DIY Fuel injection thread.

[Roman]

Yeah to be honest I might end up going MAP anyway in the long run because I want to take some restriction out of the intake.

As currently between the air filter and the throttle body it's 1/4 inch smaller pipe than the throttle body inlet pipe size. (2.75" vs 3")

And the MAP sensor shows a pressure drop above 6k rpm, probably because of this.

But it's been interesting seeing the difference with this, and getting a better understanding of both.

 

The main things that suck about MAF are that an air leak ANYWHERE ruins your day, and if you change one single thing in your intake it throws its toys out of the cot.

But if a 3" or 3.25" pipe with MAF bodged in works well enough after being calibrated I might stick with that because my brain can make sense of the tables a little easier.

 

Turns out my tables can have 20 rows of resolution so it's not really an issue running out either way.

Good fun though! Cant wait for next trackday.

[eke_zetec_RWD]

Gone of the days of my first link that had 5 rows I think. Still managed to tune quads with a supercharger blowing on them via tps only haha.

[kpr]

glory days of hand controllers and low resolution maps.  then there was a haltech of some sort, that went completely the other way. with some  massive non user definable maps, that sucked even more to tune.

 

starting off with low amount of rows, then adding some if needed is the way to go.   unless megasquirt you have to guess from the start. since cant just jam one in the middle afterwards

[eke_zetec_RWD]

glory days of megasquirt you have to guess from the start. since cant just jam one in the middle afterwards

Ohh really! Fuck that

[downtrail]

Shouldnt be looking at main fuel map whilst cranking anyway so adjust your cranking pulsewidths in order it should go cranking table then once its passed the cranking threshold then main map + after start for x amount of time / cycles.

[kpr]

On a link, its still based on the fuel map.  There is a crank enrichment table, which adds fuel to number in fuel map based on engine temp.

 

Ohh really! Fuck that

 

Yeah, i even suggested to to them for a new feature  when i had ms3. They basically said no cant do it.   From memory to add a new row/column, you had to tack it on the end, then re adjust the existing rows to get your new row where you wanted it. Which obviously messes up your tune.    Can adjust the existing rows, but can't just put  a new row in anywhere and interpolate it like link.

[downtrail]

sorry yes thats the way link does it was thinking megasquirt for some reason. So just play with your 500rpm column untill you get what you want

[Rookie]

On a link, its still based on the fuel map.  There is a crank enrichment table, which adds fuel to number in fuel map based on engine temp.

 

 

Yeah, i even suggested to to them for a new feature  when i had ms3. They basically said no cant do it.   From memory to add a new row/column, you had to tack it on the end, then re adjust the existing rows to get your new row where you wanted it. Which obviously messes up your tune.    Can adjust the existing rows, but can't just put  a new row in anywhere and interpolate it like link.

The easiest way to do it is to add a row on the end and then copy the table over one, then change the header values, hardly a big deal.

[kpr]

[Burntrubber]

im getting an inlet backfire at most startups, what would cause this?

I was getting alsorts of backfires on startup with my megasquirt on my Gemini. Turned out been voltage drop while cranking caused by old starter Motor got it fully recoed new windings etc. I had tried everything to stop the misfiring with no luck and it ended up been something that seemed fine.

[Roman]

Hey guys I'm trying to think of the best way to tune for cruising / fuel economy.

 

I was thinking same stretch of flat straight road, do a run aiming for a constant 80kph on GPS. Datalog all the things.

Adjust a variable (ignition timing, VVTI, or fuel)

 

But then what would be the best measure of economy?

A lower average throttle angle perhaps? More bang out of less air.

 

The factory ECU seemed to want to run right to the limit of detonation, so perhaps slamming in timing right up to that point is best for this motor.
 

Some people say running leaner is a better idea, but then others say you end up needing more throttle angle to keep the same speed if you do so.

EDIT:

 

Aahh! This article is pretty awesome on the topic.

 

http://www.superstreetonline.com/how-to/engine/0510ht-fuel-economy-tuning/

[Rookie]

Surely the best measure of economy would be injector pulse width? IE exactly how much fuel you are squirting into the engine.

[eke_zetec_RWD]

Surely the best measure of economy would be injector pulse width? IE exactly how much fuel you are squirting into the engine.

Assuming fuel pressure is constant?

Could log pressure and pulse width together

[Roman]

Yeah fuel pressure will be constant, my FPR doesnt reference manifold pressure.

 

Yes initially I was thinking that logging injector pulse width makes sense.
 

But then for some reason that now slips my mind it seemed like it wouldnt be accurate based on info that I had.

 

However, I've just noticed if I change my fuel calculations from "traditional" to "modelled" it allows you to enter engine size, injector size, fuel pressure etc...

Once you've filled in these things it gives you some better information about fuel consumption, estimated air/fuel consumed, etc etc.

Hoorah!

 

Either way, as per the Honda article it looks like advancing the cam 15-20 degrees and advancing timing to 35-40 degrees and 15:1 or there abouts will be the best starting point.

There's not much point trying to tune any further at the moment though until my new IAT sensor turns up and I get my MAP sensor calibrated properly.

 

Because I tune it up nice one night, then the next day its all over the show again.
Needs some more accurate info to get the compensation tables working correctly I think.

[Rookie]

I tuned MOTT's crusier to 16:1, fuel economy improved dramatically to less than 20L/100km.

[h4nd]

You efi guys, I'm curious, does fuel temp affect your mix much?

[Rookie]

Yes and no. Yes it does, but no it isn't a practical worry unless you have massive fuel pumps, a tiny tank and very little load.

[Roman]

Aahh so I found that if I changed the fuel modelling from 'traditional' (which doesnt care about injector size etc) to 'modelled' then it works out fuel based on engine size, injector size, fuel pressure, blah blah.

 

Which is what I need to do in order to access the Litres per Hour figures and other similar measures of fuel consumption. (Even though Effective pulse width is same same I guess)

 

However it's a kick in the balls because all of the fuel values on my existing map need to drop down by like 95% for it to work.

 

So I think I'll put that one on hold until I get my proper IAT in and configured, and map sensor calibrated properly.

I'm thinking that once I've got gear detection setup, I can just filter the log files to only show values for 6th gear, TPS at say 10-20% and then have an Engine RPM vs Litres per Hour graph to compare between runs.

Irony noted of using lots of gas to make car use less gas.

[Testament]

google maps sensor calibration you mean

[yoeddynz]

This is a bonnie wee thread. I like where its going.

 

I would love some kind of economy meter - ie one that measures the amount of fuel passing by a sensor. Problem is the system is a constant return setup. So what rookie just said about injector pulse width makes so much sense if other things are constant. Could one setup some sort of bmw like economy meter that works off the pulse width. ? How do cars that have economy meters get their figures?