DIY Fuel injection thread.

[Carsnz123]

@Rookie The good news I was pretty good at physics in highschool. If you have any books/web sites that would help me I'm all ears. Mega Manual has a lot of good info.     

[NickJ]

Been doing some more experimenting with the carbs on the Niva, after leaving the car to idle for 5min last weekend to help another vehicle, it instantly died when I tried to head off again, nothing immediately odd, but when I thought about it, i've been having trouble getting it to idle anything but lean, so why would that cause it to die? this weekend I hooked up the wide band and left it to idle, sure enough, wideband showed it as lean, after 5 min I gave the throttle a poke and sure enough it died with the wideband indicating a shift to off the scale rich! So far my explanation is that the long runners (due to making an equal length manifold that snakes around the distributor) is not keeping a high enough velocity that fuel is dropping out of suspension and pooling, this pooling is then made worse by the horizontal section of each runner! 

Right now i'm not sure how I will solve this issue, further looking at the set up I believe the lean spike at ~3000rpm is due to resonance in the runners stalling fuel flow in the carb as it happens at the same rev range independent of engine load, while trumpets could shift the band somewhere else, i'm short on room cos of the bonnet! In my digging i've found info from guys who have messed with emulsion tubes to solve similar ills, but ultimately $300 to re-cert for fuel injection is starting to sound very viable! 

[Roman]

Perhaps you have the problem where the longer runners mean you need more initial enrichment, as in, when you stab the throttle your engine wants to intake (for example) 400cc of air per cylinder but your runners are say 500cc in volume so you're getting that initial gulp of air before the any extra air/fuel comes out of the carb.


 

[NickJ]

2 hours ago, Roman said:

Perhaps you have the problem where the longer runners mean you need more initial enrichment, as in, when you stab the throttle your engine wants to intake (for example) 400cc of air per cylinder but your runners are say 500cc in volume so you're getting that initial gulp of air before the any extra air/fuel comes out of the carb.


 

Yes, that was my initial thoughts, simple answer and all! 

There is roughly ~0.35L per runner (including port and carb) feeding a 0.4L  cylinder that i'm surprised at how responsive it is during normal operation, it jumps off idle nicely, just after it sits idling for a while is where things get weird.

 

[Roman]

So when it is "steady state" is it fine though?

Like if at 3000rpm constant throttle does it have same problem? Or only in transient conditions.

[NickJ]

Yeah no issues at steady state, just rising off idle if its sat for a decent period

[Roman]

Carbs affected by heatsoak / lack of airflow through them somehow perhaps? 

Under vaccum the boiling point of petrol becomes quite low, and the hot intake manifold helps vaporize fuel very well.

So I would not imagine puddling to be the first thing to suspect under those conditions.

Do your carbs have a return line? Or does fuel just sit in them getting hot.

[NickJ]

Good point about fuel under vacuum, that would discount that theory with quite some weight.

Heat soak doesn't seem to be much of an issue with this set up, the runners don't seem to hold much heat at all, let alone the carbs. Having a fan on the water pump keeps airflow through the engine bay and being cross-flow keeps exhaust heat on the other side.

It doesn't have a return line, that is something easy to try and prove, just vent the fuel to an external tank for the purpose of a test?

[kws]

I need some advice lads.

The Rover has a Speeduino ECU fitted, and everything is mostly peachy, except for warm starts. Cold starts itll start within a couple of turns of the engine but when warm it takes ages of cranking to finally come to life, and usually needs about half throttle the whole time. Once started it runs and drives fine. No smoke etc on start.

I have tried playing with the priming pulse, cranking enrichment etc with little to no change. Im guessing, from the half throttle thing, it might be overly rich when trying to start warm?

[Yowzer]

Where's your intake air temp sensor located? Make sure that isn't heatsoaking. 

Check your fuel rail temp too, if that gets too hot from sitting it can vaporlock and make things hard to start too

[kws]

In the intake pipe after filter. I have all corrections for IAT turned off (but the issue was also present when they were turned on). IAT readings dont get that high, its one of the fast reading GM sensors and sucking in cold air.

Its on the underside of the black pipe near the strut tower, between the two hoses.

[yoeddynz]

I'd say heat soak on the temp sensor too...just like I was having.

 Why not try altering your iat  compensation table like I did a page or two back.  It's made a huge difference ! Its sorted it and it's so good now at hot starts.

[kws]

This is my IAT table so far, but will look back and find your post and see what i can change.

[kws]

@yoeddynz based on your post on the previous page, this is my updated IAT table. Will try it out and see if leaning it out a bit helps

I also need to remember to log and see what my IAT reads after a period of sitting when warm. The sensor is very close to the LH exhaust manifold, so i wouldnt be surprised if it gets toasty.

[kws]

Can conclude it didnt help, if anything it made it worse because my car hate idling lean and the intake temps didnt drop quick enough when running.

Starting the car in flood mode (injector cut) helps a lot, as does half throttle.

[kws]

Played around some more. Bumped the cranking advance to 25 degrees, and messed with the dwell and cranking enrichment. I also added some fuel in the IAT correction, as it was leaning out when the IAT was high. Its not 100%, but now it starts in about half as many turns of the engine as long as i give it a jab of throttle. Think ill leave it at that for now.

[downtrail]

if you need to open the throttle plate whilst cranking to get it to start you are too rich. 

You need to tune the fuel table at running temp with no corrections applied. this means looking at the temps and watching what they are doing and waiting if too high or low to make you next changes. once you have hit your desired afr then you can start working on corrections such as inlet air temp getting hot. this is best done with logs as its hard to see in real time. adjust to meet the desired afr at those temps only do not touch your fuel map. same goes for the warm up enrichment. you will need to start it whilst cold and adjust the warm up curve until your desired afr is met. perhaps a little richer whilst stone cold is normal. but there will be a point where you reach your target afr and the enrichment table will be at 100%. Same thing apples to cranking and after start enrichment these will generally be temperature based as well so you will need to do it at a few points to get it smooth. start with low numbers and add fuel for these do not start high and try go the other way you will run around in circles. 

 

As i said before if you need to introduce extra air you have too much fuel. If it coughs and splutters you need to add fuel(or your sooo rich you've fouled the plugs)

[downtrail]

And dependent on the wideband you are using and where it is placed you may need to go the cranking by feel and smell rather than relying on an accurate reading from the wideband unit. Slow exhaust speeds can do weird things in certain setups. 

[kws]

I cant use my WBO2 on start anyway as it takes a few seconds to initialise (only powered when fuel pump runs), so its all been feel and smell. Cold start and normal running is all good, its just the darn warm start.

Adding more timing and more fuel has actually made it easier to start when warm, but its still not perfect. The reason i added more fuel, was that when i was watching the logs i could see that when i gave it some throttle when cranking, it would increase the injector pulsewidth and then start. Pulling fuel from cranking made the car worse. Its all just a guessing game really.

[yoeddynz]

It's a daily trial too for adjusting cold starts too because you'll only ever get one real cold start in the morning to adjust. I tweeked my setup over a week or two of really cold winter starts to get it warm. 

Downtrail is right about really honing the table while every thing is warm.  I find that sitting for ages trying to get idles really mint can become a bore because things change as an engine sits idling. I think it's perfect, go for a drive, stop and then it's a bit fluffy again. Just takes a few attempts.  Never helped by the fact I'm running batch injection on a v6 with a very very light flywheel.