[Car_Upgrade] Yet Another Mi16 1.9 Build

[petert 588]

That's interesting about thge LS1/2 coils. I've always used them simply by grounding pin C at the ECU. See attached. Thus they obviously work either way. I'm wondering by using the OEM S16 ignition modules and TTL, you could achieve the same?

I think one of the reasons S16 coils have poor reliability is because of being heated to death under that cover.

LS1

[Andy 105 1 Cars]

Thanks for this diagram. Really interesting . Cannot quite figure out why grounding the gate gets the mosfet to conduct, but clearly it does. Looking at my s16coils in situ, it will be easy to make some neat brackets  to hold them in place without using the cover. I went looking at multichannel ignition amplifiers and came to the same conclusion as you. A quad channel driver, pulsed by the ecu will drive all four s16 coils. Reckon that is my way forward. Others reading this will be think ‘ for Christ’s sake  Andy, get a coilpack, some leads and stop messing about ‘ However , Having got sequential injection I really would like to finish the job and get sequential COP ignition if I can. With hindsight, getting some LS1/2 coils in the first place would have cost me less and have the job done by now. That might still be the solution, depending on the cost of the quad coil driver ( and the extra wiring I will have to do ) and how neatly Ican install it. 

 

[petert 588]

Yes, but LS1/LS2 style coils are messy. You need to mount them on something and get plug leads made, which is a pain going from Mi16 on one end, to LS1 on the other! Here is a locally produced quad channel igniter. If it were mine, I would buy a four channel CDI, eg M&W Pro-14. It will accept the TTL input and drive the S16 coils. Power to punch through any situation you can dream of.

 

https://www.haltech.com/product/ht-020006-quad-channel-oem-igniter-dumb/

 

https://www.mwignitions.com/pdf/Pro-14 S3 instructions.pdf

 

 

[welshpug 1,661]

I have seen the S16 cover chopped into 4 separate pieces on a group A 306, gti6 coils may be possible to adapt?

[Andy 105 1 Cars]

Thanks Mei. Looked at one of several I have and can see how it could be done. Two strips of angle, about 4cmx1cm and 30cm long would also do the job. I have satisfied myself that there are several ways of holding these coils in place.

Thanks for the links Peter . That MWigniter is a hefty piece of kit . Probably used one to ignite the Saturn 5 rockets back in the Apollo days . It is butch enough . 

My ecu does have high current drivers installed in the output stages for the coils . So on the face of it, dead easy solution. But, when I built the loom, I planned on TTL logic coils, so used wire for the coil outputs suitable to trigger coils with inbuilt drivers, and not so sink 10plus amps . It is 17A thinwall wire so probably o.k , but there is no way that I am undoing the loom , so some kind of quad ignition amplifier would be erring on the safe side . I do take your point about the installation of the LS1 coils . I looked up such an installation ( the one I saw had the coils slung underneath the strut brace ) Neat is not the most obvious adjective that sprung to mind .

Back on it tomorrow .

[petert 588]

I doubt the coils will pull 10A? How did you calculate that?

[Andy 105 1 Cars]

Cold resistance of the primary coilis 0.9-1.0 ohm, so with a PD of 12v, then 12A. However, with a duration of only 3or4 ms , the back emf will be appreciate , so the current draw will be significantly less than the ohms law calculation, so probably more like 5or 6 A in reality . I have spoken to the ECU supplier this morning . Going to drive the coils directly from the ECU. Wiring they reckon will be absolutely fine, and I am worrying needlessly( which is really what you are hinting at here ) . If the quiescent current drain really was 12A, I would have cause to worry, but the actual current load will be far less than this, I am assured . 

Next job is to make a coil retaining bracket and then try starting the damn thing 

[Andy 105 1 Cars]

One  COP holder in primer later. Do not imagine that there will be any demand for more of these, so I will pause further production !!!

[Andy 105 1 Cars]

Waiting for the ecu to come back so I finished the front . Fog lights were a bit scruffy but cleaned up o.k ( paint helped ) and a rewire . The aftermarket number looks o.k but the bolt holes in the damn thing were not very good at lining up with either the valance or the number mounting brackets. I ended up  having to cut, extend and re weld them to get the bumper to fit . 

Not concours but it all looks o.k 

[Andy 105 1 Cars]

For ‘number’ read ‘bumper’  Apple predictive text !

 

 

[Andy 105 1 Cars]

ECU arrived. After a bit of a struggle, it starts and runs. And no obvious leaks after 15-20 mins at 1500 plus rpm . One or two little relay issues that caught me out for a while , and now lots to do with balancing the bodies . Oh, and got a sync issue with the cam and crank sensors so running wasted spark and grouped injection just for the next day or two. The traces on the scope ( in the software ) are absolutely as expected (60-2 for the  crank and 1 pulse (5v -0) every two crank revolutions for the cam , but they are not syncing correctly at the moment . Firm talking to tomorrow . 

Then a quick drive up the road , probably only to discover that my brakes don’t , he said pessimistically, even though the maths says they should be fine 

[petert 588]

Perhaps the cam sensor tang is not positioned correctly? It does vary depending on the make of ecu. For OEM and Haltech it has to happen before #1 timing window, yet for Autronic it has to happen before #4. 

[Andy 105 1 Cars]

Hi Peter,

 you are on to something. Each 720deg of crank rotation, I get a brief sync as the cam hall sensor output falls from 5v to zero . I have the tang lined up at tdc on number one cylinder , but that is on the induction stroke, so effectively the cam pulse is timed on number 4 firing stroke. I think I am 180degrees out on the cam tang . I will move it and try again. 

Cheers Peter. I think this is probably to root do the problem 

Andy

[petert 588]

The window for #1 would start at say 60 deg BTDC. That gives the ecu plenty of time to work back from your static setting through to maximum of say 36 deg at cruise. Thus the cam sensor needs to fire sometime before 60, and most importantly, be a falling edge, otherwise it can still overlap if the events are close. The sensor fires when the tang is half way through the sensor. There’s ample space on a 4 cyl. If you position it at say 120 deg BTDC you should be fine. 

[petert 588]

A smaller three legged puller, with the feet ground finely, can get in behind the tang, enabling you to remove it

 

I just remembered you have a 60-2, which starts at approx 112 deg BTDC. Thus the cam reference has to finish before then.

Edited August 16, 2019 by petert

[Andy 105 1 Cars]

Thank you Peter, most helpful..I think if I put the tang around 180deg btdc that will still leave 30deg headroom on 114 deg on the crank sensor plus 36deg advance . I will move it and try it .Thanks for the puller tip. 

[petert 588]

Your 114 is included. Thus you have 114-36 for the computer to calculate. Heaps of time................

[bluesleeper 3]

On 4/15/2017 at 4:26 PM, Andy said:

Thank you Peter. I have two types of heat exchanger, the only real difference is the direction of the water outlets and a top hose like the one shown in the first picture you posted, which pretty much determines which cooler I should use. Only dilemma as that the new cooler has the wrong outlets for the hose I have and I have no easy way on checking that my older cooler is undamaged. It certainly looks ok, cleaned out well and is now on the car. Plumbing for the oil filler, breathing and catch tank will be a bit more of a challenge!

Andy, 

What is this oil setup? I'd love to do a similar setup on my mi16 engine. 

Cheers, 

Gareth 

[Andy 105 1 Cars]

Hi. The heat exchanger and remote oil filter came from a 405 Mi16 ( I think ) I made up the hoses using -10 braided hose and speedflow fittings Some 205gtis had the heat exchanger directly underneath the oil filter and longerfeed and return water hoses that were moulded in to the top rad hose a

[Tom Fenton 1,542]

Some late 205 GTI’s had that exact arrangement, well without the fancy hoses, but otherwise as shown. Maybe post 92 cars.

[welshpug 1,661]

have seen that on a 1990 car, was quite surprised myself, motronic and power steering on it, vin number matched up!

[petert 588]

It’s not Mi16 but rather standard fitment to all GTi’s with A/C. Thus every GTI delivered in Australia.

[petert 588]

Here's one I've been working on with more of the standard piping. I cut off the standard hoses, silver solder on #10 fittings either end and iuse push-on hose.

Edited August 20, 2019 by petert

[Putte205 4]

Standard equipment on all swedish 205 gti’s afaik, never seen one without. Heat exchangers works very well both summer and winter.

Bonus is changing oil filter is easy, no need to even raise the car, just reach under the front and unscrew 

[mmt 15]

All standard components? Never seen the oilfilter relocation thing before. The heat exchanger looks to be langer than the standard mi16 ines? 

 

I too World love such a setup. Anyone WHO could sat more about the exact components? 

 

Thanks :-)