 times

KLAS
Posts: 93
Joined: Thu Oct 20, 2005 4:01 pm
Location: Germany

Post by KLAS »

mr_bridger wrote:KLAS, checkout the other attachment "CR Master.xls", great for working out your CR
thanks for that offer, but i have a profesional program to do the full math for me (and yours giving the same results). i've even tripple checked my data. thats why i wondered the factory specs was way off the mark.
i know that the A series is usually a bit away from the factory CR, but never had one nearly one point off before.

mr_bridger
Posts: 52
Joined: Wed May 16, 2007 10:29 pm
Location: Coventry, UK

Post by mr_bridger »

sorry, i misread your original post, i thought you were trying to work it out :roll:

Ross.

charlie
Posts: 16
Joined: Fri Jan 19, 2007 8:18 pm

Post by charlie »

Klas

I have to admit I took some educated guesses at ring land and deck volume......so it could be way off.

I did measure the combustion chamber volumes though, and did the maths with those and book values for piston dish, ring land vol, and deck and gasket vol. It gave me 10.45-10.55 with the varying book values I had, so it's not that far off. As it's still a fairly standard engine (original pistons, same bore) I wasn't too worried. If I'd changed the cam or tried to up the CR I probably would've taken more time over it.

Charlie

KLAS
Posts: 93
Joined: Thu Oct 20, 2005 4:01 pm
Location: Germany

Post by KLAS »

Charlie, you won't be way of if you measure and calculate, even with some data not spot on.
i needed more than 1mm piston to deck clearance to loose 0.9 point of cr, thats like fitting 2 head gaskets.
but i know some people that take what the factory said for granted and than wonder the engine won't perform

Post Reply