Anyone know Honda K series? Can't get swap to start, no injector pulse.

I'm finishing a K24A2 swap into an 89 EF hatch. 06 Element wiring harness, PZD ECM, KTuner in ECU installation from a dealer.  

The issue we have now is there is never any injector pulse, no matter what we do. Here are some of what we have done and checked:
-Added additional grounds, even from G101 direct to the head and chassis.  
-Confirmed there is power and ground to every single sensor, injector, and coil
-Run the injectors with straight 12V power and fuel and confirmed they are operational 
-The motor will chug a bit with starter fluid, so it has spark
-Replaced crank position sensor, confirmed the crank trigger is in place.  

I've tried the tune from my dealer and an 03-05 Accord MT base tune as I have a cable throttle body. Neither make a difference. I've confirmed the "disable immobilizer" is checked on both, removed EVAP, etc. None of it matters. 

Of note, the tune I have on the ECU from the dealer has no values in any of the tables (fuel, spark, cranking fuel) and no values in the 2d map, but does show a 3D map with values, which looks similar in profile to the Accord maps, which do show values in the table and 2D map.  

I'm leaving for a race weekend in 3 days, my team is all paid and we need to get this going but I'm out of ideas and open to any suggestions.

I recall reading somewhere that the K24A2 has a different crank wheel layout than other K24s, so you would need an -A2 specific program in the computer.

Not sure what that could mean given that it apparently does have spark.

Do the injectors match the ecu? Ie correct impedance, not relying on a external resistor pack or something?

Can you watch the ecu "live"? 

That might give you some good info. Even without engine running you can check that the temp sensors are getting realistic data. During crank you should see cam/crank pulse, etc. 

Does the ecu need to see a signal back from the fuel pump? Or a fuel pressure sensor? 

Maybe something dumb like knock sensor reading a bunch of false knocks?

I'm just shooting from the hip here, no specific k knowledge. But I also think it should do better than just chug on starting fluid. It should damn well run. The parking lot build mr2 would chug on starting fluid, and then we realized the firing order was 90degrees out. Once fixed, it ran great on starting fluid.

Stock trigger wheel to the motor, stock 310cc injectors shared among all K24s, no resistor pack needed.  I can see most sensors live but  when cranking the voltage drop (within spec) makes it tough to get instant readings.   Known good readings from MAP, TPS, IAT, ECT.  Replaced the crank position as a desperate move, no change, but let me confirm that the crank trigger wheel is in there.  Confirmed pinouts on each sensor to make sure it is all correct with the variety of parts I’m using  

no O2 sensor hooked up yet (was trying to get it running and then button up a few details) but a lack of 02 wouldn’t matter for this, would it?  

it shouldn't.

In reply to Sonic :

When you say you have known good readings of the sensor, are you getting an engine speed reading when you are cranking?  That will confirm that you are getting a good crank sensor reading.

And is there a way to figure out what functions have no values in them?  I'm confused when you say there are no values in the functions or tables, but the 3D terms have values- the tables are 3d...  so what, specifically, has no values in them?  

(the O2 sensor should not effect crank)

WAG, I had a similar issue with my Infiniti G20, turned out the injector driver circuits in the ECM were fried.  Swapped in a known good ECM/ECU and the car fired right up.  Any way you can get the ECM tested?

If it is like most other things i have worked on, you should 12v+ to one wire on each injector whenever the key is on and during cranking, and it pulses through the ground side. Which of those are you missing? One common mistake i have seen on custom jobs is forgetting that some 12v+ circuits turn off when cranking, if your injectors are mistakenly fed through one of those you will have a no start.

Update: I took a half day from work and have been working on the car.  

I did get to speak with the tuner and did a datalog.  We've narrowed it down to a lack of engine speed reading to the ECU.  That's good.  I also confirmed that the reason I wasn't seeing values in teh ECU tables was that the tune he gave me has them hidden as it was a tune he wrote.  It's there though, confirmed by the 3D table view.  

I have no idea what to do about it though as I've done everything I can think of: 

-tried the crank sensor that came with the low milage junkyard motor, and a new one from the FLAPS

-removed the chain case and confirmed the crank trigger wheel is there, and installed in the correct direction, and that the timing chain is still in the correct position. 

-Verified I have power and ground and signal wire continuity to the crank position sensor.  

Any other ideas on why I wouldn't have a crank position sensor reading?  

In reply to Sonic :

Did you wire the sensor, or was it part of a harness?  If you did it, it needs to be a twisted pair, and (oddly enough) the polarity matters- so you may need to swap the wires.

This is a sensor that reacts to a ferrous metal passing by a magnet, which then reacts and sends voltage up and down the wires.  Not really a signal and ground, and given what it does, having a twisted pair make sure that there's not much that messes with the signal.

If you have access to a scope, you can check the signal you are getting from the sensor.

OEM wiring harness for this sort of engine from sensor to ECU, no changes made.   I had no idea about the polarity, good tip.  One of the only things I can think of is to try a different power source for the sensors and see if that makes a difference for any reason.  

ECM not reading RPM, or no sensor output as measured on a scope?

The ECM doesn't see a signal.  I don't have a scope to check it with. 

Is there an air gap spec on the sensor? If it's too far away from the triggers, it won't see them.

No air gap apec that I can see anywhere in Honda’s tech info or anyone ever referring to.  There is no adjustment other than I suppose you could put a washer between the timing case and the sensor body to space it out.  It is not contacting the wheel now.  The sensor voltage at the ecu does change if I remove the sensor depending on how close it is to something metallic.  When my wife gets home I’m going to try to see the voltage change at the ecu with the engine cranking.  

Tonight I also tried using a larger battery to make sure voltage drop wasn’t too much, and tried wiring the power feed for all of the engine sensors directly to the battery with a thick gauge wire to make sure there is no issue there.   Still no difference.  

Im not sure what else I can even try at this point.  The only thing I haven’t tried is another ECM, but I can’t just get one from the junkyard as they need a board soildered in to be able to tune it and remove the immobilizer.  

Again, there should not be voltage fed to the sensor- UNLESS it's a Hall Effect Sensor.  I don't know what Honda uses, but I do know we use a variable reluctance sensor, which generates it's own voltage, and should not be tied into the sensor power system.  I can't recall where the ground line goes, but, for sure, like the base O2 sensor, there's nothing driving the sensor- it generates it's own voltage without help.  Oh, NO CONTACT- just a small air gap.  I've seen what happens if they touch- it's a mess of very fine copper wire....

Does any of your team have access to a scope?  I know you are in a massive hurry, given the upcoming race.  But before changing the ECU, you should try to see it generating the tooth pattern sent to the ECU.

BTW, if the engine is spinning, the ECU should see it.   I'm 100% sure that the base "tune" is capable of a very slow crank at -20 F with thick oil and poor battery and it will still start.  So nominal or "slow" spinning at current ambients should be good to go.  Focus on the sensor and how the ECU sees it.

The sensor is definitely supposed to have a 12v source, ground, and then one signal wire going to the ECM, I've checked the Honda schematics and plug pinouts.  I don't haev a scope here (can have on one Friday, but that's a bit late).

The last thing I have checked is leading me to think the ECM itself has an issue.  I have the ECM cover off so I can read the voltage inside the ECU itself, ruling out all of the connectors.  By doing that, and turning the motor over and monitoring the voltage, I can see that there is a voltage pulse getting into the ECM from the crank sensor, but the software in the ECM doesn't see it.  Without an engine speed signal it will not fire the injectors.  

At this point I feel pretty comfortable that I've ruled out everything but the ECM.  

-ECM is not seeing crank position sensor voltage

-Tried 2 different crank position sensors (CKP)

-Confirmed cam/crank timing is correct and the crank trigger wheel is intact and correctly installed

-Confirmed continuity between the CKP plug and ECM (this portion of the wiring harness is unchanged from how Honda made it, and wires have been unloomed and individually inspected along their length with no abnormalities.

-Confirmed voltage and ground at the CKP plug

-Confirmed power source is solid and voltage during cranking is adequate (10.2v min)

-Confirmed that there is a CKP pulse entering the ECM

So, now I guess I'm trying to find another KTuner 05-06 Accord/Element/RSX ECU, so I'll email my tuner tonight who provided this one, and another moderately local (2.5 hours) shop that might have one.  

You're near Philly?  Surely there's a K-Series FB group that could help you out.  I'd let you borrow my ECU but I'm way too far.