Rod's build thread - new title - TSCi
Turbo Mini Forums

Haynes - Engine Management - By Dave Walker - Note there are two but the one to get is the dave walker one. And Engine Management Advanced Tuning by Greg Banish. Tells it all in simple english and even has a section on Megasquirt. I havent read it all yet but good so far. Both are quite often available on Ebay. I hope that helps

Lockfast, that's good - I hope it encourages more people to give this a go...

****************

Tonights FI porn is just the final construction of the phase sensor - the mechanical bits I did ages ago, just wired it up today....


The optoswitch is a bit compact (a bit like th Dsub37) and I've put the LED side current limit resistor in the wiring, buried in the black heat shrink - I need to change the input resistor in the MS too as we're now running these bits at 5V not 12.

Assembled,

Ignore the "spare" hole, a long story...

And with the cover on


The cute cat that eats the fish hasn't actually jumped on my keyboard yet tonight....

The tin is a perfect fit though, just need to spray it red....

Schrödinger's cat - so which one am I ???

NO!!! leave the tin!!!

Yeah leave the tin. Is it dolphin friendly tuna?

Schrödinger's cat - so which one am I ???

Progress on the EMS today...

All electrics hooked up on the test engine on the garage floor


I've made up an adapter to tee off some of the key signals to the scope when it's all installed in the car

Jean has something similar on his website but I wanted this to be small enough to use in the car.... and I've now decided I enjoy fiddly soldering...

I checked all the basic inputs on Megatune (no photos) but the one I was worried about - TPS - was the right way around.

Unfortunately, when it came to the cam (phase) pulse, despite all my carefull measurements assembling the opto switch, it was in the worst possible position


Right on top of the missing tooth of the 36:1 wheel

Luckily I made the slotted disk easily moveable without having to turn the old dizzy housing


The real hassle took several hours to sort out - I couldn't get the MS to synch up at cranking speed on the engine although it obviously synchs perfectly on the JimStim (hence all the scope plots on Paul's threads).

At first I thought it was down to too big a gap between the VR sensor and the toothed wheel - for some reason it has ended up bigger on the "test" engine compared to the real one - and the scope shows only 1 - 1.5Volts (and a non-centred wheel) but this should still have been enough.

Eventually, I found that the potentiometer settings on the MS mainboard have to be substantially different between the JimStim input and the real input. I need to look into this a bit more as the settings on Joe's JimStim that I originally borrowed are different to mine (V 1.3 to V 1.4) ???

Anyway, once I'd figured out how much the pots have to be set different on the engine compared to the JimStim, I finally got it to synch up and at cranking speed on the actual engine,
injector pulses


Note the MAP sensor wasn't physically connected for this, I did all the electrics but not the vacuum tube...


280 RPM on the starter (plugs out)....

I just need to finish the fuel rail and high pressure fuel system and then run it for real.

All this testing on the scope first is because I intend it to start on the first turn of the key when all the other bits are complete - I don't want to be fault finding with 3bar of fuel in front of me....

Schrödinger's cat - so which one am I ???

Nice job! And it is a very good idea to make sure everything is doing what it should before having the fuel and ignition going.

And it's almost always the case that the trim pots on the VR circuit have to be move when going from the JimStim to the car.

One other thing to check is the polarity of the ignition. Make sure that the coils are powered during the dwell time and not the reverse. That will avoid burning the drivers and/or the coils.

Jean

http://www.jbperf.com/

Thanks Jean - testing the ignition is tomorrow's job.

I am very concious of frying the chips or coilpack as I'm using the newer ***** chips (can't remember the number) not the VB921s and they aren't current limited so I've already built an extra fuse into the system....

Once I can get an ignition pulse on the scope shortly after the relevant injection pulse, I'll finish the plumbing and try it for real....

Schrödinger's cat - so which one am I ???

Looking good Rod.

I like your approach. You are certainly being methodical and leaving nothing to chance.

Without a scope, I took a slightly different approach. Got the MS running the ignition on EDIS first. Then went over to Jean's code and ran the ignition direct from the MS. Finally fitted the injection, turned the key and off she went. I had a few minor teething problems, but relatively straightforward.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Well..... I can only just say.... Do you want a "free" holiday week in Portugal???

Food, beer and bed for free!!! ahahahah

Think about it!!!

Paul, yes methodical as I'm trying to do it all in one go, rather than your staged approach. Normally I would have done it a bit at a time yoir way, but I need to catch you up.......

Jurguen, when it runs, you never know, I might need a holiday ... but try it yourself first - it's not that hard, just a lot of time....

So,

Final wiring inside the MS

Two coilpack driver packages, Boost control transistor, launch control circuirty, and..... well the other transistor for something I thought would be a good idea ages ago but I've forgotten what...

But,

Sparks....



Photo is a bit crappy as I had to set the shutter speed low to get them in, not easy with cheap digital cameras, but they're there...

Fuel rail next.

Schrödinger's cat - so which one am I ???

very exciting ... thanks rod.

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

Who needs a holiday?

I'm sitting in the Motorhome on Dartmoor in the pouring rain - thank god for TM to keep me amused.

Looking good Rod.

Have you worked out what the extra transistor is for yet?

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

You won't like me telling you it's quite bright, and certainly dry, in my part of East Anglia today then......

From memory (remember I started building the electronics a year or more ago), when I decided to put an extra heatsink in for the coil driver chips, I hadn't decided whether I was going to use a stepper motor IACV or a PWM drive one. As you have to replace the little transistor on the mainboard with a big one for a PWM idle valve, I would then need a second one for boost control.

But with the Extra/Siamese code you can use the use the PWM "idle" drive (big transistor) for boost control providing you're using stepper motor for the IACV.

At least that's how I think it will work.....

So the second one is just a spare now.

I'm just doing a load of tedious scope plots at the moment so I can move my cam/phase wheel to exactly replicate the JimStim to make comparisons on the scope easier.

And waiting for a couple of last bits to arrive in the post to finish the fuel rail.

Schrödinger's cat - so which one am I ???

Fuel Rail.....

Basic bits after a few hours machining


and how it crams into the engine/plenum


I need a 180 degree bend to make the inlet routing safe in the car but this will do for a test.

The inlet housings need cutting down and two small clips making to the plenum, that way the injectors are sandwiched so I don't need the fancy clips that are usually used. Also the regulator will have a bracket to the end of the plenum,

Obviously the injector/runner bosses aren't there yet, I left them until last to make sure the alignment of this all works - it's a really tight fit the way I've done it compared to Paul's swan necks, and it will be even worse when I do the four (ie, staged) injectors.

The idea came from someone else on this forum (can't remember who, some college project) but his was steel, silver soldered, mines alloy, threaded and will be held together and sealed with my favourite Hi-Tech engineering product, Loctite....

A fuel proof grade obviously - but sometimes the simplest ideas are the best.....:)

Schrödinger's cat - so which one am I ???

somthing i came across the other day thati'd not seen before is an VW/Audi 5cyl inlet plenum, it has plastic screw in bungs for the injectors.

Sounds usefull - you would still need a suitable lump of metal to drill and tap the hole in though ???

Biggest problem I had with my first attempt at the injection bosses was I drilled them 14mm and then welded them on, but when I went to drill through the runner to complete the job, the drill wandered and the hole was too big. That's why I've left the injector boss welding to last now, the hole is currently only 13mm so I can't "test" assemble it properly.

Fun....

Schrödinger's cat - so which one am I ???

Why don't you drill the runners before welding the boss? Even if the hole is larger and/or not exactly centered you have enough metal on the boss that you have some margin.

Jean

http://www.jbperf.com/

Valid point Jean, but not in my case....

To get the injectors as close as possible to the ports, the final position of the injector O-ring is actually on the interface between boss and runner - so they have to be finally drilled 14mm as one.

The fact the O-ring is on the join is bound to lead to other problems......

Schrödinger's cat - so which one am I ???

I see.

But does that mean that the join is on both sides of the O-ring (on different parts of the O-ring) or is it just on the O-ring? In the first case I would think it could cause an air leak unless the join is sealed up somehow.

Jean

http://www.jbperf.com/

This is from an earlier attempt (where the drill wandered) but the O-ring will still end up on, or across, the interface.

If there is an air leak, epoxy resin and reaming the hole afterwards will be my solution...

Schrödinger's cat - so which one am I ???

Runners tack welded (MIG) so will finish properly with TIG tomorrow



This appears to be the most I can cram in over a standard T3 Metro exhaust manifold without hitting the clutch master cylinder.

Only time will tell if it all fits in the engine bay.

Test run of engine on MS now estimated for Monday...

Edited by Rod S on 31st May, 2009.

Schrödinger's cat - so which one am I ???

Looking good, roll on monday

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Right, whilst others are breaking things, I'm still making things.....

I've essentailly missed my Monday deadline for running the engine on MS because getting the HP fuel rail set up suitably took longer than I thought.

Anyway, final test - outside, on grass, fire extinguisher and hosepipe ready......

In my industry the keywords are diversity and redundancy..... you have to think about that one



The fuel rail took a lot more modifications to get it and the injectors sitting square than I first thought



My main problem is this


I can't get the fuel pressure below 3.2bar....

The reg works fine (sucking and blowing on the differential connection moves pressure up and down as expected) but the minimum is lower than I would have liked to start testing with.

The return line is big enough



welding ferritic again was so good, even though the access was appaling...

and the flow into the tank is fine



It just looks like the pump is too high a capacity or the regulator to low..... that's why I haven't added a proper support bracket yet.

The injectors are now as close as I can get them to the runners



So we shall see...



Tonight's dilema is which to put this all on....

The "turbo" engine has all the NA dual widebands fitted and working and a cooling system but no phase sensor and an un-finished 36:1 wheel. But it ran fine on a carb.

The "test" engine has all the electronics but a turbo hanging of the back (for getting the dimensions right for the final build) and no cooling system....

I need to go and count how many spare parts I have....

Schrödinger's cat - so which one am I ???