Megasquirt ECUs

[proximity]

Anyone had any experience with them?

Im thinking of converting my bike to fuel injection, i think it would be pretty cool. Has anybody tried this with success or do i not know what im letting myself in for.

J
____________________
2004 Aprilia Tuono

[0ddball]

Not worth the hassle imo. A pain in the ass to even get it running on the base map and then finding someone who can map it properly is tricky. It's better now there's more users and support but you still need to have a very good idea of how FI works.

Emerald are doing a limited function cheaper version of their M3d ecu. No boost control, double injector drivers, traction control ect but includes all the features you'd ever need on a bike. Plus the best bit, Dave will fully map it for £235 no matter how long it takes.

But all in all it's a lot of hassle for such small gains. There's other bits to sort like making a decent loom, mounting a crank sensor and trigger wheel, tps, ats, cts ect. Unless you run a competition bike where you are always changing and fiddling with things to get that extra 1% mappable ecu's are a waste of money.

[Pete.]

As Spannermonkey said you're looking at a lot of time (as in many hundreds of hours) researching, building and setting up your fuel injection. I know of a guy who has taken zx12r fuel injection and fitted it onto his old GS thou, perhaps it would be easier to find a bike with similar engine output and characteristics and do the same. Would most likely be cheaper in the end too.

[bish777]

Looking into most of the costs (as i did when i thought of FI'ing the GSXR1100) youd need to get an electric FI pump-normal fuel pumps for carbed models dont have the muscle, make a loom, aquire a set of throttle bodies and injectors to fit, throttle position sensor etc etc

at the end of it i reckon its not worth the hassle, for the real end cost you could buy a brand new set of mikuni RS34 carbs and get dyno time.
____________________
GSXR1100J

[Pete.]

So long as you could find throttle bodies with near enough spacings, the GSXR or hayabusa stuff is pretty easy to chop over. The fuel pumps have an internal pressure regulator, throttle bodies, injectors, injector loom, tps all come as a "module". You would still have to make a crank trigger and cam trigger and sort the various sensors though.

[proximity]

What do you mean a crank trigger and cam trigger?
____________________
2004 Aprilia Tuono

[0ddball]

The ecu need to know the position of the crank so it can fire the injectors and plug at the right time. It's a toothed wheel like this

https://www.seanhylandmotorsport.com/online/images/8400.jpg

That need to be mounted very precisely to the end of the crank, i guess it would have to go on the flywheel on a bike.

And a sensor like this

https://www.siemensvdo.com/NR/rdonlyres/B151E880-0F79-4D15-B831-1055312575BF/0/sv_S107230001Z_192x146.jpg

Which has to be mounted so that the teeth of the trigger wheel pass it closely which either creates a sinewave signal by induction, or switches an applied voltage to create a square wave depending on whether you choose to go analogue or digital.

The missing tooth is what the ecu uses for reference. You mount the wheel at say 20 degrees btdc on the crank then tell the ecu that the missing tooth is at that position. That way it knows everytime it senses the missing tooth the crank will be at 20 degrees btdc and it can reference all it's calcualtions based on that.

The primary trigger most often used is the crank sensor, you can also use a cam sensor (same idea but mounted on the end of the cam) either as the primary trigger instead of the crank sensor, or as an aditional reference for the ecu is you want to have sequential injection.

[ROB_NSR]

Well if you decide you want a megasquirt, I have one all built up but never used, with megasquirt 'n spark - extra software on

I bought it for my 205 turbo project but I've given up on it!

You can have the ms for £180 (or the full car Sorrento green 205 16v turbo, unfinished, for £900!)
____________________
CBT: PASSED!! THEORY: PASSED!! PRACTICAL: PASSED!!

[Pete.]

I know of a guy using GSXR600 throttlebodies and Mini-megasquirt to convert an XS1100 to fuel injection.

https://www.mini-ms.com/

If you need a trigger-wheel or sensor connectors/sensors, you can get them here at very good prices:

https://www.trigger-wheels.com/#575lightweight

[edd]

you only need a crank sensor if you are running sequential injection, non sequential injection doesnt require one. TBH for the extra effort required to go sequential I wouldnt bother. A great deal of setups use non sequential injection to great effect. You also dont need a crank sensor for ignition If you are happy using the existing one. (presuming your bike runs electroninc ignition, which in your case I know it does.)
____________________
Bluespark Automotive - Diesel Tuning for Performance and Economy

[0ddball]

[proximity]

I just found this on the megasquirt site;

[edd]

Im not really sure what you mean by a reference signal. You can build a 3d map using just either throttle position, MAP or airflow sensor plus RPM (more factors help you to create a more accurate map, but these are all that are neccesary) . Your RPM is extremely accurate form the halls sensor on the crank, it has to be otherwise your spark wouldn't occur at the right time. What it doesnt tell you is when to inject for a sequential injection system because it only gives one pulse every revolution. Sequential injection does most certainly require a cam or crank sensor to determine position. Without this sensor there would be no way to tell whereabouts the intake cam was, mostly because the intake valve only opens every 2nd revolution.
____________________
Bluespark Automotive - Diesel Tuning for Performance and Economy

[DucatiEVO]

Don't know about a bike, but I know a few peeps using MS on their cars, I intend to get one also for my project Quattro. Darn sight cheaper than a motec and just as tuneable.


____________________
aKa: Ducky

[0ddball]

[Pete.]

Sorry DucatiEVO but comparing a Megasquirt to a MoTeC is like comparing a primary schoolchild's finger-painting to Constable's The Hay Wain. To supply, fit and map a MoTeC M400 on a bike in the UK will cost you in excess of 3 grand. Do you really think they would sell any if the same could be done with a few hundred quids worth of Megasquirt.

Now don't get me wrong I'm not knocking the MS, I'm just saying that it doesn't even come close to stacking-up to the MoTeC.

[proximity]

Well i understand what you are saying spannermonkey, but the megasquirt doesnt even have an input for a crank sensor. It doesnt support sequencial injection.
Also i dont know why you would need dyno time, with a wideband o2 sensor and a laptop you could do it all yourself with an empty peice of road.

To me it seems to have plenty of software and support in the community to help get it set up right.
They are only $140 for a kit, which is pretty cheap compared to a motec.
____________________
2004 Aprilia Tuono

[Pete.]

Yes you could do it with a wideband and laptop but to stop you going insane you'd be well advised to get hold of a Datalogger too. A wideband commander would not only give you real-time A/F readingsbut ha a datalogging feature also.

[edd]

Id agree that EFI would be better with the system you propose, and I also realise you know what you are talking about. I have read up quite a bit on EFI, and I'm pretty sure early EFI bikes did use a halls sensor and non sequential injection. For example the GPz 750 R1. I know your EFI system must react fast, but your spark must be in exactly the right place every revolution. Even at 3000 RPM you are doing 50 rpm per second and as such getting 50 sparks a second. If you take the RPM signal from the CDI and run that into your megasquirt it should be able to calculate the gap between pulses and give you a result in 1/25th of a second (processing time is negligible). Thats not uber quick, And probably wouldnt give you as smooth fuelling as you'd like. At 6k I think it would probably be relatively unnoticable. Again I know it wouldnt give the butter-smooth EFI you guys are probably accustomed to, but I think it would work. At least I cant see any reason why. I may well have got these calculations wrong if I have please tell me.
____________________
Bluespark Automotive - Diesel Tuning for Performance and Economy

[0ddball]

It does have an input for a crank sensor, that is certain. Going by the info you found it doesn't have an input for a cam sensor which tbh is a pain in the ass to fit to a bike and apart from lower emmision and slightly better economy sequential injection has no real benefits over banked injection on a standard'ish engine.

Plenty of people think it's easy to map it out on the road, it's a different story when they try it and a wideband kit isn't much use unless the user knows what values to aim for. With the right equipment (don't forget you'll need det cans too for the ignition) and a good basemap it's the best way to map. But it's tricky in a car with a passenger working the laptop, god knows how you'd do it on a bike. In all likelyhood it would probably take you a couple of hours just to get the engine fired up and ticking over smoothly. You then have to make an educated guess to roughly fill in the first few load sites of the fuel and ignition maps so that it can drive slowly up the road without cutting out or coughing and spluttering. Easy enough if you have experience, a complete nightmare if not. A basic wideband set up and det cans cost more than the dyno time anyway so unless someone had other engines to map it makes a lot more sense to just trailer the car/bike to the mappers.