DIY Fuel injection thread.

[mjrstar]

Lots of factory cars in the 90's before the newer generation of EFI had some simple little controllers/power supplies to do this. Also PWM controllers to make the pump quiet until you plant it.

 

My evo had a factory dual voltage setup, when i put a larger fuel pressure reg on at the low voltage setting (idle) i couldn't get more than about 25psi out of it. have since just jumpered the high voltage relay and get all of the volts all of the time, pressure is now a happy base 43 psi.

I wonder if some of this is to reduce pump noise also? 

[Roman]

I think it's mainly to extend the pumps life and also yeah to reduce cabin noise from the pump.

[yoeddynz]

So how many pieces of toast, cups of tea/coffee and cake will it cost to have you stay here a weekend and check my dead times.

I can show you the chouse photography locations too.

[yoeddynz]

I think it's mainly to extend the pumps life and also yeah to reduce cabin noise from the pump.

Although noisy pumps are a way of pretending "hey bro..I've got four 044 pumps in line to fed my 1200bhp monsta"

Or not.

[Roman]

So how many pieces of toast, cups of tea/coffee and cake will it cost to have you stay here a weekend and check my dead times.

I can show you the chouse photography locations too.

Oh man I would be super keen for a holiday to Nelson!

<3 South Island.

 

Do you have a spare injector, the same as what's in your car? 

 

If so, send it up! and I'll test it for you.

[mjrstar]

My falcon runs an 044 with an under bonnet surge thank and it's stupid loud, there must be a bunch of quieter options out there...

[yoeddynz]

Ok cool. I have two or three spares sets of rails. When I tested them some were spraying cleaner patterns than others and luckily the ones I had been using in the car were by far the best if the lot. Possibly just because they had been in use and kept free and clean.

I can post up a rail with the appropriate plug to make it very easy for you.

I can't even remember what dead time I set if at all. Possibly just a default value that was on ms extra code loaded on (which I think are all based around a rover v8 or some thing because James who wrote the extra code is a rover nut)

Or I might have nabbed the relevant values from the mx6 forum. Can't remember but will check...

[Roman]

I finally got my new wideband installed over the weekend.
So switched over to modelled fuel. (This models the engine size, injector size, air mass, AFR goal etc etc as part of the fuel equation rather than just numbers on a table)

 

First thing I notice is how stupidly fast I went from nothing to having a mostly sorted fuel map! 

 

I believe the numbers in the fuel table represent the engines volumetric efficiency at each rpm/load combination.

 

Which makes it easy to make a base map because for an NA engine you set the WOT row to 100% efficiency and then just drop the numbers down the rest of the map and then interpolate it. Doing this, I had a drivable car that didnt buck or run crazy rich/lean even before I'd driven it.
Some spots were quite different (Idle area 70% efficiency at high load, and the peak torque rpm area was about 115%)

One thing that's cool that I noticed, is that there is a table with your goal AFR is part of the initial fuel equation.  
So once your numbers are correct, it's super easy to adjust AFR because you just raise or lower the AFR goal.

Rather than, having to adjust an arbitrary number in a table up or down in every cell via 02 feedback, and checking what AFR that results in.

Also the engine seems to run smoother for some reason, possibly that this is more accurate lambda readings from the sensor now?
Maybe I was previously actually running leaner or something like that. Not sure but it feels really smooth and sounds good which is cool.

Running only open loop and tuning just with the Mixture Map I very quickly managed to get a most of the low load cells to a stable 14.7 in a very short period of time.

The fuel economy information is dissapointing though, it just gives cc per minute rather than l/100 or k/litre or MPG or whatever.
But this still makes life a lot easier.

I remember the first time I tried running on modelled fuel setup, it was a total disaster because I had so many of the background settings wrong. 

But its one of those things, when you've put in the groundwork ahead of time the rest just falls into place.

Looking forward to some more tinkering this week!

[h4nd]

Nice work Roman.

 

Re Yoeddy and Ned fuel pumps: Do you guys run your fuel rails like this?

 

[Tank]---+---[pump]---[injector Rail]---[Pressure Reg]---+

         |                                               |

         +-----------------------------------------------+

 

edit: (courier new isn't fixed width?)

re-edit: Huh??

[fuel]

you would want the fuel return to come back to the tank preferably away from the fuel pump pickup to give the returned fuel enough time to mix with the rest of the fuel and dissipate some heat. Many people have found if the fuel return comes back into a surge tank or really close to the fuel pump pickup that the fuel gets super-heated.  

[Ned]

tank -> filter -> pump -> filter -> rail -> reg -> return to tank

[Roman]

I've been having a bit of a think about compensation tables lately and how to build them in a meaningful way.

I'm not convinced that the default IAT compensation values necessarily suit my engine.
But It seems quite dififcult to know how to change it and check the results.

But I came up with a good plan. (maybe)

Instead of mucking around with the IAT compensation table and randomly changing values.
For now I've just turned it off, and all other comp tables.

So I'll get the car as heat soaked and high temp as I possibly can, so that the temps of everything are already stable on the "high" side.
Then run through my map and get fuel values spot on as best I can.

Thennnn let the car cool down completely, start it up and start a datalog once the engine temp reaches say 85deg.

The drive around for ages and build up a big datalog file of how the fuel changes over time on account of heatsoak etc.
 

I can add a second fuel table that's an overlay of the main one, and that adds or subtracts from the main table.

 

Buuuuutttttt the interesting part is that it doesnt need to share the same axes on the second table as the main one.

 

So for example my main fuel table works based on rpm and MAP sensor.

 

However on my secondary table I could have it working based on RPM and engine run time. or IAT and engine run time. 

 

Or RPM and intake manifold temp sensor readings (if I had one) Or whatever else.

 

Instead of needing to figure out how to build this map. (and whether it makes any sense)
I can load the datalog file and use the mixture map table to build the overlay map to fill the 2nd table in.

If there's a genuine trend happening, you'll see it in the shape of that overlay map. 

If it's just random spikes up and down, then I'll try something different for the axes and rinse repeat until I find what it is that causes the trend, without having to run the car at all until I've convinced myself that I've made a worthwhile comp table.

 

Then if this works, I can run the car with that comp table in place and do same thing again to build up the next comp table on top of that (if need be)

[yoeddynz]

you would want the fuel return to come back to the tank preferably away from the fuel pump pickup to give the returned fuel enough time to mix with the rest of the fuel and dissipate some heat. Many people have found if the fuel return comes back into a surge tank or really close to the fuel pump pickup that the fuel gets super-heated.  

This ^ = me.

 

I must sort it out. Send return back to main tank.

[yoeddynz]

I've been having a bit of a think about compensation tables lately and how to build them in a meaningful way.

I'm not convinced that the default IAT compensation values necessarily suit my engine.

But It seems quite dififcult to know how to change it and check the results.

But I came up with a good plan. (maybe)

Instead of mucking around with the IAT compensation table and randomly changing values.

For now I've just turned it off, and all other comp tables.

So I'll get the car as heat soaked and high temp as I possibly can, so that the temps of everything are already stable on the "high" side.

Then run through my map and get fuel values spot on as best I can.

Thennnn let the car cool down completely, start it up and start a datalog once the engine temp reaches say 85deg.

The drive around for ages and build up a big datalog file of how the fuel changes over time on account of heatsoak etc.

 

I can add a second fuel table that's an overlay of the main one, and that adds or subtracts from the main table.

 

Buuuuutttttt the interesting part is that it doesnt need to share the same axes on the second table as the main one.

 

So for example my main fuel table works based on rpm and MAP sensor.

 

However on my secondary table I could have it working based on RPM and engine run time. or IAT and engine run time. 

 

Or RPM and intake manifold temp sensor readings (if I had one) Or whatever else.

 

Instead of needing to figure out how to build this map. (and whether it makes any sense)

I can load the datalog file and use the mixture map table to build the overlay map to fill the 2nd table in.

If there's a genuine trend happening, you'll see it in the shape of that overlay map. 

If it's just random spikes up and down, then I'll try something different for the axes and rinse repeat until I find what it is that causes the trend, without having to run the car at all until I've convinced myself that I've made a worthwhile comp table.

 

Then if this works, I can run the car with that comp table in place and do same thing again to build up the next comp table on top of that (if need be)

....aaaaaannnnnd when you have finished your car you can do mine I'll supply beer.

[Roman]

It was coming up to dinner time last night and I was mucking around on the laptop

 

Girlfriend: "Hey, how much longer are you going to be drawing graphs in that program for?"

 

Me: "About 20 years"

 

Girlfriend: "Okay"



 

I chopped the sub 1500rpm section off my main fuel map and setup a high res table between 700-1300rpm and have been refining it with datalogs/mixture map. Idle is way more steady now, still a work in progress though

 

[kpr]

2weeks of my life and about 1k later.  looks like i've gained a couple of kw.   na lyfe

 

 

pink old.    more inj d/c at same afr + rpm starts to pull away.

[Celica RA45]

why do you guys run 044 bosch motor sport pumps ,that pushes 300 litres through your fuel rail per hr ,wouldnt it be better to use a 910 bosch motor sport pump that pushes 130 litres instead , 

on my race car i had to many problems with my fuel overheating from 75 lbs fuel pressure and pushing 300 litres per hr ,went back to the 130 litres per hr and has been fine ever since 

[cute wee gem]

What did you change, Kris?

[kpr]

Bit bigger cams.   and some more diy porting.

 

Long version: It didn't really respond too well to the cams.  I started to pull it apart so could dial wheel the cams, as i was initially too lazy to do it and the last set of kelfords were spot on.   Got a bit carried away and decided to do some more work to the ports, as head seems to be the restriction.  

When i put it back back together, dialed the cams.  Exhaust cam was pretty close.   but kelfords had ground the Inlet cam  20degrees retarded..  which happens to be 1 tooth out from factory marks. 

 

So before i pulled it apart i had advanced the  inlet a lot to try get it to make some power ..  20degrees,  so was running at there recommended settings,  while i was thinking it was way advanced..

 

After the porting.  set the cam timing back to where it was.  didn't really gain much. but didn't go backwards.  advancing the inlet 10deg after this gained a decent amount through midrange

 

Log above is new cams (193c) + some more porting  vs  old cams (193b).    both old and new cams  dialed at  10degree retard exhaust and 10degree advance inlet. over there recommended settings.

 

 

really needs a big valve head and some good port work to go much further.

[Roman]

really needs a big valve head and some good port work to go much further.

Beams motors have bigger valves