rpm signal way too high!!! need this running soon

[formula_sae]

Thanks for the help!!! Tonight ill check out the tach masking and work with some of the variables you suggested. I origionally had the cranking trigger to calculated so ill switch that back as well. Im not sure about the wires so ill check them as well. To be honest i just took this project over from another team member whos priorities dont seem to be with geared towards makeing this motor run so i wouldnt be supprised if there are values that are wrong in the setup. After i got the tach working right on the engine, the test light was my next step on the setup process. Thanks again for the help, ill be in touch later tonight with some results.

[formula_sae]

hey, i'm the bozo that's not here enough. (we're using one screen name for our whole team).

i got the crank pickup worked out. somebody messed with my original settings. 18 teeth, 1 hole, 36 skip teeth (1 cyl, 4 cycle). now it reads like it should, ~525rpm cranking. i have no spark currently, although the coil is getting +13.5V. i don't know if MicroSquirt isn't grounding the other side or if there is some hardware/wiring issue at hand. i've got the neg side of the coil hooked up to IGN OUT 1 and positive side to a confirmed, switched 12V source. i am directly driving the coil inverted. i seem to remember setting that quite a while ago after a little research. the plug looks fine, and is not covered with fuel since i unplugged the fuel pump. we're still working on the spark issue, but i figured i'd go ahead and give an update in case anybody just can't wait to lend some [experienced] brain cells.

[Martin Wagner]

If you don't have a second vr input, microsquirt won't know which crankwheel revolution is compression and which is intake/exhaust overlap. I would recommend setting skip teeth to 18 and fire the coil every revolution. If that messes the tach, call it a 2 cylinder.

[grippo]

I hate to tell you this but wheel decoder is designed for edge detection - rising or falling, but not both. Your wheel looks like it is designed for level detection. In wheel decoder, a tooth is the time between edge detections. In your case, taking the lengths as times, you have times between falling edges of 4,4,4,4 even across the irregular tooth. But if you look at times for when the level is high you get 1,1,2,1,1 which is perfect for a single missing teeth. If you look at falling edges you get 4,4,3,5,4 which may half work if you set the tolerances right, but it will not give you precision timing across this region.

It's possible, knowing how it works, that you may be able to grind a tooth and maybe end up with 2 missing or something like that.

[formula_sae]

If you don't have a second vr input, microsquirt won't know which crankwheel revolution is compression and which is intake/exhaust overlap. I would recommend setting skip teeth to 18 and fire the coil every revolution. If that messes the tach, call it a 2 cylinder.

thanks. makes sense now.

I hate to tell you this but wheel decoder is designed for edge detection - rising or falling, but not both. Your wheel looks like it is designed for level detection. In wheel decoder, a tooth is the time between edge detections. In your case, taking the lengths as times, you have times between falling edges of 4,4,4,4 even across the irregular tooth. But if you look at times for when the level is high you get 1,1,2,1,1 which is perfect for a single missing teeth. If you look at falling edges you get 4,4,3,5,4 which may half work if you set the tolerances right, but it will not give you precision timing across this region.

It's possible, knowing how it works, that you may be able to grind a tooth and maybe end up with 2 missing or something like that.

honestly, i understand what you mean about the rising and falling edges, but i'm not really sure what you're meaning by the times between rising and falling edges. we have removed the irregular tooth, and now have a 18 evenly-spaced teeth, one of which doesn't actually exist any more (for TDC indexing). can i set it up to look see high level instead of rising or falling edges? why wouldn't it see the blank area where the missing tooth is as a blank space? sorry, this is a bit more in-depth into the very basics of engine management than anything i've done before.

[grippo]

I based my statement on the diagram of the wheel that you had shown in an earlier post. If you have set up the wheel so that the distance (or time at steady rpm) between falling edges forms a sequence like t,t,t,2t,t,t,t.... where 2t is the missing tooth, then that will work. I didn't realize you had modified the wheel. I will look at your msq if I get time during the day, otherwise tonight.

[grippo]

You can have 1 input in a crank mounted, missing tooth, wasted spark configuration where you fire every 18 teeth. But if you are not seeing any spark at all, yet you have an rpm, then your immediate problem is probably electrical. There are all sorts of wiring diagrams in the megamanual for different types of coils and setups. If you have gone by these and are sure of the wiring, then you need to probe the output with a scope from the coil back through the driver circuit to the point where that goes to the processor. If that last point shows no activity, then it is a configuration problem. If there is activity there, but nothing out of the driver circuit, then you may have a blown fet driver. If in fact there is output from the circuit, then you should have spark, assuming the coil works. The timing may be way off, but that is a configuration problem which can be tackled later. Grab a double-e, promise him lots of beer and girls, show him the circuit diagram and put him to work.

What I am saying is that if you have an rpm that is reasonably steady - doesn't keep dropping to 0 constantly, then I am almost certain you will have a spark output from the processor (assuming the port on the processor isn't blown out).

[grippo]

I looked at the msq and the only thing I saw was that the skip teeth was 0, but I know that was changed. For this wasted spark setup it should be 18 teeth.

[formula_sae]

Grab a double-e, promise him lots of beer and girls, show him the circuit diagram and put him to work.

haha, funny you should mention it. we have one. he's in his 9th year, but graduating this year i'm pretty sure. he doesn't show up much because he's got a very abrasive personality, but at least we know where to find a good one.

by the way everyone, the spark problem is soon to be remedied. a really nice student at NorthWestern, named Grant Mahler, clued me in to how to get this engine working with this ECU. he's spent the last 16 months trying to get the two to cooperate. he told me how to get the factory Suzuki coil to fire, and that even once you get it to fire, spark tends to break up above 4000rpm. so we will do as he suggested and run an E34 BMW 5-series coil, as that's what that team runs and we happen to have a team member who works for BMW. said coil is an "always high" coil, and will be wired as such. Mr. Mahler says it works for them, so that's what we'll shoot for. after we get it in and get it wired, we'll find the exact ignition offset with a timing light. hopefully, after that, we'll be all set and ready to tune.

thanks again, everybody, and we'll keep you posted, for the sake of those interested and for posterity's sake.

[Maxzillian]

Judging by your date of posting, you are getting ready to attend the VIR event, correct? If you guys are still having problems, find Pittsburg State University (team 47) and ask for Eric Herbers. I should be able to lend you a hand assuming we get through tech inspection quickly.