RX7 FD3S - Cold start and idle issues when cold; no issues when hot.

Good Afternoon,

I have noticed swapping to the link g4+ from the apexi pfc, my cold start and idle is quite poor. For e.g on the pfc the car would start up instantly and really quick and hold idle at 12-1300rpm.

On the Link ecu however, It cranks for what feels like an eternity (5s) before actually starting up and than idle quality does not match my idle target table of 1300rpm. On hot starts no issues whatsoever start up within 1s on half a turn of the ign key. Please can someone help point me in the right direction.

Please can anyone point me in the right direction to

1. Getting the cold start up within asap; Not certain why it wont on the link ecu.

2. Why it wont respond to my idle target rpm of 1300rpm?

I have done a quick log and can see its cranking at 147rpm for 5s and than gets to 250rpm and higher and than starts up. So I suspect if i can get it to crank at 250rpm from the start....it will start up immediately on a full cold start up....but not sure how to do this in the software?

If your engine is in fact only cranking at 147 rpm, you need to investigate your power system (grounds too) and starter. Low, low RPM is the death of rotary compression, hence real compression numbers are normalized to 250 RPM by Mazda and aftermarket compression testers for actual diagnostics.

Look into swapping RX8 2kw starter guts into your REW housing, or there are a few vendors that do it for you.

Engine compression is not the issue, its a rebuilt engine with compression in low 8s, as mentioned on the pfc starts up just fine on a cold cold start but on the link it doesn't. I believe this is a software over sight on my part somewhere

What David is saying is that the low initial cranking speed is usually due to low voltage across the starter - this can be on the power and/or ground/return side, and/or a weak/tired starter motor and an aftermarket starter will crank the engine faster. It could be a tired battery that isn't able to supply the required current while holding voltage - it is NOT something 'software' can fix. Personally, because I'm Scottish, I'd pull the starter and clean the commutator with some fine W&D and electrical cleaner and lubricate the bushes first, as that can make a BIG difference in the cranking speed, current draw and voltage drop.

As David also said, the rotary really needs to be spun at a good speed to ensure the dynamic compression is where it should be - they are notorious for poor seals resulting in hard starting.

That said, it may indirectly be the cause of your problem - ALL ECUs have a minimum operating voltage below which they may not operate, and if you've changed from one that was more tolerant of the cranking voltage drop to one that's less so, the higher initial voltage drop may not trigger it until the engine is spinning fast enough that the voltage drop from cranking rises above that threshhold.

When it does eventually start, does it idle well, that may also point towards whether it's a mapping problem or not.

As you're coming out of the Northern winter, the fuel may cause some issues, too, with starting if you're using a high ethanol content one.

Copying my reply to your same question on the Link forum for everyone else's info:

Your low cranking RPM is due to your low battery voltage. I would generally consider anything less than 10.5V a problem, you only have 7.0v.

For the idle speed issue we will need a PC log with all parameters logged and a copy of the tune.

Thank you for spotting this Adam. I will try another battery and see if this resolves the cranking issue. However, the battery I am using is fully charged and functional so this is strange to see. I'm glad you have picked this up on the log.

Will make another log of the starting up issue and tune.

Thank you everyone for helping, really appreciated.

Looking at that voltage graph, it looks like the ECU is seeing less than 12V even before cranking - this is less that I would expect from a charged battery unless there's a significant drop in the power and/or ground wiring for the ECU. I would suggest reviewing the wire gauges and connection points while you're at it - for example, it you are taking the power for the ECU off the same terminal on the starter the battery cable is attached to, the ECU is seeing the drop across the cable, but an independent, oversize (to minimise that wiring voltage drop*) feed to the ECU directly from the battery will avoid that drop, if you treat the whole wiring loom like that, you may avoid other problems elsewhere with voltage drops.

*I'm quite fussy about voltage drops and for pwer circuits I really like to see close to 0.1V copper (wiring) losses, with less preferred. There is a small financial and weight cost, but I consider there are times it is worth it.

Gord, thank you very much I will need to investigate why the voltage is dropping like that. The battery in question is a Odyssey PC680 and I have it on a trickle charger and its showing the battery is in good condition. Adam pointed out the ECU is def seeing a drop in voltage as confirmed by the log.

I used jump leads today to connect my daily driver to my odyssey PC680 and right away the cranking was much better and I had my laptop connected to the ecu logging the cold start and this time the ecu did not disconnect from my laptop. Usually if im connected to the ecu via laptop on crank the ecu disconnects from the laptop. The previous log of the cold start was a log file i downloaded from the ecu as on cranking the laptop will disconnect from the ecu.

As requested, Adam I have attached a cold start log of the car coming up to temperature and the map. I feel the cold start could be much better optimized. This time it started from cold in approx 4s from the key turn; was hoping to improve on this and get it to start much more quickly.

Any input appreciated.

Ok, You still appear to have a starter/wiring/voltage drop issue. It was showing 14.48V before cranking which suggest the ecu is seeing the correct battery voltage, but it is still dropping to 8.2V during cranking in that log. There is also a spike in TP and a few other analog channels right when you start cranking which may suggest the issue is a bad ground.

Also something odd going on with fuel pressure, it varies a lot with battery voltage so that suggests fuel pump is seeing the same voltage issue. You have only 326kpa differential press during cranking Vs 423kpa when idling.

Your idle speed doesn't sit on target because you have idle control set to open loop and the base position table hasn't been tuned - so the idle valve isn't open enough.

In the map attached I have made a few small changes. I have given it more dwell at low batt voltages since it was flat before, I have enabled pre-crank prime. I have changed idle speed control to closed loop and adjusted the base position table a little. You can use the file compare function to see what I changed.

After loading this map you will have to do a TPS calibration as it is currently showing 1.7% all the time so idle control will never enable like that.

Do another cold start/warm up log and we can use the logged closed loop values to tune the idle base position table.

Starter/starter wiring/grounding still needs to be looked at.

Adam - Thank you very much, I will investigate the grounding issue which seems to be effecting the parameters you mentioned and a closer look at the starter and battery grounds.

Will do another log as you've requested. Once again thank you.