DIY Fuel injection thread.

[Roman]

Havent tried one, but I follow their page on FB. 

From what I've seen it looks like it's the most actively developed out of the Tuner Studio compatible DIY style options.

But price point will make it or break it.
 

[M.H.]

Has any one tried running a modernish oem ecu on a old engine? I was thinking of using the bosch me7.5.5 from a vw 1.8t on the fiat 1.6 engine for my X1/9. I already have everything so just thought it would be a cheap way of running sequential injection and spark, wideband, drive by wire ect. It can be directly tuned over the obd port

[Yowzer]

I see no reason why you couldn't, people have repurposed the Honda ECUs for other jobs with great success so it'd be very much the same.

[JustHarry]

A workmate has a supercharged fist 128 running toyota 4agze ecu wiring and sensors.

Makes sense if you can find a good factory system on an engine of close size and power.

 

[shrike]

17 minutes ago, 00quattro00 said:

Has any one tried running a modernish oem ecu on a old engine? I was thinking of using the bosch me7.5.5 from a vw 1.8t on the fiat 1.6 engine for my X1/9. I already have everything so just thought it would be a cheap way of running sequential injection and spark, wideband, drive by wire ect. It can be directly tuned over the obd port

People run Barra Ecus on the sohc AU engines as well

[Yowzer]

Seems like a waste, they should be swapping the whole engine.

[kpr]

Few days late on this one,   but going back to the overlap thing, with air going straight through the engine.     I thought this may have been the case  when i was road tuning my engine.  300deg ish cams,   running approx  5mm of lift on the inlet cam and 3mm lift on exhaust at TDC.  So you can basically look straight through the engine at TDC.   
I was tuning with a peasants version of @Roman's  MAF tuning;    A tps (alpha n) tune.   So the throttle plate is metering the air. The ecu doesn't have any compensation like a map tune.  Which will add fuel as manifold pressure increases.       well other than barro correction .     So  at the same throttle opening if any more air goes into the engine it will result in a leaner afr.  Which in turn will need  X percent more fuel,  which = x percent more power. 
So getting to the point of adjusting cam timing.  going from a lowish overlap setup  then advancing the inlet  cam, which increases overlap.  AFR  would go lean through most of the midrange and little richer right up top. cool did i just gain power  or  loosing part of the charge out the exhaust?  same deal with exhaust cam.  till ending up with settings far more aggressive overlap wise than what the cam manufacture recommended.  

Then  confirmed  on the dyno that those cam settings were actually gaining power.   So seems the theory that any big amount of air going straight through the engine, isn't going to happen unless in an extreme case 

 

[ajg193]

 I finally decided to have a go at setting idle control up on the microsquirt. This feature was so low on my priority list that I just never really felt like playing with it.

I only have the factory oldschool auxiliary air valve on the engine:

 

No fancy steppers or PWM valve or electronic throttlebody to take advantage of.

I have the valve setup so it will give me about 1200 RPM when cold and settles at 800 when warm.

Anywho, the ECU has a function for idle control where you specify an idle RPM as a function of temperature and you can set it to just vary the ignition timing as a function of RPM delta from the desired value. I always had this option turned off and just let the engine idle wherever it wanted to based on my ignition and fuel maps, so it would idle happily at 800 RPM all day but would drop 50-100 RPM when headlights or fans were turned on. A couple of minutes of fiddling with the adaptive timing settings (basically just advance timing if below desired RPM and retard timing if above desired RPM) and the car now sits rock solid at 800 RPM no matter the load on the alternator.

It's just a small change but it makes the car feel at least 5 years newer, and didn't cost a cent in new valves or wiring stuff in. So I'd say it's a win.

Hopefully it works reasonably well at lower temperatures while the engine is warming up, but there are a fair few settings available to toy with. I'd say the range of RPMs obtainable with this setup will be quite small as it's always just getting the same amount of air into the engine (but still varying with the AAV closing as engine warms up), so I won't be able to demand 1200 RPM when hot unless I adjust the idle bypass on the throttlebody - but why would I want so much adjustment anyway?

 

 

[BlownCorona]

I did not know that feature existed! 

I was considering going to e throttle when I put the motec in to avoid havnt an iac valve, but if the megasquirt can do it with timing then hopefully the motec can and the future ugly electric throttle body can fuck off. 

Cool! 

[ajg193]

A couple of random numbers from my ECU that may be of interest to people considering EFI:

 

Idle fuel consumption

1. Fuel only with factory distributor: 14 cc/min (840 cc/hr)
2. Untimed injection with ECU controlled timing: 11 cc/min (660 cc/hr)
3. Timed semi-sequential injection with ECU controlled timing: 9 cc/min (540 cc/hr)

 

 

[Yowzer]

Oh @Roman I was pondering your exhaust tuning question re maf sensor, and it occurred that it may not be so simple.
I don't know the dynamics of exhaust pulse tuning well enough to know when the pulse is actually supposed to be timed for maximum results.

For example, if you have a low pressure pulse hitting the ex valve as it opens, you'd aid cylinder extraction and reduce pumping losses, but this wouldn't have any effect on the intake.

However you could also have this pulse during overlap and aid scavenging, which potentially would show on the maf

Or if you have a high pressure pulse hit the valves during scavenging to push over-excited intake charge back in the cylinder, much like a twostroke

In short, I have no idea and I'm just thinking out loud. Make some graphs and report back

[Yowzer]

Here's some science or something. Pretty basic article but meh.

https://www.onallcylinders.com/2015/10/30/header-theory-part-1-looking-at-the-science-behind-exhaust-header-tuning/

I did read a very in depth article ages ago regarding this on old F1 engines, but it was high level stuff and I wasn't all the sober at the time.

[Roman]

2 hours ago, Yowzer said:

Oh @Roman I was pondering your exhaust tuning question re maf sensor.
. Make some graphs and report back

The results are going to be interesting either way. 
I just need to get my friggen car going again then this is #1 on my list.

Personally I think all of the accumulative phenomena going on in the exhaust either help pull air out, stop air being pulled out, or some varying combination of these things. In which case, it'll show on the MAF. 
Empirical evidence is king, for me... The "how or why" doesnt matter so much as "does, or does not". 
If this theory gets proven wrong, thats really interesting in itself.
But I'll be chuffed if I've got a good method of quantifying exhaust side changes this way.

So much nerding to do coming up. Bung will be at max fizz when my car's going again

[BlownCorona]

HP Academy Tuning Fundamentals course for only $1!

https://www.hpacademy.com/1-dollar-efi-tuning-fundamentals?utm_source=promotions&utm_medium=email&utm_campaign=1-dollar-efi-tf&fbclid=IwAR3PpzbTmC7nwQMMKTVITUorFz8Vz2GSpCc55uUqaKIL1fusMSENWqQ4zs8

dunno how long this sale goes, but how can you go wrong. apparently this course is really good and lots to be learned. 

[Roman]

Dang, thats good! 

Definitely do it, when I worked there I needed to go through all of the courses so I'd be able to answer peoples questions. 
There were plenty of "wellll, shit! that makes sense! / Never thought of it that way!" moments even after messing around with EFI stuff for many years. 

Reccomend.

EDIT: I see it gives you 3 months of gold membership gives you access to all of their webinars too. Which have some awesome info in them.

[BlownCorona]

its a pretty bloody generous offer for a buck.

i flicked through the list of videos this morning and theres shit loads of stuff there. 

[Transom]

Just now, BlownCorona said:

its a pretty bloody generous offer for a buck.

i flicked through the list of videos this morning and theres shit loads of stuff there. 

Signed up for that thanks for link gotta be able to learn something from it 

[maxted]

Thats a good link just signed up too

 

[h4nd]

On 20/04/2020 at 20:46, ajg193 said:

A couple of random numbers from my ECU that may be of interest to people considering EFI:

Idle fuel consumption

1. Fuel only with factory distributor: 14 cc/min (840 cc/hr)
2. Untimed injection with ECU controlled timing: 11 cc/min (660 cc/hr)
3. Timed semi-sequential injection with ECU controlled timing: 9 cc/min (540 cc/hr)

Calc from pressure and duty cycle? I'd be curious to see that measured on actual, so as to get an idea of how injector calibrations line up (open / shut times). I recall Roman did some work on that?

[ajg193]

Flow rate from Tunerstudio, over a whole tank the predicted values are within a percent or so of actual fuel consumption.

However, I don't have absolute faith that my deadtime is bang on so the accuracy at low flow rates may be questionable.