Fuel Trim Questions

linneje

Well-Known Member
My son has a 2013 5.3L Silverado LTZ. It has always run perfectly, except when starting a cold engine at very cold temperatures (like, below -25 C). No codes thrown. I have access to a good scan tool, so I thought that I would try to look at Fuel Trims etc. when trying to do a cold start at such a temperature. Vehicle did start for me, but these are some typical readings right after start:

Intake Manifold Pressure 9 psi
Long Term Trim Bank 1: -16%
Long Term Bank 2: -19%
Short Term Bank 1: -2%
Air Flow Calculated: 11.84 g/s
MAF Sensor: 14.27 g/s

Also, after driving around (warm engine):
Idle: Long Term -12%,-14%
Short Term : -3%
IMP: 9 psi
Air Flow: 5.93 g/s
MAF: 6.74 g/s

1500 RPM
Long Term -16, -19
Short Term -2
IMP 10psi

2500 RPM
Long Term -5, -10
Short term -11
MAF 21.40 g/s

I am new to this technical a diagnosis, but it seems to me that such high negative values indicate a problem which is leading to his difficulty starting; when it is so cold. I realize that these values are just snapshots, and not a graph, but it seems like the PCM is trying to lean out the mixture. However, I thought that fuel trim isn't used when the vehicle starts (not until closed loop operation).
1) If the engine is really cold and the air is more dense, shouldn't it still start okay if the mixture really is too rich all the time?
2) And aren't those high negative values really strange? Or is that because it is so cold here.
3) Why do I have negative short term and negative long term in a snapshot? That doesn't make sense to me, probably because I don't understand enough.

Can anyone get me a little smarter on what is happening here? Do I need to graph a longer drive?
 

Mooseman

Moderator
You can't look at the cold engine trims as it is running in open loop and not using O2 sensor data until it goes into closed loop (warm). Basically, it's like running on choke. However, your warm LTFT is a little high negatively. Basically it's taking fuel away from the mixture thinking it is running rich.

Maybe someone better than I can delve deeper into the reasons why this may be happening as that is a little beyond my capabilities except to say maybe the O2 sensors are getting lazy or there is something causing the slightly rich condition.

What has been the maintenance like? Plugs, air filter? Once the engine is warm, it runs OK? Some engines just don't like the extreme cold.
 
OP
OP
linneje

linneje

Well-Known Member
Thanks for the information. This confirms what I thought, the trims don't matter when starting. My feeling is that the mixture is somehow too lean at startup. However, when running it seems like it is too rich.

I will try to take a test drive and look for other factors, like oxygen sensor performance.
 

Chickenhawk

Well-Known Member
Can you also let us know the coolant temperature when warm? This can tell us if the thermostat is sticking open. This will definitely affect your long term FT.

The other issue with Silverados at those temperatures is the block heater. Are you plugging it in? If so, has the block heater or block heater cord ever been changed from the factory cord?

GMs use a special block heater cord that senses the outside temperature and prevents the heater from warming the coolant unless it is below about -18C. If owners ever change to a non-OEM cord, it causes real starting problems because the computer thinks it is below -20 and pours in lots of fuel, but the actual coolant temp is much higher and it ends up running far richer than it should.

The solution is to stick with the OEM block heater cord, or to plug it in on a timed outlet. (At -25C, it only needs about 3 hours to warm.)
 
OP
OP
linneje

linneje

Well-Known Member
Good points. I will check that out, since he bought this used a couple of years ago. He hasn't changed the block heater cord, but the previous owner may have. He always plugs it in when -15C or below.
 

Kelly@PCMofNC

Well-Known Member
Fuel trims are something that work over time, so you would really need to log them over a 15-20 minute period or so and then look at any trends there. With that being said though, the newer trucks are pretty sensitive - if they are rich/lean more than 12.5% or so average it will set a rich or lean code. I'm assuming it doesn't have any codes stored?

But secondly as someone else said, it is not using the fuel trims at start up anyway, as it is in open loop. It has a set of predetermined spark/fueling from the factory (also based on IAT/ECT) that tell it what to do.

When you say it has trouble starting, is it just turning over slow? Or does it sound healthy turning over, just refuse to fire for a bit? Some other things to consider are the block heater as one mentioned, but also that the cold can be taxxing on a battery that's on its way out. Also wouldn't hurt to put a fuel pressure gauge on it and see that it has 58psi at start up and the whole time it is trying to start in the cold.
 
OP
OP
linneje

linneje

Well-Known Member
Thanks for the good suggestions. It is a factory cord for the block heater. No codes stored. We ran the truck for quite a while - trims are all normal (both long term and short term) except for idle. At idle (closed loop) the LTFT is over -10%, usually -12% or more, and the STFT is about -2%. Very strange.

There are things that we can try with the scan tool, like resetting the fuel trims and letting the LTFT repopulate over time. But the only real issue he has is with the very cold start. My hypothesis is that the intake manifold has a small leak that only occurs when the engine is very, very cold. This causes some extra air which bothers the start, but as the engine warms the leak seals off and so it will start properly even after turning the truck off. Monkeying with the fuel trims is not going to change the starting issue, so we will probably leave them alone.

I realize that the rich AFR mixture for a cold start is set by the manufacturer - we could go to the dealer and have them reprogram this, but I am not sure that this is a good solution. I think these AFRs are pretty standard for startup so it should work with factory settings.

The battery tests good. It turns over well, but just kind of coughs a bit and doesn't want to stay at idle. I think we could also try a check of the fuel pressure to see where it is at. Typically it will start anyway - if it is really cold it might take him 4 or 5 tries to get it to run without dying immediately. It is so tough, because it only does this in extreme cold, and of course you don't want to be doing engine work out in the extreme cold.
 

Mooseman

Moderator
Not a bad idea to check fuel pressures, cold and warm. Throttle body clean? I know it's not the same as on the I6 but maybe it affects idle air control. I do believe there is a relearn to be done after this.
 

budwich

Well-Known Member
As suggested, check your fuel pressures. Inadequate fuel pressure will result in "rich running" because of poor "spray pattern"... thus resulting in "poor burn". It could also result in poor start as perhaps the pressure is not there to provide proper fueling at start. You can try the "turn the key to on, then to off then on then off " and ultimately to "start". This does multiple "fuel pump charging" to build up pressure. If it starts better after this "effort", that may suggest a poor fuel pump and or pressure regulator.
 

MRRSM

Lifetime VIP Supporter
With the exception of 'Getting it Straight From The Horse's Mouth' from @Kelly@PCMofNC ... THIS Video from The Wells Tech is THE Best Primer on How Narrow Band Oxygen Sensors, Open and Closed Loop Conditions and Long Term vs. Short Term Fuel Trims all work in any PCM-ECM F/A Management Systems... and posed in an Excellent Shop Classroom setting:

 

budwich

Well-Known Member
another area to check... if your vehicle has a MAP sensor. Its relatively cheap to change out and may have direct impact on some of the run results. Your trims, both at idle and at "speed" seem to be indicating rich running which potentially indicates that the system is "over reading" the amount of air in the system (not withstanding fueling issues like pressure). Both MAP and MAF are possible candidates.
 
OP
OP
linneje

linneje

Well-Known Member
Good point. On the scan tool it gives both MAF reading and the air flow (calculated). How is it calculating the air flow if not using the MAF reading? The two values are not the same (check out the snapshot above) so there must be an independent way of calculating the air flow.
 

budwich

Well-Known Member
I do believe some form of the MAP is used to "map" back into the density tables to "guess" at run data especially at idle... if my understanding is correct. The MAP itself doesn't provide any "air values". I think, in my "readings" that the MAP can have significant impact on "fuel economy" which is a reflection of both fueling and air. You can see the impact by unplugging the module... basically the vehicle will likely not run. Again... it depends on whether your vehicle has both MAF and MAP or not.
 

Online statistics

Members online
1
Guests online
93
Total visitors
94

Members online

Forum statistics

Threads
20,530
Messages
596,645
Unanswered questions
2
Answered questions
2
Members
13,671
Latest member
aczlan
Top Bottom