Mega squirt users

steell wrote:

All of the embedded microprocessor code executed for MegaSquirt-ITM was hand-written directly in assembler, not compiled from a high-level language

Assembler still gets compiled. And the GNU 68HC11 toolset would still be used for this. If you want to skip compilation entirely, you have to write in machine language.

steell wrote:

Heck, there is even a free demo of Code Warrior that will work, it has a size limitation but MS is well within the limits (I have the CD sent to me by Motorola )

I realize that the reason you missed this stuff is because the whole site has simply gotten huge, and to understand the whole MegaSquirt thing really takes a lot of time for someone new to it.

No. I actually tried to make some changes to the MS-II 2.886 code and compile it. That's where I learned that the current code set is MUCH too big for the free version of code warrior. Here's the thread where Al Grippo told me that: www.msefi.com/viewtopic.php?f=69&t=31952

steell wrote:

You stepped into the MegaSquirt World at the top level with the assembled MicroSquirt, you learn a lot more when you start at the bottom with the MS1 kit.

Perhaps I didn't get the experience of assembling the kit, but review my project page: I fabricated the intake manifold, fuel rail, and trigger wheel from scratch; I disproved Bruce's VR conditioner theory, provided considerable additional detail about the workings, and potential problems, of the Microsquirt VR circuit, and eventually tacked in a replacement conditioner with custom trigger clipping. I also had the dubious privilege of getting one of the first 10 Microsquirts released, which meant I got to discover and work through a lot of the initial code problems, and fun bits like the swapped VR input pins.

I wasn't impugning your experience, I was referring more to the history and background of MegaSquirt. Heck, anyone that has a MS up and running on their bike obviously knows what he's doing

Since you were lucky enough to get one of the first ten MicroSquirts you got to have the fun of chasing bugs

Surprising that the MicroSquirt still had all those problems after the extensive beta testing that was done

Fortunately, most (but not all) of of the KZ's have a tail piece that will hold a MS1 or MS2 quite nicely, so we get to do it with the cheaper versions.

I had a 1981 Yamaha XV920RH (Yamaha's bastard SECA) so I know how little space is available.


Edit:

Finally got the link to work (site is really slow) and read the thread, damn code sure has multiplied since I was messing with it


The MS Extra code was recently ported to MS2 (recently being sometime in the past two years), so the MS2 Extra crowd may be able to help you compile the code since they have recent experience.

Yeah. Most of the problems they had with Microsquirt were around the VR input and the ignition drivers. They now have a 2.0 revision out with improvements in these areas, though I think they still have VR input problems. Bruce has aluded to some work they're doing on a better one-size-fits-all VR conditioner solution.

Code-wise there were issues around IAC control, injector test mode, and dual tach handling (that come to mind off-hand). I got to chase two out of these three. Still I'm quite happy with the unit; I was just a bit frustrated at being the first to have seen a lot of issues on the forum. I can say that Al and Bruce were very helpful and responsive, though. If I do this again it will certainly be a lot easier with the lessons now learned. I think I've convinced one of my co-workers to FI his '57 MGA... that should be a fun project

Sorry if I came off a bit touchy, but the comment about missing stuff on the site kind of irritated me. I may not know the code as well as the MS2Extra crowd, but I'll bet I know it better than almost everyone running Microsquirt. I am an electronics technician and software engineer by background, so my first instinct when troubleshooting a number of the issues I ran across was to crack open the code and see exactly what MS-II was doing.

I also started arguing with you about the MS-I CPU because your post was in response to a question from a guy that knew nothing about Megasquirt. I wanted to clarify that there are a lot of differences in the available functionality and the method of implementation between MS-I and MS-II, even though they use the same base board.

Interesting to hear that there's more room to work with on the Kawasaki's. I did see one guy that put MS-II on his Yamaha XJ1100 ( users.hal-pc.org/~dhutch/ ), but he got very creative about mounting the case under the seat. Microsquirt worked out really nicely on mine in place of the ignition module it replaced. Actually, I looked at JAW, and even the size of the controller board (and finding a case for it) would be a concern for me on the Seca... I also found a deal for the LC-1 kit for $178, which would seem to me to be a better choice for a permanent Yamaha installation. At this point, though, I'm far enough down the dyno tuning path that I think I'll do one more session and leave it at that.

Cheers,
Paul

And how has the dyno tuning worked out? I'd have to truck the bike about 70-80 miles to do that, the local dyno guys just want to cut new fuel maps for late model bikes, they know nothing of MegaSquirt and have no desire to learn.

How's it running, power and economy wise, with the MicroSquirt? I thought about buying one, but couldn't see any sense in spending the extra $$ when MS1 will do all I need for the bike. Had to spend the extra $$ for an MS2 for the car because I needed stepper motor control for the IAC.

I did spend an extra $40 or so for a Peak and Hold driver board that fits in the case with the MS so I wouldn't have to use resisters or flyback with low impedance injectors. I couldn't see any sense in wasting the bikes meager watts as heat from the resisters, I need every watt I can save


Oh yeah, upper right side of this page has a thing that says "Private Messages" click on it and you can read the message I sent you earlier. It's right under the "Logout" button.

Luckily there's a local dyno guy that really understands engine management systems and is interested in projects like mine. He was a lot of help with the first dyno session I did. The project page has the full throttle curves from that session. Power in that run was probably on par with stock, given rear wheel losses and the fact that I'm at about 6000 feet. Economy-wise I was getting about 45mpg last time I checked. I think I can still do better.

That run was before the bike was run over, before the 4-1 MAC exhaust, before the rebuilt injectors, and before the 36-1 wheel; I also need to check (and adjust?) my valve clearances, so there should be some room for improvement.

On the other hand, I was doing some work on the bike about a week ago. It was pretty close to 0 degrees Fahrenheit in my garage, and the bike started with a simple three second press of the starter button. That alone, regardless of my general hatred of carburetors, makes the whole project worthwhile to me (not that I'm planning on doing much riding at 0 degrees!).

You keep saying that MS-I will do all what you need. Are you doing fuel only? To me, the big drawback of MS-I (besides size) would be the inability to read a trigger wheel. With dual tachs (single trigger "finger" triggering the VR pickups off the crank) my timing was jumping around +/- 15 degrees or so of where it should be at idle. This was due to the light crankshaft, lack of flywheel, and all the variation in speed of the crank that happens at low RPM in a motorcycle... the code just didn't have enough information to properly predict when to dwell and fire. By switching to the 36-1 wheel, my timing accuracy is spot on at idle... much smoother.

The one other feature of MS-II that I may yet use (and it's one of the ones that's leading me away from MS2Extra) is X-Tau. I may try playing around with that a bit at the next dyno session.

While I won't use a Dyna ignition, I'm real happy with the stock Kawasaki electronic ignition, so I don't plan on running spark on the MS. But the MS1 will do ignition if I was so inclined, and Dave Sloan has been playing with ignition on his MS1 equipped KZ650/810.

From the Extra Code Manual:

MS1 Extra code is a version of firmware that has been developed by Philip Ringwood and James Murrey over the past few years. This code can be run on any version of B+G Megasquirt PCB (V1.0, 2.2 or 3.0) using the MS1 (68HC908) microprocessor. It is based on the B+G original V3.0 code and Eric Fahlgren's Dual Table (DT).

Generic Wheel Mode allows the use of a multi-toothed trigger wheel such as 36-1 (commonly fitted to Fords), 60-2 (used mainly by Bosch and therefore very widespread in Europe.) Custom wheels such as 6-1 or 4-1 also work.

www.msextra.com/manuals/MS_Extra_Basic_Configuration_Manual.htm

Too many people locally (dyno owners) already know everything and are not interested in learning anything new

X-tau is one of those things that people either love or hate, some swear by it while others swear at it. I haven't tried it, or studied it, so I don't have an opinion.

SQLGuy wrote:

I disproved Bruce's VR conditioner theory,

OK, I have to admit that I, at least, need to bow down to a superior intellect.
My background is Mechanical Engineering, the only course I flunked was electrical circuit theory, I can write code, but I always get lost reading somebody else's. I have learned just enough MegaSquirt to make it do what I want. So I haven't spent much time on the MS forums once I got a handle on what I thought I needed to know. So...
tell me, which VR conditioner theory are you talking about? Does it change my vr conditioner circuit I built two years ago?

One of the problems that most Microsquirt users had was the inability to rev past some RPM (in some instances quite a low RPM) without the system completely freaking out. This behavior was because the VR input was being overdriven, and the initial solution was to install series resitors between the pickup and Microsquirt.

Bruce's theory was that the problem was the result of D5 (a resistor in the VR circuit that clips the VR input at 5.7V pretty early in the input) going into reverse breakdown. His recommendation was to switch this to any 100V or higher reverse rated diode. I tried this with a 1N4003 and still had problems. I then showed that, unless Q4 is destroyed, it's not really possible for D5 to be driven into its reverse breakdown region by the VR signal. My theory was that at least some of the overdrive problem was due to recovery speed of the devices being used, and this seemed to bear out in that a 1N4148 worked better than a 1N4003, even though the 1N4003 has a higher reverse voltage rating. In my testing, I found that removing the filter capacitor at C30 also helped a lot.

If your VR conditioner is working fine, then none of this changes that... although, if you have external resistors you want to get rid of, you might want to check this thread: www.microsquirt.com/viewtopic.php?f=87&t=23063

Cheers,
Paul

steell wrote:

From the Extra Code Manual:

MS1 Extra code is a version of firmware that has been developed by Philip Ringwood and James Murrey over the past few years. This code can be run on any version of B+G Megasquirt PCB (V1.0, 2.2 or 3.0) using the MS1 (68HC908) microprocessor. It is based on the B+G original V3.0 code and Eric Fahlgren's Dual Table (DT).

Generic Wheel Mode allows the use of a multi-toothed trigger wheel such as 36-1 (commonly fitted to Fords), 60-2 (used mainly by Bosch and therefore very widespread in Europe.) Custom wheels such as 6-1 or 4-1 also work.

This is surprising to me. I guess most of my impressions of MS-I were based on the feature set of the Megasquirt code and not the MSExtra code. MSExtra does add pretty much everything to MS-I. Since MS-I can do most trigger wheels, and wasted spark, and PWM IAC, there really isn't anything I'm doing that couldn't be done by MS-I (other than fitting the unit into the space available).

The only other clear benefit I can see to the MS-II systems is the higher resolution it offers for most variables, which I could see being a benefit for my setup, where the injectors are a bit large for the application, so fine resoltion of PW at low RPM's is helpful.

MS2 has one big advantage for me, it will handle the stepper motor IAC in my car

Just finished reading your website.

Next time you need to cut/drill/machine copper, try milk for a lube.

If you were going to do it again, would you make your own intake or use a set of throttle bodies off a bike?

KZ's have it made, the 650/750/900/1000/1100 all share the same carb spacing, and three separate models (750/1000/1100) shipped with stock EFI.

I see you have the same missing watts problem as the KZ's, most KZ charging systems are rated at 20 amp. LED's help

Are you using oil temp instead of cylinder head temp? There are thermocouples available on eBay for $3-$4 that can be screwed into the head, or just epoxied to it. I think most recommend against using oil temp because it takes so long to change temp.


I applaud your determination, and admire the execution.

Thanks for your comments.

The main reason I made the intake manifold was that I wanted idle air control, and I didn't see any reasonable ways to do that using separate throttle bodies.

For this project I also bought a head temp sensor from Dakota Digital, but I didn't use it because it supposedly doesn't provide any useful information below 150 degrees Fahrenheit. The oil temp sensor provides good information about the cold end of the warmup curve and how much enrichment and air are initially needed. It may be preventing me from leaning out quickly enough when the engine has been running for a couple of minutes, but the IAC correction based on oil temp seems right. If I can find a head temp sensor that covers 20 to 350 degrees or so, then I may try recalibrating for that.

Temperature Measuring Range


-50-1500C

Is what the seller specified for the three I bought, 2 for the bikes and one to measure egt at the output side of the turbo on a Cat 3406B. They're standard K-type thermocouples that I read with my multimeter. I think they were around $7 each shipped from Hong Kong.