Hot Start issues

[MarcusP]

So I have been fighting starting issues and with the help of this forum most everything has been worked out except erratic hot start.

So it will start great over and over then will get finicky backfire while cranking then usually start.  Another odd issue it seems really picky about cranking fuel enrichment, 1 or 2% change on the table will go from backfire to crank no start etc.  Please take a look and let me know if anything stands out. 

 

I also have a cold start with warm up but the log is too big to upload and I can't figure out how to chop it up. 

 

[Adamw]

A backfire would suggest a trigger issue rather than a fuel issue.  Looking at the little info I have on the RZR triggermode I can see it uses the MAP sensor for sync.  For MAP sync to work reliably the ECU needs to see the correct amount of MAP depression during cranking.

Unfortunately I dont know these engines or vehicles well and we have little info in our library.  I have looked at a couple of other RZR maps we have and it looks like the trigger offset is normally 0 on these engines, but in yours it is set at 270.  This makes me think you are mis-syncing a lot of the time.  

Is the MAP sensor stock?  

Are they normally E-throttle stock?

Do you have access to an oscilloscope?

 

[MarcusP]

10 minutes ago, Adamw said:

A backfire would suggest a trigger issue rather than a fuel issue.  Looking at the little info I have on the RZR triggermode I can see it uses the MAP sensor for sync.  For MAP sync to work reliably the ECU needs to see the correct amount of MAP depression during cranking.

Unfortunately I dont know these engines or vehicles well and we have little info in our library.  I have looked at a couple of other RZR maps we have and it looks like the trigger offset is normally 0 on these engines, but in yours it is set at 270.  This makes me think you are mis-syncing a lot of the time.  

Is the MAP sensor stock?  

Are they normally E-throttle stock?

Do you have access to an oscilloscope?

 

MAP is a Link 4 bar, still using the factory tmap for IAT.
Yes its a stock throttle body

I do not have access.

 

Polaris factory turbo motors are a 270 firing order. The Factory NA is a 180...  This is a 270 motor. But that does not mean the trigger offset is correct thats how the original tuner set it up.

 

[Adamw]

Let me get a bit more info off the engineers how this trigger mode works, im not sure if it uses raw MAP sensor voltage or if it converts it to a pressure drop first and what type of pressure/voltage drop it requires to satisfy the sync test.  Might be a day or two as they are neck deep in it at the moment.  

[MarcusP]

I appreciate it thank you!!

[Adamw]

Ok I got some info from engineering yesterday about how this triggermode works.  What I would like you to do is temporarily move your AN Volt 1 wire (MAP sensor signal) into the trigger 2 position.  This is so we can capture the crank signal and MAP signal together using the built in trigger scope while cranking.  I will then be able to confirm if the ecu is getting the correct MAP signal to allow reliable sync.  

I would suggest disabling fuel so it doesnt attempt to start and do a few trigger scope captures like this. 

[MarcusP]

35 minutes ago, Adamw said:

Ok I got some info from engineering yesterday about how this triggermode works.  What I would like you to do is temporarily move your AN Volt 1 wire (MAP sensor signal) into the trigger 2 position.  This is so we can capture the crank signal and MAP signal together using the built in trigger scope while cranking.  I will then be able to confirm if the ecu is getting the correct MAP signal to allow reliable sync.  

I would suggest disabling fuel so it doesnt attempt to start and do a few trigger scope captures like this. 

I will give it a try. 
 

since it seems to be a lot worse when iat is above 106° should I warm the motor up first or can you see what you need with a cold motor?  

[Adamw]

Both cold and hot would be good.  Vacuum can be lower when cold and you also generally have the throttle open more when cold to make matters worse.    

[MarcusP]

.

[MarcusP]

I dont seem to be getting anything on trigger 2.  I moved the AN v1 wire to the trigger 2 spot.  My gauge is reading -13.5 so I know its the correct wire. 

 

 

[Adamw]

Can you temporarily set trigger settings like below.  Also, make sure you only click the capture button when the engine is turning, not before. 

[MarcusP]

here you go

 

 

[Adamw]

Hmm, still doesnt look right.  Not much movement in MAP there and it is showing about 0.8V, it should be a bit above 1.0V for atmosphere.  

Have you moved only the ANV 1 wire like so?

[MarcusP]

3 minutes ago, Adamw said:

Hmm, still doesnt look right.  Not much movement in MAP there and it is showing about 0.8V, it should be a bit above 1.0V for atmosphere.  

It’s pulling 2 or 3 lbs vacuum. Want me to pull the vac line off the sensor? 

[Adamw]

No we want to see the vacuum signal.  But it is showing a dead flat line, which would be very unlikely which each cylinder taking gulps of air out of the manifold.  

[MarcusP]

On 3/23/2022 at 5:54 PM, Adamw said:

No we want to see the vacuum signal.  But it is showing a dead flat line, which would be very unlikely which each cylinder taking gulps of air out of the manifold.  

On the gauge it doesn’t change much during crank. With a 270 firing order they gulp pretty close to each other. Only 90° from 2 back to 1. And the throttle body is at 0.5 to 3.5% from 0 to 1000 rpm. 

 

Here is the current tune and yes I verified I moved the wires as indicated on the drawling

 

The motor has cooled off to about 130

 

 

[Adamw]

I think I see the problem.  I think the trigger scope must have a low resolution voltage measurement when in hall mode.  In one of those latest scopes you done I can see the voltage is only changing in steps of about 0.16V.

Can you try changing trig 2 to reluctor mode like below as that may use different measurement hardware.

 

54 minutes ago, MarcusP said:

On the gauge it doesn’t change much during crank. With a 270 firing order they gulp pretty close to each other. Only 90° from 2 back to 1. And the throttle body is at 0.5 to 3.5% from 0 to 1000 rpm. 

You can see significant MAP oscillation and a general trend downwards during cranking in your earlier logs (first post) and this log is only 40Hz - so we should see much more detail in the triggerscope which is a few thousand Hz.  A mechanical gauge has significant damping so wont show much.  

[MarcusP]

I don't see much difference, hopefully you see something.

 

tried playing with the ethrottle to force it to do something

 

[MarcusP]

Is it possible that since I’m running a 4 bar map sensor that while cranking it’s only a 4-5% swing?  The factory is a 3 bar so cranking would use 8-10%.   Can I wire the factory sensor to the trigger 2 and use it to sync and leave the 4 bar to an v1 for map?  
 

 

[Adamw]

Ok I just done some testing on the bench.  Unfortunately the G4+ triggerscope is not going to give us anything useful.  With trig 2 in hall effect mode the scope has a relatively low resolution, something like 0.15V is the minimum change it can see, this is not good enough to see the detail we want.  With trig 2 in reluctor mode it has much better resolution but it mustnt reference ground so it cant measure AN Volt 1. 

I cant think of anyway to validate that the MAP signal meets all the requirements to give reliable sync without a standalone oscilloscope.  Unless you know someone with a G4X ecu you could borrow - they capture AN Volt 1 natively in the built-in scope.  

 

Attached below is your original tune that I have made a few tweaks to to hopefully give a stronger MAP depression when cranking.  In your earlier logs I can see it cranks around 600RPM, so I have shifted the 500RPM column in your E-throttle target table so that we target a smaller throttle opening below 700RPM, then as soon as it fires the throttle will move to a more normal position.  I also changed the 70deg cell in the idle base position table so that wasnt opening the throttle too much more on a hot start and I lowered the start up step in the 80 & 100deg cells.  Your trigger offset may be 0 or 270 - it is hard to know if it was syncing correctly or not before so it may or maynot have always been "wrong".  Try it both ways to see if this map gives less backfires etc.  

 

 

3-21-22 test1.pclr

[MarcusP]

2 minutes ago, Adamw said:

Ok I just done some testing on the bench.  Unfortunately the G4+ triggerscope is not going to give us anything useful.  With trig 2 in hall effect mode the scope has a relatively low resolution, something like 0.15V is the minimum change it can see, this is not good enough to see the detail we want.  With trig 2 in reluctor mode it has much better resolution but it mustnt reference ground so it cant measure AN Volt 1. 

I cant think of anyway to validate that the MAP signal meets all the requirements to give reliable sync without a standalone oscilloscope.  Unless you know someone with a G4X ecu you could borrow - they capture AN Volt 1 natively in the built-in scope.  

 

Attached below is your original tune that I have made a few tweaks to to hopefully give a stronger MAP depression when cranking.  In your earlier logs I can see it cranks around 600RPM, so I have shifted the 500RPM column in your E-throttle target table so that we target a smaller throttle opening below 700RPM, then as soon as it fires the throttle will move to a more normal position.  I also changed the 70deg cell in the idle base position table so that wasnt opening the throttle too much more on a hot start and I lowered the start up step in the 80 & 100deg cells.  Your trigger offset may be 0 or 270 - it is hard to know if it was syncing correctly or not before so it may or maynot have always been "wrong".  Try it both ways to see if this map gives less backfires etc.  

 

 

3-21-22 test1.pclr 167.05 kB · 0 downloads

I’ll try everything in the morning and report back. Thank you for all the time. Do you think the G4x ecu would do a better job with the rzr?  
 

or should I add a cam position sensor to the system?  

[Adamw]

9 minutes ago, MarcusP said:

I’ll try everything in the morning and report back. Thank you for all the time. Do you think the G4x ecu would do a better job with the rzr?  

The G4+ should be able to do a similar job in terms of syncing reliably, just the G4X has a better scope for setting up and diagnosis and the MAP thresholds etc are user adjustable.  Having said that, the way the RZR trigger code works is pretty smart and should be quite fool proof (ie it is designed to work with non-stock range MAP sensors and partially flat battery etc). 

 

10 minutes ago, MarcusP said:

or should I add a cam position sensor to the system?

That would certainly be a valid option too if you have the engineering capability.

[MarcusP]

11 hours ago, Adamw said:

Ok I just done some testing on the bench.  Unfortunately the G4+ triggerscope is not going to give us anything useful.  With trig 2 in hall effect mode the scope has a relatively low resolution, something like 0.15V is the minimum change it can see, this is not good enough to see the detail we want.  With trig 2 in reluctor mode it has much better resolution but it mustnt reference ground so it cant measure AN Volt 1. 

I cant think of anyway to validate that the MAP signal meets all the requirements to give reliable sync without a standalone oscilloscope.  Unless you know someone with a G4X ecu you could borrow - they capture AN Volt 1 natively in the built-in scope.  

 

Attached below is your original tune that I have made a few tweaks to to hopefully give a stronger MAP depression when cranking.  In your earlier logs I can see it cranks around 600RPM, so I have shifted the 500RPM column in your E-throttle target table so that we target a smaller throttle opening below 700RPM, then as soon as it fires the throttle will move to a more normal position.  I also changed the 70deg cell in the idle base position table so that wasnt opening the throttle too much more on a hot start and I lowered the start up step in the 80 & 100deg cells.  Your trigger offset may be 0 or 270 - it is hard to know if it was syncing correctly or not before so it may or maynot have always been "wrong".  Try it both ways to see if this map gives less backfires etc.  

 

 

3-21-22 test1.pclr 167.05 kB · 1 download

unfortunately it would not start on that tune.  I went back to the tune I was running before the scope tests and it took a little while but it fired up.  It acted like it was flooded.

[Adamw]

Suspect possibly just not enough air now.  You could try putting the E-throttle target back to how it was but leave the start up step and idle base with my settings.

You actually have a lot less cranking fuel than most engines need so it is hard to imagine it is flooding but less air may have an influence.  

[MarcusP]

On 3/24/2022 at 3:01 PM, Adamw said:

Suspect possibly just not enough air now.  You could try putting the E-throttle target back to how it was but leave the start up step and idle base with my settings.

You actually have a lot less cranking fuel than most engines need so it is hard to imagine it is flooding but less air may have an influence.  

There were a lot of weird things with the tune you sent. Errors, Ignition turned off etc. Im not sure if they were intentional or not so I didn't play with it.  Here is the tune I have been running all day.  and some logs

 

 

I also have a bunch of Hot start logs but I can't upload them because of size restrictions.

I suspected the low cranking fuel was because im running a pair of 1700 cc injectors on a 925 cc motor.  Factory is only 850cc injectors.