DIY Fuel injection thread.

[Roman]

If you have a turbo, when you are higher alitude each spin of the turbo is only able to inhale less air, so you need to work the turbine harder to reach the same boost level.

So this can impact engine efficiency in a way that a map sensor doesnt compensate for "automatically" as your turbo can go out of its efficiency range.

To a lesser extent with an NA engine at higher altitude the engine has a harder time breathing in and an easier time breathing out so the efficiency changes a bit at same map pressure.

However to put this in context, Taupo race track which is quite high is still 95kpa atmospheric pressure so its not like you're driving at 40kpa or something like that.

Pikes peak gets to about 70kpa at the top, from memory.

Not having baro compensation isnt the end of the world especially for normal driving conditions / especially if you've got closed loop wideband running.

Buuttttt if wanting to experiment with it then.... post results and graphs

[Alfashark]

18 hours ago, ajg193 said:

Apparently Lycoming engineers reckon that a barometric correction table is not really needed as the fueling should be within 5% of calculated for all realistically encounterable conditions.

Just remember that an aircraft engine is a million miles away from the conditions an automotive engine is under - nothing Lycoming or Continental make runs faster than 2750rpm and even then only for very short periods.  No changes in effective load on the engine, and a minimum of 3 seconds from idle to full throttle.  Not particularly hard for them to map their fuelling requirements to a close tolerance compared to the constantly changing conditions you'll use on the road.

[NickJ]

Printed out the profiles for the manifold full size to check I had spacings correct only to find I had the itbs upside down! Few clicks of the mouse later and to good surprise the bolt spacing came out much better, I have a few more gentle tweaks before sending it to the water jet but the design is pretty solid, I'm hoping to have them done by christmas, but she's sneaking up fast!

[kws]

I’ve just picked up a new car with an old Link LEM G1 and hand controller. Who thought it was a good idea to tune cars with a calculator?

looks like another speeduino or mega squirt will be on the cards. 

[Roman]

Alternatively you could consider upgrading to a Casio FX82

[Roman]

I live quite close to my neighbours, and my cars idle is very noisy - Especially when the car is cold.
So for sake of not seeming like a jerk I've been wanting to make my car a bit quieter when coming in and out of the driveway.

Based on some other time I've been looking to optimise ignition timing, I've found that when you dont have enough timing, the car is noisier - As more energy is going out the exhaust instead of pushing the piston down. Sooooo I tried advancing the timing at idle to about 20 degrees, (which makes rpm go up) and then pulling fuel out to 16:1 and closing throttle down to bring idle speed back down.
Which gave a really nice stable idle at 700rpm which is awesome... And its really quiet too! Awesome.

But now the next problem, which is that the car stalls a lot when idling this low.
Like if you rev the engine up, and revs fall back down, its fine. But if you are driving along, clutch in, and coming to a stop the car will stall. 
Also when you try to first take off, the car will lurch and die sometimes. 

My first instinct here is to just raise the idle speed again, but I want best of both worlds damnit! 

So I found that I can have a 3D table for idle speed goal and throttle angle. So it says "If car is moving, idle higher"



Now no stalling, and super stable low idle. Great success! 

[Ash Kelly barr]

MS1, v3 board, wasted spark, 36-1 crank.

Only used for ign.

 

Ok so I've been told to use the VR circuit for my cherry Hall effect sensor?

[Ash Kelly barr]

Can't find anything for the ms1, only pages for ms3 on adjusting pots

[ajg193]

22 hours ago, Roman said:

I live quite close to my neighbours, and my cars idle is very noisy - Especially when the car is cold.
So for sake of not seeming like a jerk I've been wanting to make my car a bit quieter when coming in and out of the driveway.

Based on some other time I've been looking to optimise ignition timing, I've found that when you dont have enough timing, the car is noisier - As more energy is going out the exhaust instead of pushing the piston down. Sooooo I tried advancing the timing at idle to about 20 degrees, (which makes rpm go up) and then pulling fuel out to 16:1 and closing throttle down to bring idle speed back down.
Which gave a really nice stable idle at 700rpm which is awesome... And its really quiet too! Awesome.

But now the next problem, which is that the car stalls a lot when idling this low.
Like if you rev the engine up, and revs fall back down, its fine. But if you are driving along, clutch in, and coming to a stop the car will stall. 
Also when you try to first take off, the car will lurch and die sometimes. 

My first instinct here is to just raise the idle speed again, but I want best of both worlds damnit! 

So I found that I can have a 3D table for idle speed goal and throttle angle. So it says "If car is moving, idle higher"



Now no stalling, and super stable low idle. Great success! 

For my car to fix that issue I just followed the MAP/RPM trace as it went back to idle and looked for where it went too lean or too rich and adjusted the cells on that path. Seemed to work quite well. My biggest problem that I am currently trying to solve is optimal MAT/CLT blending as my MAT sensor is right up in the airbox away from the engine. I seem to have gotten most of it right now.

[Roman]

My wideband responds super slowly at idle conditions, I think it's because it's quite far from the head and gas volume is so low in the pipe.
Perhaps also because the tip of the sensor is just flush with the edge of the pipe rather than in the flow. (Which is how its supposed to be, but maybe not ideal for idle readings)

So I dont really get anything reliable on heavily transient conditions. But you can also see affects of wall wetting stablising directly after transient conditions as well.



 

[Roman]

38 minutes ago, ajg193 said:

My biggest problem that I am currently trying to solve is optimal MAT/CLT blending as my MAT sensor is right up in the airbox away from the engine. I seem to have gotten most of it right now.

Okay so are you just able to adjust this as a fixed number, or can you adjust based on rpm and load? 

One problem here is that you've only got two points of reference but there's a 3rd variable in the middle which is intake manifold temperature. 

Like you could have 20 deg IAT and 90deg ECT, and intake manifold is 25 degrees.

So at low load/rpm you see a temperature gain of say 5 degrees or 10 degrees more than IAT.

But then you could also have 20 deg IAT, 90 deg ECT and intake manifold is 45 degrees because you've been sitting in traffic / engine running a long time / etc. 

In this case even though your two data points are identical the temp increase could be 30 degrees (or whatever)

So it's easy to chase your tail around. Although I guess it's fortunate that the areas mostly affected (low load/rpm) are very easy to trim effectively with wideband feedback in realtime.

[Yowzer]

Have you played with VVT angles at idle yet?

[Roman]

10 minutes ago, Firetruck said:

Have you played with VVT angles at idle yet?

Yep adding overlap at idle means it ruts like a car with big cams and you need lots of air and fuel to make it idle, and a higher idle speed. (1000+)

It sounds cool but it's annoying to drive with and very noisy. So entirely contrary to my goals haha.

Adjusting the injection timing was one of the best changes I made for reducing amount of fuel used at idle and improving idle stability.

[ajg193]

Mat coolant blend is a function of map*rpm.  At the moment I have it taking mainly coolant temp until above 60 kPa at 2500 rpm. Above that it is reasonably linear down to mat. These values seem somewhat reasonable as it is a small displacement engine with a fairly large volume intake system (long residence time) on the exhaust side of of the engine and the runners get hot during operation. 

[Roman]

This is a rerun of an old test that I did.
Going full throttle with different cam angles, to see which works best.
But this time logged at 100hz and going from full range of intake cam advance from 0 degrees through to 55 and logging results with MAF sensor.

The scale on the graph here is "Air per cylinder measured" so its effectively MAF reading divided by RPM. So its somewhat representative of the torque curve of the engine. 

It's pretty easy to see the results, which ever line is the highest up gives best airflow at that rpm. 

The colour scale down the side is degrees of cam advance. 



So a few slight changes to my VVTI settings it looks like there are some gains using about ~10 degrees advance up top rather than pulling back down to 4-5. And it wants as much advance as it can get at ~2000rpm. 

It's amazing to see what a compromise static cam timing is by comparison, even when only considering full throttle conditions.
And using a MAF for this gives such amazingly clear resolution compared to trying to look for changes in oxy sensor readings. 

[Ned]

for the fuse box looking people, these are cool... can put in fuses and relays etc wherever you want!

https://www.kinnettickreations.com/product-page/48-position-pdc 

[Yowzer]

That airflow is pretty neat, looks like the timing can compensate for reversion a bit there too. Now experiment with different trumpet lengths again.

[NickJ]

Whats the go to hall effect sensor for cam/crank angle? i've only found threaded rod style industrial ones or vehicle specific options that make mounting tricky, i'm leaning to the industrial style for ease but thats fairly chunky, something small and discrete would be nice, ideas anyone?

[Roman]

Any reason heart set on a hall effect rather than a reluctor? 

Might give you some more options.

[NickJ]

21 hours ago, Roman said:

Any reason heart set on a hall effect rather than a reluctor? 

Might give you some more options.

Only reason is perceived ease, reluctor needs a few more bits to run with the speeduino, but i'm happy to be convinced otherwise.