My folly - supercharged megasquirt 2L TC or die trying

[TimpanogosSlim]

This past saturday i acquired the correct TPS from a volvo 850, the obd1 escort crank position sensor, and the saturn SL coolant sensor.

And then i spent hours trying to extract a supercharger from a previa all-trac. and failed.

[v6spider]

This past saturday i acquired the correct TPS from a volvo 850, the obd1 escort crank position sensor, and the saturn SL coolant sensor.

And then i spent hours trying to extract a supercharger from a previa all-trac. and failed.

Wow! I think your spider is going to be almost as much of a Frankenstein as mine! Lol

Cheers!
Rob

[TimpanogosSlim]

Went back and got the SC today. Looks like someone else had tried and failed. The bypass valve has to come out and the output plumbing has to come off, which is no simple task. Got it tho.

Also pulled a flex fuel sensor off a 2001 sonoma. I will have to hook it up to my oscilloscope to test it. Only 6 bucks if it works, and then i can have a tune that takes advantage of e85 - i can buy it a couple miles from home.

[TimpanogosSlim]

So, another complication is that the bypass valve for the supercharger is a stepper motor. If i want to use it, I'll have to come up with a method of control. This may get interesting.

[Zippy]

This may get interesting.

Yeah, because it's been boring up until now, LOL.
I hope you get this thing going. It should be fantastic.

[burgandy81]

MS handles stepper motors ok.

You can probably figure it out with a little input from the MS forum.

[TimpanogosSlim]

MS handles stepper motors ok.

You can probably figure it out with a little input from the MS forum.

Since I'd intended to use microsquirt for convenience and compact size, i'm reaching the limits of how many wires actually stick out of the thing before i have to come up with some other solution.

I have a little arduino experience so plan A is a nano board + l298n so it can use an analog control signal (0-5vdc) to move the bypass valve from full open to full closed.

I guess it's possible that i can use the LED pins to connect a stepper controller. I'll look into that.

[TimpanogosSlim]

It looks like there are advantages to positioning the throttle body ahead of a supercharger.

given that the throttle linkage is missing from my 2L throttle body anyway, I might go back and get the previa tb.

'course, that means it would also be much easier to find a feasible throttle cable if it's mounted low on the cold side, bumping the alternator over to the hot side where it would have been pre-77.

The SC14 is internally oiled and it's possible that i can mount it at an angle, which would help with fitment i think, if i keep the original plumbing.

I'll probably want to see about having someone with a TIG welder modify the output plumbing so that it doesn't exit through a silencer behind the SC anymore. That is, if the stock output plumbing and bypass valve even fit in the engine bay.

I may have to pull the intake manifold and alternator off of my 1800 so i can eyeball it at least.

[TimpanogosSlim]

It does look like there's a pwm to stepper addon for megasquirt that can control the bypass valve.

Went back and got the throttle body from the egg van. Not real happy with it because someone took the TPS and the IAC and broke one of the TPS screws doing it. I might have a spare toyota TPS that will work, need to check. I can gut the integral IAC and use the bosch one i already pulled from a volvo.

went to another yard and came across a bmw 525i that had lost it's head, so the fuel rail was just hanging out in the engine compartment. So i took all six 21.5lb injectors and the injector harness. These are probably overkill, which means they will have a real slim duty cycle most of the time on this engine.

Then found a 1989 ford f250 and got a bosch inline high pressure fuel pump. Since the fuel components were just sort of dangling and the pump is much cleaner than it's bracket, I assume (and hope) that this wasn't the original pump. Got the bracket too.

Progress i guess.

[MrJD]

Any particular reason you picked that supercharger?

[TimpanogosSlim]

Any particular reason you picked that supercharger?

Mostly the size and the fact that the plumbing that comes attached to it has round intake and output ports and has an integral recirculation valve. That, and it was less than $200 out the door at the junk yard. There are a lot of them out there if you check car-part.com.

It's also kinda small, has a short nose, and can be mounted in any position.

It's kinda nifty that it has a clutched pulley but i am unsure whether i will keep that feature.

It turns out that there is NO availability of service parts for these. The seals and bearings are NLA. Koyo will make a run of the bearings if someone wants to buy qty1000. The issue is that they have to be rated for 10,000rpm. You can get bearings the right size, but they aren't rated for even half the speed. I think someone may have a solution for the seals.

They won a design award for the lobes - they are cast aluminum, which is then milled for the basic shape and weight, and then a fluoropolymer (e.g. teflon-like plastic) is cast over the aluminum. This means that they are very precisely shaped but if the surface gets damaged there is nothing you can do about it but replace it.

Thankfully the one i pulled seems to work great. Just needs the oil topped up.

One problem is that the output goes to the rear of the unit and goes through a silencer. Makes sense when the engine is behind the thing, but i might end up paying someone to cut off the leg that goes around the back and tig on a forward-facing output port.

If i had more money, I might get the charger used on the mini cooper S.

I have no real updates on this project. Weather has been rainy so the engine block has been in the shed all this time. I need to strip it bare and get it to the shop, but I'd prefer to do that on a dry day.

I did take back the previa throttle body I pulled at the self-serve yard and get a better one complete with IAC and TPS from a junk yard, for $5 less.

This means that i have a Volvo 850 TPS and a Volvo 740 IAC that are up for grabs if anyone wants them. I should probably test them and list them on ebay.

Perhaps a job for the weekend is to assure that the BMW pink top injectors i acquired will really fit in the intake manifold, and figure out if i want to fab a new fuel rail or fab / buy o-ring to hose tail adapters. I know they can be bought - for $25/ea (ouch!) - but i suspect i can fab them with some copper fittings.

I've got to reconsider whether i want to just use that lancia intake manifold. Might price them out. Since i won't be using the original fiat throttle body anymore, I don't have to find that linkage, and i can remove the butterfly and probably mount the IAT sensor in there.

[TimpanogosSlim]

Non-update, blog type musings, whatever:

Still no actual progress. My DIY time has been sucked up by other things, like fixing cars that aren't even mine.

I have acquired a Lancia intake manifold. I think that's ultimately going to work a lot better. Need to send it to the shop to be baked out, find a blockoff for the old EGR port. Will need to fab an inlet adapter (did not come with TB).

I may have been talked into going with LS type coils instead of the mustang COPs. LS/LQ series GM engines are coil-near-plug. Each spark plug has it's own somewhat substantial coil, with it's own igniter, at the end of about an 8 inch spark plug wire.

Since those are V8 engines, they have a bracket on each side of the engine that holds 4 coils. Those four coils have one wiring harness that terminates in a 7 pin weatherpak connector. Very clean to implement.

I think i would basically have to find a way to mount a bracket between the cam towers. Or out over the exhaust manifold.

At any rate, it looks like I am going to try and find a chevy/gmc truck or van at the yards with an LQ4 engine and take one rack of coils + wire harness. In wasted spark configuration on a 4 cylinder, only two logic level lines are required to drive them, just like EDIS.

The guy who talked me into it has a megasquirted volvo 242 turbo. He was also showing me his mitsubishi cam angle sensor, doing crank position sensor duty in the distributor port of his block. Very interesting. He bought an adapter from Yoshifab, but some people have done DIY conversions using the bottom half of a distributor mated to the CAS. Since these are parts anyone can pull at a junk yard, i might look into it.

[TimpanogosSlim]

I'm still deeply mired in other projects while continuing to put the cart ahead of the horse. I understand that this is how Ferdinand Porsche arrived at many of his early designs.

Also I'm disgusted that i was already mired in the particular not-even-my-car project i was as last time. There is light at the end of the tunnel though.

At any rate. I had an opportunity to buy a partially assembled Megasquirt v3.00 motherboard for sort of cheap. So i bought it.

MS is in an odd phase, in my opinion, where you start by buying a board that was designed 10 years ago and then modify it. And then plug a daughtercard into it, usually.

Speeduino, where you buy a Arduino Mega2560 board off the shelf and plug a brand new daughtercard into it, makes more sense to me. But that project is not really ready for what i want to do. At the moment they are holding off selling any more v0.4 boards while v0.5 is finalized, and it lacks things like boost control.

I'm gonna continue to try to be a cheap bastard with my ECU. By that i mean, when the time comes, i will troll around for a used MS-II daughtercard that will presumably cost less than the $130 asked for a new one. And I'll troll around for MS-II case parts from someone who has upgraded to MS-III.

I found one of those DSM cam angle sensors on a '91 Eagle Talon TSi in a junkyard, and acquired a used cambox-mounted points distributor that will be the basis for mounting it. Still playing with ideas, but here's the current thought:

The distributor shaft is 12mm. The mounting plate for the advance mechanism is pressed onto splines that are a hair bigger than 12mm.

I plan to cut the distributor shaft short right above those splines.

The DSM CAS looks like this on it's business end:



That shaft i have eyeballed at maybe a bit less than 12mm, with the barrel of the spinny thing at something like 16mm.

One thought: I could take a small chunk of black iron pipe with a nominal 1/2" ID (real ID closer to 17mm) and cut a slot in it to fit the wings on the spinny thing, and tack weld that to the advance plate. The difficulty there will be centering.

Another thought: I could pull the advance plate off of the shaft and press on a bronze bushing with an OD slightly larger than the barrel of the spinny thing, then take another bronze bushing with an ID matching the OD of the first and cut a slot in it, and press it onto the smaller bushing. Potentially solder it in place as well. That should take care of centering.

Near as i can tell, that's basically how Yoshifab's CAS adapter works.

Other options include off-the-shelf shaft couplings.

There will then need to be a spacer between the CAS and the body of the distributor - bolted to the distributor the same way the distributor cap attached (just longer screws), with the CAS bolted to the spacer plate.

I could prototype that in MDF and then eventually have it milled out of aluminum. Or i could just schellack the MDF and call it good, though that would be pretty ghetto of me.

The picture above is a reference to aligning the CAS for it's intended installation. What I've gleaned from the megasquirt forums is that MS doesn't care what angle your missing tooth or notch or window is - you just figure out the angle and program it in.

Also, I found what is probably the cheapest wideband o2 controller out there:

http://www.wide-band.com/product-p/apsx_d1diy.htm

$25. Sans-enclosure. Could probably be mounted inside the Megasquirt enclosure, though they are using a hall sensor or a tiny reed relay or something for their calibration switch, which would make installation in a metal enclosure potentially problematic.

[timinator]

Just can't help myself from making a few comments. My opinion is that the Ms part of your project can wait. Build the blower to the engine with everything stock Fiat. This is no small feat in itself. Low boost is doable.

With a running car you will get a better understanding of what it takes to make the engine you are looking for. With prototype engines the problem is that "you don't know enough to know what you don't know". After the engine is running you will see where you can or cannot place the up-grade components you are planning.

Never place any electrical or electronic device anywhere near the exhaust. Heat shields are your best friend. Blower engines make unreal heat so plan for it.

I'm on your side. Hope you don't take these comments as criticism.

[TimpanogosSlim]

Just can't help myself from making a few comments. My opinion is that the Ms part of your project can wait. Build the blower to the engine with everything stock Fiat. This is no small feat in itself. Low boost is doable.

With a running car you will get a better understanding of what it takes to make the engine you are looking for. With prototype engines the problem is that "you don't know enough to know what you don't know". After the engine is running you will see where you can or cannot place the up-grade components you are planning.

Never place any electrical or electronic device anywhere near the exhaust. Heat shields are your best friend. Blower engines make unreal heat so plan for it.

I'm on your side. Hope you don't take these comments as criticism.

Of course it can wait. I had the opportunity to acquire the board. I took it.

I know the lancia intake manifold will fit because others have used it.

LS coils turn out to be mounted over the exhaust manifold in their OE configuration on GM vehicles. Yes there are heat shields.

I might have to move the alternator to the exhaust side to fit the blower. Will have to see.

As for building it stock - I don't even have the stock L-Jet stuff (aside from the stock intake manifold and plenum, sans-throttle body i might add (someone had already taken it from the donor vehicle)), and i don't want to waste money acquiring it so i can briefly use it to bootstrap the engine.

I have enough parts to build it carbureted but i don't see why i'd bother. Prove to myself that the stock distributor does it's thing and fuel atomizes in a venturi, and that when the sprockets on the front of the engine are properly aligned, it will turn over and fire? I know those things already.

I honestly don't understand what I'm expected to learn about the engine by building it stock. Internal combustion engines are not hard to understand and in the grand scheme of things, all common motors considered, I'm not doing anything particularly outlandish. Maybe not so many with a lampredi twin cam, but generally speaking i am treading where many have trod before me.

I also don't see how building an engine in the machine work sense and then immediately building a custom ECU around it is remarkably more difficult than taking a running car and building a custom ECU around it, and I've actually watched people do that. The challenges appear to be exactly the same, aside from one of these tasks being a lot greasier and more cramped than the other.

I think I've said before that there is a good chance i might bootstrap the engine w/o the ignition or SC upgrades.

This isn't my first engine modification. It's more complicated than the last, but I do have a 1984 jetta with a diesel engine comprised of parts ranging from the original '84 block, to a 1996 turbo and intake plumbing, 1989 airbox, 2006 intake manifold, 1993 breather plumbing, and various other odd combinations that just happen to bolt together just fine. The only hiccup from engine stand to car was discovering that I'd routed the turbo's oil drain exactly where the passenger CV shaft wanted to go. Took a handful more fittings to reroute it.

It ran pretty well until something in the (bizarre, huge mistake to purchase) injection pump broke. I've got a replacement pump mostly installed but have been distracted by more pressing projects.

Old diesel is a lot simpler because there is no ignition system and everything but the starter and the fuel shutoff solenoid is strictly mechanical.

Physical fitment in the vehicle is naturally a big question mark for the SC itself.

The sensor network and actuators associated with the ECU can be characterized without any kind of engine present and in fact this is standard practice for custom ECU work. All this junk will be laid out on a table hooked up to a laptop before it ends up in the car. There is, additionally, a stimulator board used to test the ECU without having to connect it to an engine.

I've debated on and off whether i want to start with an engine stand. I have an extra cross member so i can certainly construct one without too much difficulty. It's basically a box and some braces, somewhere to put some controls and gauges. Gatorade bottle for a gas tank and dumpster dive at the muffler shop for an exhaust system.

The long road is the tuning map. It remains the tuning map whichever way i approach this. And I've already found a few examples of good starter maps for the 2L lampredi twin cam in an NA configuration.

I think you are overestimating the variables that are unknowable beforehand. EFI is a relatively mature technology, and Megasquirt in particular is at least 15 years behind the state of the art. It's archaic compared to the 10-year-old factory ECU in my daily driver. But it turns out that pretty much all of the recent advances in engine control are of an extremely subtle nature and intended to maximize efficiency and minimize emissions. I can think of no other reason why my car has fully dynamic ignition timing for each cylinder individually, for example.

For the mechanical side, assembling an engine and just expecting it to work as advertised is something machinists do every day. It's expected that there will be hiccups along the way, but most of them won't damage the engine, and for those that might there are strategies that minimize risk.

If you're worried that I'm just going to spend too much time without a running vehicle, the carbureted 1.8 in my '77 is some brakes and wheel bearings and gas tank full of varnish away from hitting the road. And I'm not entirely sold on the concept of swapping the 2L into it rather than finding a 79-up rolling chassis with a lot less rust.