Jump to content
will15

Cold start e85

Recommended Posts

On 6/15/2020 at 8:17 PM, finnigan001 said:

Sometimes I wish I knew where to find the actual functions that showed how these variables operate. But then I’d be down another rabbit hole. 😂 I’ve had issues with trying to work out other variables like fuel pump. Still don’t think that works how advertised.

Any other advice on the cold start up?  Looks like just the multiplier and maybe crank spark seems to be enough(I haven’t touched yet).

That is where it gets difficult! if you are curious this is what we have to decipher to do the writeups we do. You can see how it would be extremely error prone as well and a lot of the time brute force trial and error with lots of logging is quicker than us trying to follow the code. We can however quickly see which scalars/tables are used in a calculation from the below logic without actually reading the code.

Here is a small snippet of the speed density logic (in total its about 100x longer than this screenshot).

 

  • Like 1

Share this post


Link to post
Share on other sites

 

15 hours ago, finnigan001 said:

I would just like to confirm these are my findings as well.  Stock numbers in the FG cranking table were all under 1.  So before reading this thread i assumed the equation for cranking fuel was cranking lambda * cranking multiplier. So being under 1 it would give a final lambda as a percentage of the cranking lambda value.  Lower being richer, higher being leaner.  However no matter how low i made the cranking multiplier it never richened or made made it easier to start on e85.  Watching the inj PW confirmed that lowering the number was indeed leaning it out.

Below stock table (fg series 1) and below that E85 cold start enriched

Hopefully helps for others sorting out e85

Thanks for this @finnigan001 you can see this is where good old experimentation wins.

These are the numbers I used in my BF with E85, starts first go at 5c° (as cold as it gets here). This is with ID1000s with 1-2% trims. With injectors with a poorer spray pattern you might need to go richer, same with if your injector slopes are out. If it gets below 0°C it will need further tweaking as I could never test it that cold.

image.png.651922ee4e12bc3eb4ba67077cfa23ce.png

I also modified the following

auF2325 cranking airflow

image.png.45c333d860167c9ef375e42a7c2fe40b.png

auF0180 Cranking Lambda

image.png.eeed07c7d0412b8baed452698a847fbd.png

auF0411 cranking throttle angle

image.png.4c41da036f5500f2fff548656151a3c1.png

auF0178 lost fuel

image.png.38130dacdff6e45805bdb01b65167237.png

Fuel base cold auF0173

image.png.1b0416fedfebe2f0811cd1688ca13501.png

Closed loop delay auF0162

image.png.842e2108234bc10859ec8bfd4fdc63a5.png

 

 

 

  • Like 1
  • Thanks 1

Share this post


Link to post
Share on other sites

The above fixed my lean tip in without needing to enable pump shot gain. I found I needed a 2 minute open loop window with lots added to the lost fuel table to fix that. Even on 98 with the ID1000s it would stumble on tip in when cold.

I can stab the throttle off idle when stone cold now and it will not misfire. Same with flattening it at 1500rpm when stone cold (not that it is a good idea to do that when cold). Drives like factory imo.

Share this post


Link to post
Share on other sites
1 hour ago, Roland@pcmtec said:

That is where it gets difficult! if you are curious this is what we have to decipher to do the writeups we do. You can see how it would be extremely error prone as well and a lot of the time brute force trial and error with lots of logging is quicker than us trying to follow the code. We can however quickly see which scalars/tables are used in a calculation from the below logic without actually reading the code.

That's awesome. Believe it or not i would actually love to trace that assembly to find out whats what with some of this stuff.  Sometimes it feels like you changing things and no matter what you do there's no corresponding measurable result.  And like you say, experimentation wins... sometimes :P 

 

45 minutes ago, Roland@pcmtec said:

These are the numbers I used in my BF with E85, starts first go at 5c°

Appreciate the time and effort to share these.  i think we all battle the same things.  Trying to start straight away, "lean" stumble which i believe is actually a false lean and possibly too much fuel/lost fuel, and cold start lean tip-in's.  It's interesting to see different ways we using to solve these issues.

Share this post


Link to post
Share on other sites
15 hours ago, finnigan001 said:

Sometimes I wish I knew where to find the actual functions that showed how these variables operate. But then I’d be down another rabbit hole. 😂 I’ve had issues with trying to work out other variables like fuel pump. Still don’t think that works how advertised.

Any other advice on the cold start up?  Looks like just the multiplier and maybe crank spark seems to be enough(I haven’t touched yet).

Can you send me your tune so I can have a look.

Share this post


Link to post
Share on other sites

I haven't changed anything other than what i've previously screenshotted.

21 hours ago, Puffwagon said:

Depending on how cold the weather is and how much ethanol is in the mix you might want to extend the time that it takes to swap from open to closed loop (base fuel enable time).

This seem to be the key with post ignition fuelling when cold.  I've lengthened the delay as you and roland have mentioned an pretty much eliminated the first stumble after start up.  Switching to closed loop must lean it out too much compared to the cold fuel table.  Doesn't mean i'm getting commanded but its not leaning out to the point of stumbling/missing.  I'll keep plugging away but is much better.

  • Like 2

Share this post


Link to post
Share on other sites
Posted (edited)

I’m pretty happy with my cold start but there is one issue I have not been able to resolve, when coolant temp is around 50c and you restart the car it starts fine but when you put it in gear the commanded lamda go’s form 0.95 to  0.7 and make it hunt for about 10sec then it will go back to lamda 1 but then it will go to lamda 1.17. I think there is a hard limit set to 1 but that gets over ridin. Wideband confirms this as well.

If I don’t drive off and put it back in park it will go back to lamda 1 but slip it back to drive and back to 0.7, it will do this in the first 1-2 mins from start up.

Edited by will15

Share this post


Link to post
Share on other sites
5 minutes ago, will15 said:

the commanded lamda go’s form 0.95 to  0.7 and make it hunt for about 10sec then it will go back to lamda 1 but then it will go to lamda 1.17.

 

What are your trims doing just before and when this happens? Is there a tip in fuel thing that is used when it gets put into gear? What fuel source is it using when this happens? Is there some funky interpolating going on at this temp? Is there a large difference between the base fuel and cold fuel map that it's possibly switching between?

Just a few things I would think of looking at, mebbe someone has something else to add?

Share this post


Link to post
Share on other sites

Long terms are turned off, this is a flex fuel tune but I don’t think that’s causing any issues because I tuned another  non-flex e85 auto and it did the same thing. Coolant temp vs lamda table looks fine, there must be some multiplier table in there doing it.

Share this post


Link to post
Share on other sites
On 6/26/2020 at 3:34 PM, Puffwagon said:

 

What are your trims doing just before and when this happens? Is there a tip in fuel thing that is used when it gets put into gear? What fuel source is it using when this happens? Is there some funky interpolating going on at this temp? Is there a large difference between the base fuel and cold fuel map that it's possibly switching between?

Just a few things I would think of looking at, mebbe someone has something else to add?

I will see if I can log fuel source. Everything else looks fine

Share this post


Link to post
Share on other sites
1 hour ago, will15 said:

I will see if I can log fuel source. Everything else looks fine

The reason I suggested logging fuel source is in case the cold map was causing the issue. The more channels you log, the more chance of noticing something happening, but I'm sure that's fairly obvious.

So the stft is not causing it to do it? The fuel trim would react to an afr change like that but it's not moving before the afr changes?

Is the nbo2 sensor all good? Sometimes the wiring can get hot and can melt, causing issues.

Just brainstorming here, maybe it helps, maybe not.

Share this post


Link to post
Share on other sites
8 hours ago, Puffwagon said:

The reason I suggested logging fuel source is in case the cold map was causing the issue. The more channels you log, the more chance of noticing something happening, but I'm sure that's fairly obvious.

So the stft is not causing it to do it? The fuel trim would react to an afr change like that but it's not moving before the afr changes?

Is the nbo2 sensor all good? Sometimes the wiring can get hot and can melt, causing issues.

Just brainstorming here, maybe it helps, maybe not.

Nar it’s all happening in open loop, I’m laptop died today hopefully have it back soon and I will do some more logging.

Share this post


Link to post
Share on other sites

I've attached a parameter file to this post which has the cold start settings for my territory on e85. The changes are to fuel cranking lambda, fuel base cold and fuel base cold decay.

Worth noting is that although I have the decay set to a minute, my tune will switch to closed loop after 30 seconds. I might need to extend it but haven't had a play with it yet.

The car started at 8C this morning with the attached settings.

E85 cold start settings, fuel cranking lambda, fuel base cold and fuel base cold decay.param

  • Like 1

Share this post


Link to post
Share on other sites

I've since rescaled the injectors (bosch 1000's) with id1050x data to suit e85 (available on the injector dynamics website) and I've removed the delay table completely. The cranking is the same and the cold base is the same and it starts every morning these past few days. The high slope is 67 and the low slope is 58.

It isn't perfect (I just haven't put in the time yet) but as you mentioned this info will point people in the right direction and should definitely gets peoples cars starting on e85.

 

Share this post


Link to post
Share on other sites

Just a side question. How are you guys powering your flex fuel sensors or are you just doing straight E85 tunes.

I ask coz i notice that the voltage to the flex fuel sensor during the switch from acc to on must be low enough for the sensor to turn off and send 0v to the ECU.  The gauge etc stays on, but the signal to the ecu drops out causing the 98 tables to be used initially. Once the car starts to crank over then the values come back to the High Slope as voltage slowly returns to 11.5-12v.

I think one of 2 things are happening. Either battery is no good and voltage is dropping to low - i have seen as low as 9-10v on startup- not sure if that is normal.  Or ecu turns off/reduces voltage to accessories during the switch from ACC to ON. The later will affect me as i just did the ole cigarette lighter power source hack.

So just wondering are your gauges wired the same or by a relay etc.  If you are using a relay, is there a power source that is constant power during the switch from ACC to ON that remains 12v that can be used as a trigger wire that you've found.

Share this post


Link to post
Share on other sites

Straight e85 for now. I'll get a flex sensor etc one day. I'm interested to see how you go with this as it's likely you aren't the only one seeing this.

Share this post


Link to post
Share on other sites
6 hours ago, finnigan001 said:

Just a side question. How are you guys powering your flex fuel sensors or are you just doing straight E85 tunes.

I ask coz i notice that the voltage to the flex fuel sensor during the switch from acc to on must be low enough for the sensor to turn off and send 0v to the ECU.  The gauge etc stays on, but the signal to the ecu drops out causing the 98 tables to be used initially. Once the car starts to crank over then the values come back to the High Slope as voltage slowly returns to 11.5-12v.

I think one of 2 things are happening. Either battery is no good and voltage is dropping to low - i have seen as low as 9-10v on startup- not sure if that is normal.  Or ecu turns off/reduces voltage to accessories during the switch from ACC to ON. The later will affect me as i just did the ole cigarette lighter power source hack.

So just wondering are your gauges wired the same or by a relay etc.  If you are using a relay, is there a power source that is constant power during the switch from ACC to ON that remains 12v that can be used as a trigger wire that you've found.

Mines powered and earthed from the rear o2 sensors, you need to use the ecu for power and earth or the voltage off set will wreak havoc on the readings

Share this post


Link to post
Share on other sites

9v is way too low. Anything less than 11v will reset the ECU. 

You likely have a dead battery. 

You can reconfigure the voltage look up table to use the e85 tables at 0v and 98 at 0.02 as a fail safe if you want. 

Share this post


Link to post
Share on other sites
9 hours ago, Roland@pcmtec said:

9v is way too low. Anything less than 11v will reset the ECU. 

You likely have a dead battery. 

You can reconfigure the voltage look up table to use the e85 tables at 0v and 98 at 0.02 as a fail safe if you want. 

Hey Roland can we re-scale some of these cold start tables to get better resolution in temp range we need. Some of them are in the -40deg i would say it will never see that kind of temp.

Share this post


Link to post
Share on other sites
51 minutes ago, will15 said:

Hey Roland can we re-scale some of these cold start tables to get better resolution in temp range we need. Some of them are in the -40deg i would say it will never see that kind of temp.

Nothing stopping doing that. Use the rescale wizard

 

 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...