[TheSteve]

After not getting much help on the MS boards, I'm thinking of a different way to achieve the same result. The MS lets you specify a resistance curve so you can use a custom sensor. If I use my current sensor but change the data points, it could read 350F but report something totally different but not maxed out. Right now if the engine is 200, its reported to the MS as 200 but itll top out at 215. Changing the curve could make it so if the engine is 200 it reports 90 (for example), and it would effectively top out at 325 while reporting 215. I'm sure there are some ill-effects from this, but I think it should work just fine. MS only uses the coolant temp for warmup and cranking so it may give it the boost it needs without hurting much else.

One of the crappy things about MS1 is that the crank pulse temps are preset (with a 160F max) and cannot be changed. With MS2, you can generate your own startup curve with temperatures up to the full 215. What this means is that to MS1, for cranking 130F = 130F, 160F = 160F, but 180F+ = 160F. For MS2, since you set the temperatures to whatever youd like you dont hit that limit.

What this means is that to make this start an engine accurately, I'd have to drag the CLT values down to maybe 180 degrees at the absolute most while HOT. A possible problem with this is that the engines cold temp might be south of the established minimum of -40f and then instead of hotstart problems I'd have coldstart problems. This problem would be helped by use of the MS2 since the values would only have to be pulled down to 215 instead of 160.

Theres also the possibility that I could set the sensor curve so that -40 (sensor) = -40 (MS) but 350 (sensor) = 160 (MS). That might create a pretty funky looking curve though leading to all sorts of weird startup issues.

This is all theoretical at this point, I haven't done anything other than post up on the MS forums yet. I'm moving tomorrow so it may be some time before I have a chance to really play with it. Couple that with the bike being my daily driver I'd rather not accidentally make it worse on a day where I need to be somewhere.

[ericej2344]

sounds interesting, let us know how this turns out.

[TheSteve]

Time for a small update. The wideband has been removed from the bike and it still runs as well as ever. I did not install a narrowband sensor, instead I just blocked it off. Since it still runs fine with no o2 correction it means the maps are spot on.

I've been pondering the startup issue and I think I've figured it out. The bike only has issues on hot starts, like after parking after sitting in traffic then trying to start it back up 5 minutes later. So the engine is probably borderline overheating. As you know, these oil cooled engines already run hot. Overheating means the engine is probably pushing 350 to 400 degrees. The megasquirt, due to limitations in programming (max value for integer storage is 256), can only understand temps up to about 215. This is because the code is set so the min temp is -40. After a hot shutoff, the engine is well above that 215 mark, but the MS thinks its no higher than 215.

The ms bases starting injections on temperature, so since a hot engine needs less gas to start, the 350 degree motor floods when given the gas it'd need at 215.

The solution is to bump up the minimum temperature that megasquirt can see. I live in Arizona, so if I can push the min temp to +40F that should raise the max temp to almost 300F which should be good enough for these hot starts. To do this id have to make changes in the MS code itself I think. I don't know how to do this, but I'm going to check the ms forums to see if its possible. Ill keep you updated as I find out more.

[ericej2344]

Even though its a small problem, let us know if anything changes with that cold start bug.

[Adequateman]

Might be worth taking a wire brush to the plugs. Or replace them, if they are old.

[TheSteve]

Well, its been quite awhile since I've last posted so I figure I'll give you guys a small update.

I believe I'm about done tuning, as every time now when I go out to tune the numbers change by less and less. The bike rides like a factory original to the point where I'd have no problem loaning it to someone for a few days (as in I wouldn't worry about it leaving them stranded). Power is very linear, no coughing or choking, usually starts on the first crank, etc etc. Mileage has been right at 45-48MPG throughout the winter. I'm not sure if the decreased mileage is from the cold or my riding style. Having just started riding maybe 7 months ago, I'm sure I started going faster now that I've gained comfort with the machine. In addition, not a day goes by where it doesn't get up to 85-90MPH on the freeway. Were I to actually maintain speed limits I'd likely be back up to 55+MPG. Might be worth a tank of gas to see if mileage goes up if I slow down.

I still have the occasional problem starting, but I'm beginning to think it isn't entirely the fault of the Megasquirt. Left to sit for a few days (especially when cold), sometimes it'll crank, fire, run for 3 seconds or so, and die. Then it'll refuse to restart without the aid of a car battery. My only guess is the afterstart enrich is fouling plugs and it really needs HIGH voltage to arc across them till it gets going. Not sure theres much I can really do there, because if I lean it out it'll die on its own, and any richer and it tends to flood. It's not a big issue unless its really cold outside.

I feel pretty confident in the current state of tune though, and am about to pull the wideband out to begin retuning my MS'ed truck, whos mileage went south without the wideband to guide it. The Katana, on the other hand, was set to NOT rely constantly on the wideband. Meaning once unplugged it should get the same mileage and same throttle response. I don't plan on replacing it with a narrowband O2 sensor. In my opinion they are a complete waste of money, especially since my tune rarely sits at 14.7:1 AFR. Instead itll be replaced with a simple threaded block off plug allowing for an easy reinstall later if needed.

So, has anyone else made any progress on their EFI Katanas? Post up!

[TheSteve]

Lol yeah.. being 20 and working at a movie theatre isn't a whole lot to brag about.. Gotta go back to school first, then hopefully I'll be doing something a hair more interesting than fixing projectors.

[GytRDunKat]

Dude, you need to find a different line of work. I don't know what you make, but it should be 3x what your making now. I would suggest one of the big 3 automakers, but that gig just might not work out. You need Monster.com or something... Sheesh!!


You've missed your calling dude, go find it again! It's calling, and IT don't know your number. This was like a mosqito in the net for you. Go get em tiger!

[TheSteve]

Hmm. Possible good news for someone out there wanting to 'squirt their katana. I picked up a new unrelated project, and need to upgrade one of my megasquirts to the 3.0 board for VR decoding. So I'll likely be selling one of my ECUs soon. It's an MS1 cpu on a 2.2 board, slightly modded to run a single spark output. While the spark output wouldn't be very useful on a katana (the kat needs two outputs) it doesn't interfere with normal operation of the ECU. If anyone is interested, let me know via PM. It's been in use without issue for just over a year now.

EDIT: Actually, I'll have one of two for sale. One is the one described above, the second is a untouched MS1/2.2 kit from diyautotune at a discounted price. One of my friends needed me to build him one, I could save myself some time by selling his kit and giving him my assembled one. Diyautotune wants $140+shipping for their kit, I'd do $140 shipped. Their price for an assembled one is $240+shipping, I'd sell mine guaranteed to work for $220 shipped (and pretuned for a Kat, just add wiring). Remember I can only sell one of these, the other goes to my buddy.

[TheSteve]

I've done a full setup before with ignition timing because I wanted to ditch the stock ECU. Here, since it was a little more complicated I figured I'd just leave the timing control to the CDI. I just didn't feel like messing with timing, to be honest. If you're out for balls-out power, you'll want to do it. If you just want reliability and a hair more power, you don't need it.

As for the GSXR/bandit/whatever ignition module: timing advance is very different between different engines. Bore, stroke, compression, operating temp, valve shape, etc all affect ignition timing. Chances are good that your engine that you're building is going to be too far away from the specs of any of those bikes to make it a bolt on power improver. It'll run, but not as well as it could.

I don't know if theres really a way to log the mechanical timing against actual advanced ignition timing either, so it might be tricky. The easiest way is probably going to be to intercept the CDI-advanced signal, and add in more advance of your own as you need it. I made this sound like a very bad way of doing it above, but honestly you'll get the same result. The only real "bad" thing is you're limited on timing retardation. The maximum limit on retard is going to be the CDIs advance (opposed to 0 degrees before top dead center). So its not gonna be great for traction control, but you could still do a mode that keeps power somewhat contained..

http://www.megamanual.com/index.html Heres where to start reading. Going down the index on the left, look for all the different ignition systems. That'll explain the HEI, DIS, and EDIS. I looked a bit already and I'm not sure it'll really apply to Katanas unless you go absolutely crazy with it and decide to put in the legwork to adapt a custom trigger wheel. And CDI is the little ignitor box under your seat. Its actually a misnomer; its NOT a CDI type ignition! No idea where it got its name, but it stuck..

[il_ragazzo]

wow what a informative response!

Its possible. Theres a Nitrous function where when the nitrous is activated it uses a different map. You could rig up the nitrous mode to run a table with very little ignition advancing at high RPM. Thatll make it pretty gutless above 4000rpm or so (or wherever you stop advancing it). You can run a simple switch and tell the MS that when it goes hot, switch to table 2. Note that this is a more advanced install: timing isnt something you want to mess with while still trying to get it running. You could set it up so that it stops delivering fuel altogether above X load and Y rpm but that could lead to issues such as the inability to downshift and move when you have to.

I believe the way the GSXR does its low power modes if with modulation of the secondary throttles and messing with the ignition timing. You can't just cut fuel somewhat, youll run lean and blow. If you set a hard fuel cut, you won't be able to accelerate out of the way in an emergency. The correct way to do a power cut is with a drive-by-wire system where the computer controls the throttle. Ford does this often nowadays, but its one of the most complained about features (lag, etc). Not something you'd really want on a bike.

Long story short, if you really want it on a katana, I wouldn't change the rev limit much. Maybe set it at 7000 or so, thatll still have you with a max speed of 90ish. I'd run dual table mode with ignition timing retarded severely. With as high as this engine revs, cutting the advance out should stifle the majority of the power, especially above 4000.

Edit: I forgot to add something. If you ARE controlling timing with the MS, you dont really need the new CDI for your motor. Since youre making your own timing map the preprogrammed map in the CDI is irrelevant. Now heres where it gets fun: No one knows the stock map. Triggering off the CDIs coil outputs is going to give the MS the advanced signal. To do timing control, you need eith3er the raw TDC timing or you need to know the stock map to translate the advanced signal to the raw one. Personally, if I was doing timing Id scrap the CDI altogether. Id make or buy a new trigger wheel and install that where the current one is. Itll have to be 4 cyl wasted spark capable, maybe the Ford EDIS system? Maybe the GM DIS type? Not sure, but you'll have to figure something out if you want to do it RIGHT. Right in this case meaning mathematically correct timing. You could just wing it with the stock advanced signal, add timing until you get pinging then back of a degree or two and call it good. Honestly either would work, but one lets you know exactly how much advance youre running, the other just says 10 degrees plus X (X being the stock CDIs advance) equals it running well.

Edit again: The EDIS ignition system is looking for a wheel with a missing tooth, of a certain count (36 teeth I think). The Katana wheel has 4 teeth, one longer than the others. Maybe it'd work with the DIS system, I'm not too familiar with it. What I do know is that the wheel can be picked up and read by other GM systems (HEI) but they need a distributor. Maybe you'll get lucky, I'll look around and see what kind of wheel it needs, maybe the stocker will work. Don't get your hopes up though.

ok well it will be a while (prolly about 4 months) before i really start my EFI swap anyways, so im not to worried about it i'll figure some way of doing it. but its good to know that the MS constroler has the capacity to do this function.

so your setup doesn't control ingnition? And whats stoping us from taking the IGN controler, ect from a gsxr , bandit?
couldn't i splice a conection from the cdi to the ign controler for the ms and have it record the IGN timing then tinker from there?

Finding a way to adjust IGN timing will open alot of doors for things like traction control and tuning...

To be honest right now all of this is above my head and i have alot of reading to do. but i dont even know what half of the systems your talking about are or what they do. I.E HEI, DIS, EDIS, and countless others in this thread. Hell i dont even know what the CDI stands for! But my passionate hatred for carbs will drive me to do this mod. thanks steves!

[TheSteve]

Its possible. Theres a Nitrous function where when the nitrous is activated it uses a different map. You could rig up the nitrous mode to run a table with very little ignition advancing at high RPM. Thatll make it pretty gutless above 4000rpm or so (or wherever you stop advancing it). You can run a simple switch and tell the MS that when it goes hot, switch to table 2. Note that this is a more advanced install: timing isnt something you want to mess with while still trying to get it running. You could set it up so that it stops delivering fuel altogether above X load and Y rpm but that could lead to issues such as the inability to downshift and move when you have to.

I believe the way the GSXR does its low power modes if with modulation of the secondary throttles and messing with the ignition timing. You can't just cut fuel somewhat, youll run lean and blow. If you set a hard fuel cut, you won't be able to accelerate out of the way in an emergency. The correct way to do a power cut is with a drive-by-wire system where the computer controls the throttle. Ford does this often nowadays, but its one of the most complained about features (lag, etc). Not something you'd really want on a bike.

Long story short, if you really want it on a katana, I wouldn't change the rev limit much. Maybe set it at 7000 or so, thatll still have you with a max speed of 90ish. I'd run dual table mode with ignition timing retarded severely. With as high as this engine revs, cutting the advance out should stifle the majority of the power, especially above 4000.

Edit: I forgot to add something. If you ARE controlling timing with the MS, you dont really need the new CDI for your motor. Since youre making your own timing map the preprogrammed map in the CDI is irrelevant. Now heres where it gets fun: No one knows the stock map. Triggering off the CDIs coil outputs is going to give the MS the advanced signal. To do timing control, you need either the raw TDC timing or you need to know the stock map to translate the advanced signal to the raw one. Personally, if I was doing timing Id scrap the CDI altogether. Id make or buy a new trigger wheel and install that where the current one is. Itll have to be 4 cyl wasted spark capable, maybe the Ford EDIS system? Maybe the GM DIS type? Not sure, but you'll have to figure something out if you want to do it RIGHT. Right in this case meaning mathematically correct timing. You could just wing it with the stock advanced signal, add timing until you get pinging then back of a degree or two and call it good. Honestly either would work, but one lets you know exactly how much advance youre running, the other just says 10 degrees plus X (X being the stock CDIs advance) equals it running well.

Edit again: The EDIS ignition system is looking for a wheel with a missing tooth, of a certain count (36 teeth I think). The Katana wheel has 4 teeth, one longer than the others. Maybe it'd work with the DIS system, I'm not too familiar with it. What I do know is that the wheel can be picked up and read by other GM systems (HEI) but they need a distributor. Maybe you'll get lucky, I'll look around and see what kind of wheel it needs, maybe the stocker will work. Don't get your hopes up though.

[il_ragazzo]

ok so ive just bought a 1052 motor cdi and harness yesterday for only 180 so my next step it to start thinking of and looking for EFI parts... I was thinking about how much power i will be putting down to the ground and thought perhaps i should have a rain setting or a half power button... is this possible to do with the mega squirt/ micro squirt system and would i need an external switch? is there a way to have external controlers pluged into the MS system?

[TheSteve]

Ok now that I'm finally at a real computer and can see the pics:

Those are definitely later throttles. These are the early type. http://cgi.ebay.com/ebaymotors/01-02...Q5fAccessories

Dont fret though, the reason I didn't want the later style was because on the 04 you cannot alter the spacing between 1-2 and 3-4, but only between 2-3. On the 01 style each throttle body is separate so you can space them all out however you want. But, I ended up only having to gap between 2-3. The other's spacing was fine, and I didn't have to touch them.

Looking through other listings on ebay, I see where the confusion came from. The literbike's throttles went to the "later" style a year before the 600 and 750. So you'd want 01-03 600/750 throttles or 01-02 1000 throttles to clone my setup. I think you'll be alright though.

Again I stress that the best place to buy these would probably be from GSXR forums. The secondary butterflies fail very often, and suzuki will not sell them separately. So people end up pretty much throwing them away when theyre still perfect for something like this.

[Newbie2it]

Yes, The STVA controls the secondaries. I didnt realise you'd removed them during the build. In this case it would be useless including the STP. My TBs are on a K4 which is a plastic fuel rail. Simular to the picture above. But the rest remains regarding twisting the "T" fitting on the plastic rails. Dunno if theres a way to "Shorten" the rail like you had to do with yours.

BTW, good to know the line you used is Gas safe.