Fireblade 893cc Engine - Idle and Start help

[Murdoc]

Hi,

I have a locost kit car I bought last year, I decided to rewire it from scratch and convert it over to injection so I'm using later 954cc throttle bodies. The rest of the setup is an early Zeitronix ZT-2, MX-5 MK1 fuel pump, MSII. I'm using an MX-5 igniter to fire the coils.



The car runs but I'm having trouble getting a consistent idle, at the moment the idle is controlled by a wax valve that seems to be a very good air trap so I have ordered a bosch 2 wire IAC. The wax valve and idle screw actuate bypass valves in the throttle bodies rather than acting on the butterflies. There are small M5 thread ports on the intake in the engine head, I have put 4mm push fittings in with some 4mm tube, this is my intended method to connect the IAC. I have just the pipe connected to the ports at the moment but this doesn't seem to provided sufficient air when in conjunction with the idle screw valves. In the original application the valves would have provided enough air for a larger capacity engine to idle. Therefore I think air isn't the problem, sometimes if the engine is idling OK the revs will increase from ~1k to ~2k for not apparent reason, could this be timing?

I has mostly used Alpha-N but today tried Speed Density, although I went back to Alpha-N for the ignition.
Starting has sometimes been difficult without a lot of throttle on, although has improved since I change the cranking advance to 0°.

I would be grateful if someone could have a look at my msq and logs to see if there is anything I'm doing wrong or any tips.

Thank you!

2017-03-08_17.16.48.msl

2017-03-08_16.56.21.zip

Murdoc_2017-03-08_20.53.32.msq

[Matt Cramer]

First, you'll have an easier time tuning if you can isolate issues - try to get it warmed up with whatever combination of throttle and fueling gets it to limp along until the engine is hot, then dial in the idle fuel and ignition at that point. After it idles well when hot, you can go back and get the cold start and warmup settings dialed in.

Also, if you set the peak torque and peak horsepower numbers too closely together in the VE table generator, you tend to get garbage numbers at low RPM. This seems to have happened to your VE table.

[Murdoc]

Thanks, that sounds like a very sensible idea.
I will have a look at the VE table. EDIT: the values at the bottom of the VE table have been tuned down by autotune (for low TPS, the numbers above are from my fiddling), I have had the engine warm once or twice and it seems OK when warm but the idle can run away. I will get it warm and generate a datalog.

I'm not entirely convinced by the ignition system I have put together so I have ordered some CBR1000 COPs and some Fairchild Ign specific IGBTs to replace the MX-5 igniter and old coil combo.

[grom_e30]

i think you may want to read up on the itb mode for load, like my bike it idles at about 80kpa so speed density may not be the best option. your ignition table is set for alpha-n but you table only goes down to 10% so between 0-10% throttle you can make any changes.

[Murdoc]

I managed to get the car hot, changed the VE table, and switched back to Alpha-N but it wasn't happy to run unless it was at 2k-3k rpm. Once warm I put an oscilloscope on the VR circuit and the cyl 1 coil low side, noticed the duration was quite long. After a stall I changed the spark dwell to 3ms, I had it set around 5.5-6ms and it seemed to make quite a difference. Not sure if the igniter was overheating. Following this I have also fixed a poor crimp on a relay which may have caused some issues.

I was looking at the ITB mode but thought it would be better to get it running at idle first. I will have a read up. I will look at making the tables go to 0.
Log, covers some of the stumbling around.

2017-03-09_18.09.05.zip

MSQ

Murdoc_09_03_17_Alpha-N.msq

Thank you for the help!

[grom_e30]

there is quite a bit of sync loss in the latest log you may want yo look at that

[Murdoc]

I was checking the sync losses. I think they are all at stalls or failures to start. The bad crimp I found was on the fuel/ign/inj relay, the crimp had cut through the wire that may have been the cause of some of the stalls. The stalls were happening very suddenly which is why I got the 'scope out.

[grom_e30]

they were when the engine was revving up