Car won't start unless cranking speed set to 1000 RPM

Recently installed new ID1050X injectors, upgraded fuel pump, Flex sensor, lightened flywheel, heavy duty clutch and fuel rails. I have adjusted the flow-rate in the ECU for the new injectors, and checked the fuel pressure at the FRP with and without the vacuum line attached.

Now the car will only start when the cranking speed is set to 1000 RPM or higher? If it is set to the normal value of 380 it will briefly fire and then just die.

Just wondering if anyone has encountered this before or has any ideas what could be causing it. Only thing I can think of is that it isn't getting enough fuel which is why it runs off the cranking table so I will try increasing the overall fuel trim and see if that makes any difference.

Thanks.

First, tell us what ECU you are using. Does the cranking fuel use the injector characterization or is it millisecond based?

Can you log the actual injector open time? What happens to this value when it leaves the cranking state, and transitions to the running state?

Can you log the lamba (or AFR) during the start? What happens to this when you transition to running?

My apologies, I'm using a Haltech Elite 1500 on a 2003 WRX. The cranking table is a percentage output, with the value for a cold start (20 degrees celcius) being set to 115% and dropping down to 65% for normal operating temperature.

So based on this, I increased the overall fuel correction first to 50% which made it run very rough for a few seconds and then die so I increased it to 70% which made it idle ok but as the post start correction table slowly reduced the additional fuel it was adding, the engine would start to splutter and then die. So I increased it to 85% and it ran fine albeit rather rich (around 11-12 AFR).

So the issue is definitely fuel related. I found a few forum posts of people having similar issues who had to input crazy injector dead times in order to avoid having to re-do every map so I tried these settings which allowed me to be able to zero out the overall correction value and then increase the main fuel map slightly in order to get it to idle at close to 14.7 AFR but transient throttle is going to need to be increased quite a lot as it stalls as soon as you open the throttle.

But obviously those dead time values can't stay like that forever, I am going to need to address the underlying issue which as this point I'm leaning towards either the fuel pump or the FPR but I've checked fuel pressure at the FPR and that is reading the correct value of 43 PSI with the vacuum hose disconnected but I have read in a few places that sometimes cheap FPR's (which this one is) can fail or cause issues when changing to a higher flowing fuel pump but I'd like to make sure before buying a new FPR.

Or should I just input the manufacturer supplied dead-time values and then re-tune all of the fuel maps with the much higher VE values that it now needs to run correctly?

Seems to be an issue with your fuel/VE table. If the car fires off then you are probably close on your cranking fuel and need to address the main fuel table and the after start tables. That's where i'd start anyways.

If I use the dead-time values that are listed on the ID web site I am going to have to adjust all of the maps in order to get everything correct. Haltech suggested the below approach (except that as I know the flow-rate it should be much easier):

As a DIY guide to find approximate flow and dead time:

Start with a normal VE table, something you can guess would look about right for the engine you have. Give or take you should be within approx. 10% by guessing. Set up your Target Lambda table to what you want to run for AFRs. Take an educated guess at the injector flow, and adjust the dead times so the engine idles. Bring the engine up under medium load and adjust the injector flow rate so your actual AFR matches the target. Come back to idle and adjust the dead times again so the actual AFR matches the Target. Repeat until the adjustments are very small, then go on to tune the engine. You can go as far as to disable the alternator and adjust the dead times at different voltages to keep actual and target at idle matching. This will get you flow and dead time data that is for the most part as accurate as you need to get the job done.

Injector Dynamics has done a lot of work and testing to be able to provide us with accurate injector characteristics. I have never had a problem using the data that they provide. It would probably be a waste of time to go through that procedure. It is however always good practice to check your lambda values at different voltages to verify.

I agree. I have spoken with ID support who have verified I have entered the injector data in correctly and it makes sense to start from known and tested base values and adjust everything according to these values rather than just entering random numbers and seeing what happens.

So I have started with a base map for my engine and entered in the dead-time values and flow-rate provided by ID and now the exact opposite is happening in that I am having to take out at least 30% of VE values in order to get the correct lambda but I'd much rather that than having crazy high distorted values like I had before. And I'm noticing with the new injectors the Post start fuel map needs as much fuel when hot than it does when it is cold or it just won't start whereas with the old injectors those values were all empty so I'm wondering if the new injectors might not be seating properly or something along those lines.

The difference in spray pattern is would result in a different post-start compensation (i.e. a short compensation while the fuel film builds up on the port walls).