Megasquirt EFI and Ignition Project

[Science Guy]

Reluctor

Is it possible to just use the stock pickups rather than making the new reluctor wheel and pickups? They are sending a similar signal to the ICU any ways. Can't Megasquirt just use those signals? Also, do you have a copy of the tuning files used on your 81? I'm doing this to a 79 and it should be just about the same.

[81xsproject]

Someone previously mentioned in this thread that it could be done with stock PU's, but I want the GM DIS system. So, I don't know about that. As for tuning tables, I would be happy to give them to you but you have to have you required fuel value set up the same (I think). If you use the use the required fuel calculator in MegaTune for 67 ci (1102cc) you should get a required fuel value of 8.8 IIRC. Or you could just manually enter that value. You also have to be set up for the same number of injections per cycle and using the same flow rate injectors and I am sure there are some other things. I could send you the whole tune and you could just go in and change your ignitions settings to use the stock PUs. Everything else would be a really good starting point. I HIGHLY recommend dropping the $40 for the pro version of Tuner Studio so you can use the 'Live' tuning feature.

[clcorbin]

Sorry about even posting that question! I went and reread your post and saw what pump you went with. I guess I was hoping you found something else that is actually available...

I have ordered a pump from an '08 GSXR600/750. From the looks of it, I can strip it of the unnecessary fuel sender and then turn down the mounting flange and make a small aluminum header tank for it to bolt in to. I'll use gravity feed to get fuel to it from both petcocks and let it return the excess fuel to the header tank. HOPEFULLY, I can make this header tank small enough to sit in the space currently occupied by the little tool holder.

I plan on building my own PC board based on the MicroSquirt board (electronics is another hobby of mine and I LOVE SMD!) but customize it to have everything i need and dump all the useless stuff (ADD: USB connector, dual high current drivers for ignition coils, stepper IAC, etc. REMOVE: CAN, serial connector, etc.). With the stepper IAC built in, I plan to use the secondary throttle shaft on my '08 GSXR600 throttle bodies for fast idle/idle control by removing the throttle plates. My board won't be as small as the true MicroSquirt (no 4 layer board, larger trace sizes for the most part, etc.), but it will be much smaller than MicroSquirt boards. Oh, I was also able to get Tyco to send me a sample of the complete Ampseal connector (PCB, plug, pins, AND seals!). Every little bit helps!

The final bit of electronics that I am planning is to build a microcontroller controlled rectifier/regulator for the alternator. It should be able to maximize the low rpm output of the existing alternator while maintaining a more stable voltage level (good for electronics AND battery). I just hope the stock alternator has enough ump in it's tail to run the injection system. If not, I might have to see about converting this thing over to a returnless fuel system so the pump doesn't run as hard at low speeds and pull so much current.

Lot's of work to do...

[clcorbin]

Is it possible to just use the stock pickups rather than making the new reluctor wheel and pickups? They are sending a similar signal to the ICU any ways. Can't Megasquirt just use those signals? Also, do you have a copy of the tuning files used on your 81? I'm doing this to a 79 and it should be just about the same.

Yes. The MicroSquirt/MegaSquirt has the option of using dual pickups, which our bikes already have. I was originally planning on going this route, but I decided I wanted to go ahead and build a 36 tooth reluctor wheel while I have access to a machine shop (up to and including CNC mills and lathes... ).

I too would be VERY interested in your tune files! Especially the volumetric efficiency tables!

[clcorbin]

I did a quick search, and found it for $25. Here is the exact one I used.

Any chance you could post the model/part number of that little number? The auction is no longer listed...

Thanks!

[Shappers]

Is it possible to just use the stock pickups rather than making the new reluctor wheel and pickups? They are sending a similar signal to the ICU any ways. Can't Megasquirt just use those signals? Also, do you have a copy of the tuning files used on your 81? I'm doing this to a 79 and it should be just about the same.

Just to clarify a little..

Megasquirt will handle the fuel injection system but can also handle the ignition system.

The guys on here have gone full hog and done both for a complete solution. If you like you can also use megasquirt just for the fuel injection system.

I have done it this way as I always wanted to get the FI side running well before handing control of the ignition side over to megasquirt. In such a case you just leave the normal ignition system of the bike alone, the pickups still tell the TCI when to fire the coils. All megasquirt needs to run the FI is a single feed from the TCI. In my case I paired the two outputs to the coils together with some diodes before sending it off to megasquirt. It works very well.

[81xsproject]

That's interesting as I went completely opposite. I built the whole thing, then made sure it ran good with MS running the timing while I still had carbs on. It was a quick one time ride. The ignition worked fine. So, I went home and started hooking up the FI system. The bike was already wired for it so I was already half there.


clcorbin,

WOW! I am very interested in everything you are doing. Almost all of it is stuff I wanted to do, but didn't have the knowledge or resources to do.
How exactly do you plan to utilize the secondaries for idle control? Are you talking about removing the primaries? It seems like it would be a lot simpler to use an idle valve to add that little bit of extra air into the intake boot ports.
I am VERY interested in your DIY board. What version to you plan to build? The size of the MS box is one of the things I dislike immensely. I am also interested in a your 36 tooth wheel. PM if you are interested in possibly making more than one of all these things and maybe I could help bring your costs down. I am trying my hand it hobby-tronics myself. I found a great site to order your board prefab. $5 per square inch gets you 3 copies of your board for a 2 layer, there is also an option for 4 layer.

[natemoen]

Do a short run of the wheels for several people perhaps and really bring costs down. That wheel would be one is my biggest obstacles for fi so I might be interested in one as well...

[clcorbin]

How exactly do you plan to utilize the secondaries for idle control? Are you talking about removing the primaries? It seems like it would be a lot simpler to use an idle valve to add that little bit of extra air into the intake boot ports.

The primaries would still be the primaries, controlled by the rider. If you look at the stock secondary setup, it has a cam on the end of the shaft that is used to control the fast idle during cold start. After that, the secondary butterfly valves are used as a persuedo constant velocity setup to keep the low speed throttle response from being TOO aggressive (among other things).

I'm not too worried about low speed air velocity with this 1100, so I plan on removing the secondary butterflies completely then ONLY using the stepper to control the fast idle cam. I'm not sure exactly how the GM IAC code works, but I bet I will be able to smooth it out to work with this setup, especially considering it also has a throttle position sensor for that shaft as well.

I am VERY interested in your DIY board. What version to you plan to build? The size of the MS box is one of the things I dislike immensely.

I would LOVE to build V3, but Bruce and Al don't plan on releasing the schematics to that version, so I am "stuck" at V2. I will base it about 90% on the actual MicroSquirt schematic including things like pin outs to the tyco connector. There will obviously be changes as I will be adding some hardware (IAC, MAP sensor, coil drivers, USB, etc) and deleting other stuff (serial, CAN, etc.), but I want to keep it as close to "stock" as possible. If I am lucky and take enough time with the layout, I SHOULD be able to keep the board down to about half again larger than the original.

I am also interested in a your 36 tooth wheel. PM if you are interested in possibly making more than one of all these things and maybe I could help bring your costs down.

I am in a ME class this semester that gives me access to the machine shop. It's not a commercial shop, so I don't know how well Dave would respond to me whipping out several reluctor wheels instead of just the one. On the bright side, cost is pretty much time+materials.

We have a very nice CNC lathe and CNC mill, and I have the CAM software, so I might work up the files to make the stem on the lathe and the plate on the mill so it just has to be welded together. I just have to make sure I get the notches lined up to the notch on the stem when they are welded together. If I'm smart, I'll design in alignment notches in the stem and wheel so they can ONLY go together in correct alignment.

I am trying my hand it hobby-tronics myself. I found a great site to order your board prefab. $5 per square inch gets you 3 copies of your board for a 2 layer, there is also an option for 4 layer.

That sounds like a decent price. I will DIY my 2 layer board for testing, but once it proves itself, I would be open to making a few more of these boards for the group (and to replace my test board!) IF Bruce and Al give me the ok. When I build a commercial board, I definitely want to make a 4 layer board. It is SO much easier to keep noise and size down with four layers instead of just two.

The only caviate here is I would only be making bare boards available. I just don't have time to assemble and test them. Given I will be using small SMD components, you will have to have access to a small reflow oven (or make your own!). I made one from an 8$ Good Will toaster oven and a PID controller of my own design that works great for everything I have "baked" so far, down to MLF packaged parts.

[81xsproject]

I'll see what kinds of bribes I can come up with for getting one of those wheels from you other than my tune files and maybe that Pro version TS. You don't need to go all crazy when connecting the shaft to the wheel. MS comes with the ability to 'correct' it in the software. For instance, I can't remember why, I probably screwed up, but I have my base timing correction at -59 deg. There is a Wizard in MegaTune that helps you do this with a timing light. If you have the abilities (and it sounds like you do) keep the option available on the board to have PMW Idle Valve Control. I do like your idea for using the secondary controller to run that cam, but I can tell you first hand that even in my coastal climate where I never start my bike at temps less than about 50 degrees F, that I have had to adjust my idle set screw farther open than that cam opens the throttle. Something to think about. PS, are the 08 TB's still 38mm?

I only have 2 complaints about MS. The size of it (obviously a result of TH components), and it's like they refuse to add a launch rev limiter (which I never have understood).

Another ps, would you share the toaster oven controller?

[Haarstad]

Do you know how much this EFI project cost you from start to finish from the stock bike?

[clcorbin]

You don't need to go all crazy when connecting the shaft to the wheel. MS comes with the ability to 'correct' it in the software. For instance, I can't remember why, I probably screwed up, but I have my base timing correction at -59 deg. There is a Wizard in MegaTune that helps you do this with a timing light.

That's nice to know. But simply from an engineering point of view, I want to make the hardware CORRECT so I don't have to fudge it in software. But it IS nice to have a backup...

If you have the abilities (and it sounds like you do) keep the option available on the board to have PMW Idle Valve Control.

Probably. I would have go over the pinouts on the PWM IAC to see what pins it uses. I MIGHT be using it for the stepper motor, but at this time at least, there are about 6 pins free on the Tyco, so it should be do-able.

I do like your idea for using the secondary controller to run that cam, but I can tell you first hand that even in my coastal climate where I never start my bike at temps less than about 50 degrees F, that I have had to adjust my idle set screw farther open than that cam opens the throttle. Something to think about.

Worse case, weld up the cam and grind it to a higher lift profile. It shouldn't be too hard to get the amount of "throw" we need out of this so the PCM can fully control the idle speed.

Oh, and one other thing I want to look into: have an output from my (future) MCU based voltage regulator connect to the PCM so it can tell the PCM to increase engine idle speed if the voltage drops below a certain level but the field is fully energized (aka: max output at that engine speed). It MIGHT be able to help keep the battery charged if the alternator just can't put out enough juice. I'll probably design the input into the PCM and then work out the software details at a latter date once everything else is up and running and I want something to mess with.

PS, are the 08 TB's still 38mm?

38/39. Depends on how I round the numbers. The spacing is 72-77-72, as measured by me.

I only have 2 complaints about MS. The size of it (obviously a result of TH components), and it's like they refuse to add a launch rev limiter (which I never have understood).

It is WAY too big for bikes! As for the launch limiter, I would assume there are some liability issues involved. But that is JUST a WAG and nothing more.

Another ps, would you share the toaster oven controller?

Sure. It was a fun project to make. I used triac controls modules (1 for each heater element to spread the power over two triacs AND to allow me to play with individual heating control if I wanted to) and a single control board that has an AtMega168 micro and a Max6775 k-type thermocouple interface IC (serial communications to the MCU and fully cold junction compensated) and a Max232 serial IC formed the brains of the package. An old wall wart gave it's all to become part of the power supply for the micro (the HARDEST thing about building projects for plug in "stuff" is the bloody DC power supply!).

I wrote the code (based on code from several other sources for the the various bits such as MAX6675 reading, UART I/O, 7-segment driver, rotary encoder, PID loop, etc) for the thing in Visual Studio and compiled it with WinGCC for AVR.

It will take me a bit to get things "cleaned up" for publication, but I would be happy to let the beast loose on the world. I could also include the PCB images so you could make your own boards (toner transfer works great when done properly).

[81xsproject]

Right now, I am only wishing we had the bowing emoticons...

[XS1100 Newbie]

I feel like i should have stayed in school past the first grade. I do love it when knowledge, ability, need, and want come together.

[clcorbin]

81xsproject,

I have uploaded the code/schematics/board images:

This file contains the Eagle schematic and board files, along with the image files I created from them to make the boards using toner transfer. The board images are setup to use 1200dpi, so if your laser printer is less than that, you will need to change the resolution of the images in some image editing software before you print them or they will be WAY too big.

AtMega168 based Reflow Oven Eagle schematic and board files

This next file contains all the source code along with the final compiled Reflow_Oven.hex file you can use to program you AtMega168 if you implement this project as is. I am fairly annal about commenting my code, so you shouldn't have too many problems figuring out what I was doing/trying to do. If something hangs you up, I would be happy to help you figure it out.

My programming environment is Visual Studio 2010 and this includes the full project files. If you use AVR Studio, you will have to create a project and add all the files to that before you can compile it. I only use AVR Studio to upload the hex files, so I won't be able to help much with that. Also, this project uses WinAVR (GCC C compiler for AVR microprocessors), so you will also have to have that installed if you want to compile it.

Reflow Oven source code and hex files

All in all, this was a fun project and I LOVE the results from it. Here are a few pics:

Overall picture of the oven:
Overall front view

The power switch, 7-segment display and rotary encoder:
Power switch, display and encoder

The encoder, display and serial port:
Encoder, display and serial port

Thermocouple inside the oven:
Thermocouple inside oven

Heatsinks on the back for the triacs:
Power module heat sinks