Long, slow Megasquirt conversion, Sad D-Jet times Options

[r3dplanet]

Kameraden -

I own a lovely, though far from perfect, 1971 914 with a stock D-jet fuel injection system. I bought the car almost a year ago, and during that year I've tried hard to make the thing run correctly. It's never happened. After spending apparently a thousand hours reading and re-reading the available D-Jet literature and forum posts and testing and replacing every D-Jet part I can think of, I've decided that the best way to proceed is to purge by fire.

My goal is simple: all I want to do is to make my car run as smooth as it is capable of running. I'm not after performance gains, but I also don't want to suffer any losses. I don't drive my 1.7 liter car to go fast, I drive the car because I love what it is: small, efficient, fun. If I need a speed fix I can just as easily hop on my motorcycle or jump into my hopped-up Barracuda.

If this process helps me think out loud and use the available wisdom here, then surely it will help others on what I'm expecting will be an increasingly common path. I dig the D-jet injection system, but the fact is that parts are becoming more rare, more expensive, and more uncertain. Another major factor for my action is that there are too many unknowns when trying to diagnose the D-jet. Just when I think I have it nailed down, the air temperature drops ten degrees and the car barely runs, which makes for a crestfallen, tongue-gnashing experience. With the Megasquirt setup, I can hook up my Thinkpad and just find out what's happening with the injection system. To me, this alone is worth the price of admission.

The unit I've chosen will either be a Megasquirt II v.3 unit or a system based on it. After reading the comparisons between the systems it seems that the MS II v.3 is newer enough for all I want to do, and also because the MS III system simply is too complicated for my needs. What are my needs? Right now I want to do a fuel-only setup, but perhaps later I'll install an optical crank sensor and go for spark/timing control.

Let's begin.

From eBay and diyautotune.com I have ordered and received a Bosch oxygen sensor, Innovate controller, dash gauge, and MS stim kit. Since my Bursch exhaust already burned off all the paint from its year-old rattle-can paint job from the previous owner, I decided to begin with the oxygen sensor install. This would be a handy thing to have even if I kept the D-jet system just for the sake of curiosity. Am I running lean or rich? Who knows? Well, the O2 sensor knows. So this morning I detached the exhaust and had a local exhaust shop weld in a sensor bung. I then began to lace the wires along with some existing turn signal wiring. I'll have pictures of this shortly.

After the bung was welded in, I made a $10 soda blaster like the one you can see here:

http://www.aircooledtech.com/tools-on-the-...p/soda_blaster/

I blasted the exhaust using four small boxes of baking soda obtained from the drug store down the street. I then shot it with a little hydrochloric acid to eat any remaining schmutz and rust, and then I hosed it clean with water to neutralize. Normally I like to send off my exhaust projects to be Jet-Hot coated, and I'm a believer in the process. The exhaust manifolds on my Barracuda and the headers on my Falcon were both Jet-Hot coated and they look brand new many years later. Its expensive but worth it. I didn't Jet-Hot the Bursch only because my neighbor gave me a can of Por15 Black Velvet exhaust paint. It's a brush-on affair where you paint on a coating the consistency of dairy "pudding," let it dry, brush on another coat, and then let the heat of the running motor bake it into place. I'm dubious, but it was free and I have nothing to lose by trying. It can't be any worse than the thoughtless Krylon job it was sporting until this morning. We took some photos of the process and I'll post them later.

Next up is wiring up the dash gauge and soldering the stim kit together.

Thanks for reading.

[type47]

(IMG:style_emoticons/default/popcorn[1].gif) Keep posting your progress, I, for one, could learn from your experience as I would like to do the same thing: megasquirt (or some modern FI system) for fuel. My engine is still apart now but the car is due to arrive today!

[904svo]

I would suggest you run MS1 extra it has all the features you want, I would also suggest that you run either ignition or the injectors first not both at the same time
to start with. This will made your installation a lot easer.

[underthetire]

I agree with the choice to do the MSII-V3.0. I'm running one with fuel only for the past couple years now. The nice thing is I have the option to do spark, and, I don't have to run resistors for the low impedance injectors. (IE stock injectors)

[nathansnathan]

I've painted the heat exchangers, stock muffler, and muffler heat shield on my bus with the por15 stuff, only did the exchangers and shield with the black velvet, other was silverish . It's gone, both, from all but the sheet metal (meaning the parts that don't get that hot). - not impressed with it.

I will have to check out the hot jet coating, I've got a stock exhaust I want to protect.

That seems pretty cool the homemade soda blaster. I'm getting my blast cabinet up and running, and what's in there is probably not as efficient even, the fitting being a straight 'T', instead of angled. nice.

I'm definitely interested in the Megasquirt conversion also. More pictures please! (IMG:style_emoticons/default/smile.gif)

[charliew]

soda will not clean rust off it is a very mild cutting powder. It's good for cleaning carbs or taking paint and sometimes body filler off thats all. I used the black velvet on a 55 gallon wood/oil burning stove and it has been good for several years but it was also very clean bare metal when I put it on.

[r3dplanet]

Okay. I've done a little work, but most of it is pointless. I just wanted to keep this thread moving.

Here's a few establishing shots. This is my car and engine bay. My equally orange cat Steve loves to work on the car, so he'll star in some of the photos.

-m.

[r3dplanet]

Since I've written last I yanked my Bursch exhaust and had a oxygen sensor bung installed, and it fits perfectly despite being the ugliest weld I've ever seen. The gentleman with the torch didn't use goggles and smoked a cigarette while welding. That was a little unnerving. Some part of me wants to cry for humanity.

Looking down at the trunk I already had two holes at the very rear. I used a 1" drill bit and expanded the hole on the right so the oxygen sensor could slip through. The sensor is just hanging in place until I can calibrate it, but it will slip into the bung without issue. I used a little POR15 chassis paint to cover the exposed metal from the enlarged hole and used a 1" grommet to seal it up.

I'm routing the wires on the left side of the car, into the engine bay, and then down through the tunnel. I haven't done that part yet, but I will soon.

More to come.

[r3dplanet]

I also stopped by Camp 914 today and bought a few goodies. I picked up some lovely fog light replacements, and what I believe is a blown D-Jet brain. I just want the case because my hope is to put all of the MS guts into a stock-looking case.

Here's some photos of the disassembly.

[r3dplanet]

More photos.

Something I'm not sure how to do is to remove the nifty mixture control knob. The early D-Jets like mine don't have one, but I want to keep all the guts in one piece if possible. Does anyone know how to remove it cleanly? It doesn't seem to want to pry up at all.

The next step is to blast it and powder coat it, mostly because I have a bunch of other stuff to powder coat and I need to make a $60 minimum. I'm thinking silver for the color.

[r3dplanet]

These are the last uploads of the evening, just in case anyone wants to know what an assembled Megastim v2.2 circuit board or an orange tabby looks like.

It's already raining without end in Portland. It's a nice time to dig into this project.

*sigh*

[Mark Henry]

Well.... I don't like your O2 sensor location. It will most likely knock against the body metal and any grommet will just melt off. No need for it to be "inside", no factory car is done this way.

I know the logic for wanting it inside a D-jet box, I did it on my SDS at first, but it turned out to be a huge PITA. Plus if moisture gets to it you're in trouble.

[JFJ914]

Try to PM 904SVO he has just finished a long painful megasquirt install with CNP (Coil Near Plug). The pitfalls are many, the expertise on line is mostly misleading if not downright wrong, the documentation could be a whole lot better. You get the idea. In spite of this, I drove the 904 replica on Wed and was blown away with its drive ability. It's a project well worth the blood, sweat and tears.

[JFJ914]

Kameraden -

The unit I've chosen will either be a Megasquirt II v.3 unit or a system based on it. After reading the comparisons between the systems it seems that the MS II v.3 is newer enough for all I want to do, and also because the MS III system simply is too complicated for my needs. What are my needs? Right now I want to do a fuel-only setup, but perhaps later I'll install an optical crank sensor and go for spark/timing control..

Megasquirt 1, v 3.00 running the Extra firmware is perfect for the Type 4.

Get a 36 -1 wheel now, see McMark in the vendor section, timing it from the coil will be an unhappy experience.

Good luck, it's worth it.

[r3dplanet]

Allow me to clarify. The oxygen sensor doesn't just thread directly underneath the hole. The heat shield underneath is perfectly intact. The bung is much farther to the right sitting on the collector, resting perhaps 15 degrees forward. The drilled hole is simply so that the sensor can drop through the body. I didn't want to mount the controller under the car because its too close to the exhaust, and that's why it now lives in the trunk. I don't foresee any heat related problems here. Also, when I jack up the rear wheel the o2 sensor doesn't strike the body. I'll get down there and get some photos.

Well.... I don't like your O2 sensor location. It will most likely knock against the body metal and any grommet will just melt off. No need for it to be "inside", no factory car is done this way.

I know the logic for wanting it inside a D-jet box, I did it on my SDS at first, but it turned out to be a huge PITA. Plus if moisture gets to it you're in trouble.

[ripper911]

Looks like my cat!
(IMG:style_emoticons/default/cat.gif)

[charliew]

Looks like my cat only yours has a tail.

[stugray]

spankmeister7,

My understanding was that an O2 sensor will go bad very quickly using leaded gas.

It will help you with initial install, but dont panic when it begins to drift.


Question: Are you going to run with the stock cam?
Does anyone know what the best cam is for the megasquirts?

Stu

[JamesM]

Who is using leaded gas?

As for cams i dont think anyone has done testing specifically with megasquirt, but with all the options megasquirt has available you could probably tune it around just about anything. I would say go with with something that gives decent vacuum, unless you want to mess around with megasquirt in alpha-n mode.

spankmeister7,

My understanding was that an O2 sensor will go bad very quickly using leaded gas.

It will help you with initial install, but dont panic when it begins to drift.


Question: Are you going to run with the stock cam?
Does anyone know what the best cam is for the megasquirts?

Stu

[r3dplanet]

Apologies for my absence. I've been putting the finishing touches on my other car before winter hits too hard, so I've been neglecting the 914. I'm also in the process of dropping a 2.0 liter gas engine into a 1981 Diesel Rabbit body, which is the project of my friend but parked at my house.

I've never heard of anyone mentioning issues with oxygen sensors in unleaded fuel cars. Come to think of it, I can't even think of one car with EFI and O2 sensors that runs on regular gas.

What I know about O2 sensor installation is to mount it at 12, 3, or 9 o'clock in the exhaust collector to keep any condensation from pooling up inside the sensor. I install these sensors in all of my vehicles, including the carburetor-equipped models just so I can make sure that I can tune it to perfect efficiency. I recently did this with my '65 Plymouth Barracuda with a hopped up 360. Even though everything sounded great via tuning by ear, I still needed to change out one set of metering rods and the secondary jets. I also found that a dirty air cleaner and a ONE degree timing shift can make a very large impact on the mixture. It's nice to see what your mixture is up to in real time while driving.

Here's what I've done on the 914 in the past couple of weeks:

(a) I bought/ordered a complete set of heater boxes and tubes from Camp 914.
((IMG:style_emoticons/default/sunglasses.gif) Made plans to drop the gearbox and send the innards to Dr. Evil for rebuilding.
© Set a date with a local bodywork guru to fix the dented fender and re-paint the car over the winter (the PO hand-tightened the lugnuts looking to upsell me on some other wheels, so 1/4 mile after buying the car the rear wheel fell off and smashed the fender).
(d) Reading as much MS info as I can
(e) hungrily reading JamesM's thread about his own fabricated MS boards because I want to go that route
(f) Plans to get the motor onto an engine stand so I can tap the threads for the CH temp sensor, add a crank trigger mechanism in case I want to go with spark control one day, add a taco plate so I can install an oil pressure and temperature sensors. Pelican is clean out of taco plates and sensor mounts. Crap.

So things are still in pre-production, but I'm eyeballing getting this done before next summer, which I think is a reasonable time frame.

My Barracuda averages 12mpg. Ugh. My 914's 1.7L gets triple that. Next summer can't come too soon.

-marcus