[rusEfi] hello all - Mitsubishi #31

[matt]

update..
my engine start and running with Rusefi!!


big thanks to russian and kb1gtt.

[stefanst]

Awesome- congratulations!
Could you please post your tune for a quick critique?

[matt]

msq attached

[AndreyB]

And a video!!!

[matt]

video i upload at facebook. i have no account yet at youtube.

[matt]

[video][/video]

[AndreyB]

How to find it on facebook?

[matt]

i try to paste it here. dont know it is working or not.

[AndreyB]

Tag @rusefiecu on fb i'll try to download

[matt]

ok, i try

[kb1gtt]

Congratulations.

[AndreyB]

[matt]

one more thing. do i need to wired fuel pump this way?

[kb1gtt]

I believe your OEM harness uses a fuel pump relay. If you provide 12V to the fuel pump relay pin in the harness connector, you can have the ECU control the fuel pump. Frankenso has the hi/low drivers which can source current. It appears your relay needs an ECU which can source current, and as luck has it, Frankenso can do that.

You will need to double check the current consumed by the relay. The hi/lo drivers can only handle about 0.125A before we expect thermal concerns start to limit the current draw. So I would say find the pin for your fuel pump in the harness, then with a multi meter in 200mA or 400mA current mode (or what ever your meter has for the lowest current), then connect one side of the meter to +12V, and the other end to the harness pin in the ECU. This should cause the fuel pump relay to click, and you should be able to measure a current. If that current is at or below 0.125A, then you can configure a hi/lo driver chip to activate this relay. If it is more than 0.125A, then you will probably need to add a heat sink to that coil driver, or you'll need to limit the max ambient temperature of the ECU. For now measure the relay current and let us know what you find.

[AndreyB]

undocumented advanced diagnostics, only documentation is the source code

Starting to document these at http://rusefi.com/wiki/index.php?title=Manual:Debug_fields

[stefanst]

I just took a quick look at your tune. Looks like a good start. I see a few things that could be looked at again:

Topic Engine:
- Ignition Settings: Dwell table is 2.01ms across the board. This really depends on your coils, but 2ms seems really short. More common would be 4ms to 6ms or so. If you have missfires, you may want to try increasing the dwell in that range
- IAT sensor: Looks like you're not using one yet. In order to properly tune a Speed-Density (MAP-based) controller you will need IAT to correct for air-density changes
- Tuning fuel without a Wideband O2 sensor is going to be nearly impossible

Topic Tuning:
- Cranking: Set "Enable Cyl. Cleanup" to "true". Helps you to clear a flooded engine by going full-throttle during cranking
- Cranking: I like batch fire for cranking the engine. Seems a little less susceptible to doing stupid stuff
- Cranking dwell also seems short at 3ms. On my 99 Miata I'm using 8ms
- Baro Correction: Hole table should be set to "1" across the board to start with

Looks like you're making good progress. Don't even bother trying tuning fuel without a means of determining your AFR. It's not worth it....

Good luck!

[matt]

I just took a quick look at your tune. Looks like a good start. I see a few things that could be looked at again:

Topic Engine:
- Ignition Settings: Dwell table is 2.01ms across the board. This really depends on your coils, but 2ms seems really short. More common would be 4ms to 6ms or so. If you have missfires, you may want to try increasing the dwell in that range
- IAT sensor: Looks like you're not using one yet. In order to properly tune a Speed-Density (MAP-based) controller you will need IAT to correct for air-density changes
- Tuning fuel without a Wideband O2 sensor is going to be nearly impossible

Topic Tuning:
- Cranking: Set "Enable Cyl. Cleanup" to "true". Helps you to clear a flooded engine by going full-throttle during cranking
- Cranking: I like batch fire for cranking the engine. Seems a little less susceptible to doing stupid stuff
- Cranking dwell also seems short at 3ms. On my 99 Miata I'm using 8ms
- Baro Correction: Hole table should be set to "1" across the board to start with

Looks like you're making good progress. Don't even bother trying tuning fuel without a means of determining your AFR. It's not worth it....

Good luck!

IAT- i not wired it yet.
dwell- last few years ago, i got burn of my coil with ms2. i dont know it cause either by dwell is over 2ms or by spark inverted. to play safe, i only set it to 2ms and had tested it around 12hours driving with ms2. BTW i'am using toyota yaris/vios coil on plug.
cranking- meaning if i set batch, it will run batch spark and run sequential when it running?
baro- ok will try

i got my innovate MTX, but i don't wired it in rusefi yet.

anyway, thanks for suggestion.

[matt]

I believe your OEM harness uses a fuel pump relay. If you provide 12V to the fuel pump relay pin in the harness connector, you can have the ECU control the fuel pump. Frankenso has the hi/low drivers which can source current. It appears your relay needs an ECU which can source current, and as luck has it, Frankenso can do that.

You will need to double check the current consumed by the relay. The hi/lo drivers can only handle about 0.125A before we expect thermal concerns start to limit the current draw. So I would say find the pin for your fuel pump in the harness, then with a multi meter in 200mA or 400mA current mode (or what ever your meter has for the lowest current), then connect one side of the meter to +12V, and the other end to the harness pin in the ECU. This should cause the fuel pump relay to click, and you should be able to measure a current. If that current is at or below 0.125A, then you can configure a hi/lo driver chip to activate this relay. If it is more than 0.125A, then you will probably need to add a heat sink to that coil driver, or you'll need to limit the max ambient temperature of the ECU. For now measure the relay current and let us know what you find.

i am 100% sure my oem harness is like this.

fuel pump.png

ig switch supply 12v to main relay. once main relay activate, it feed 12v to pin85 at fuel pump relay. so we need a ground signal at pin36 to activate fuel pump relay. my issue is fuel pump is no switching. or do i need to add some resistor at green wire to lower the current?

BTW my multimeter mode is 10A. i am not able to measure.

[kb1gtt]

Sorry I miss read the schematic. I thought 85 and 36 were a normally closed contact that would open when the relay was activated. I did not recognize that as the coil to activate the relay. I understand it now. You can use an unused injector driver to drive this relay.

Does your 10A meter function down to 0.1 or 0.2. I was suggesting the smaller mA scale for increased accuracy as well they have an internal fuse which helps protect the wiring. Can you take a reading with the 10A scale and see what you get. However in this situation, connect that ECU pin with the meter to GND, not to the 12V. If your wiring is like that circuit, it is low side drive. So any time the key is on, and you connect that wire to GND, the fuel pump would be functional.

[matt]

issue solve. i found something is going wrong at inj 6 channel. i move fuel pump signal to other channel and it work great.
where should i wired for launch control input? it is via op-amp channel?

[AndreyB]

where should i wired for launch control input? it is via op-amp channel?

what do you mean launch control input? clutch pedal? some button?

not much launch control logic in rusEfi at the moment

[matt]

yes. clutch pedal type.

[AndreyB]

yes. clutch pedal type.

Not much logic in the firmware to use clutch pedal signal but yes, at the moment clutch input is implemented via ADC inputs I believe.

[matt]

done some on the road tuning just now. everything is running good.
hopefully this weekend i can dyno my car.

[AndreyB]

hopefully this weekend i can dyno my car.

looking forward a video!

[matt]

i saw in tunerstudio, trigger type nissan. it this trigger for 360 slot nissan cam angle sensor?

[AndreyB]

i saw in tunerstudio, trigger type nissan. it this trigger for 360 slot nissan cam angle sensor?

It was supposed to be that one but the guy with that Nissan is not very active so it was never finished. http://rusefi.com/wiki/index.php?title=Manual:Software:Trigger#Nissan

[matt]

i am planning to build frankenstein board, and my target engine is nissan sr20ve neo vvl.
for now this engine is running with ms2 with 4g63 trigger. i swap nissan cam angle sensor with 4g63 cas with little bit modification.

[matt]

done dyno tuning today with rusefi. got some issue at rpm 5500 above. seems it related with trigger/ignition.

[kb1gtt]

Cool to see another dyno run. You're moving it right along.

I see you used a DB style connector(s). These are not rated for the voltages of VR's or currents of injectors. There is a strong chance the problem is electrically based.

A VR can reach voltages of 300V at high RPM. I seem to recall DB connectors are only rated for like 50V, then after that you are getting significant leakage currents via creapage issues. These creapage issues at higher voltage could be causing issues with your crank signals. However they are typically more of an issue with leaking currents to other circuits, like the 5V rail. High voltages to the 5V rails can cause all sorts of problems.

The injectors at 1A will likely cause your connector to overheat, but the contact resistance can also cause a tank circuit. A tank circuit is known to allow your injectors to see 300V, but the ECU only see's 50V. However connector problems there typically result in electrical fire or injector failure. So that's not your problem yet.

Do you have TS logging, console captures, or scope captures to show if the problem was seen by the the Frankenso board? For example, if we see funky pulses on the crank, we can make some guesses about what is wrong. Also if we see the fuel pulses were as expected, we then can start to point at fuel rail pressure issues. Those logs can show allot about what happened.