Calculated Load vs. Absolute Load

" The VE test can't make a distinction between a pumping problem and a measurement problem. But, neither can Load."

If I'm understanding you correctly here, I would have to disagree.

If you assume...
WOT is either enrichment mode or CL
An O2 will switch high > stoich
For a skewed MAF to affect VE to any useful level, it will have to have a significant error

Then I think the problem is solved. I have been using and teaching this with examples for a long time.

"Conclusion: Absolute Load is kinda useless."

Per the J1979 definition, it is a "normalized" value.

Normalization of a data set I believe brings a good bit of value, but it does take some explaining.

I have a long term project on just this topic and it will likely end up being part of a new class.

I don't really care for generic data as it is so slow, so I've always used whatever definition of engine load the OE side of the scan tool uses and use other pids to give it context. The reason for the project is that I have to address this stuff industry wide, not just for myself.

I think I saw that you posted numerous GM examples. I have found GM to use at least 4 different engine load definitions, all with the same pid name. Other OEs seem to be a little less haphazard

Ah snap...

Sorry Tyler, I didn't realize I hadn't read to the end of the thread

An less techie name for volumetric efficiency would be cylinder filling effectiveness, how much "weight" of fresh air is actually getting in to refill the cylinder.

Flatrater wrote: Ah snap...

Sorry Tyler, I didn't realize I hadn't read to the end of the thread

:lol: No worries! As you can tell, I've been fumbling around with Load PIDs for awhile now. If I'm not drawing the right conclusions, I'm happy to hear about it.

I have a long term project on just this topic and it will likely end up being part of a new class.

Very interested. :ohmy: How do I get on a mailing list to get notified about this?

I have found GM to use at least 4 different engine load definitions, all with the same pid name.

Would love to hear more about this.

Flatrater wrote: " The VE test can't make a distinction between a pumping problem and a measurement problem. But, neither can Load."

If I'm understanding you correctly here, I would have to disagree.

Not trying to be argumentative, but would you mind checking my understanding? I had a Lexus that I believe was a good example of a measurement error causing a low VE calculation.

'99 ES 300, 3.0L, dirty MAF numbers:





And after cleaning:





In this situation, both Load and VE showed low values. With the VE test alone, I'm not sure how a distinction could be made between a measurement or breathing problem. Not unless the upstream sensors are considered. Or, was that your point? :silly:

Not unless the upstream sensors are considered. Or, was that your point?

I'm going to start out by apologizing for how thrown together this will be. Hectic morning and a lot on my plate

Yes, that was what I had in mind. I think you're already doing the same thing, I'm just suggesting it can be easier. Perhaps I even misunderstood you. Sorry if that is the case

Perhaps this will help someone else, so I'll continue. As I'm sure you know, many technicians struggle with trim codes and power complaints. I think it mostly comes down to exposure, they just don't see enough problems to get good at it.

I can't read your data (too small), but let me describe how I would deal with what I assume was going on.

I'm guessing you had a P0171/P0174 and perhaps a lack of power. I would setup for a data capture and take a short drive. Idle in park, idle in drive, light load, moderate load, and WOT.

I would then reset the trim and do it again. Many systems use LTFT as a modifier in power enrichment mode and by resetting trim, I get a better idea of how healthy the MAF is.

Something fun to think about - how do we get LTFT corrections for an area of operation that disables trim learn? Sorry, just had to throw that in.

Back at the shop I'd take a look at the data. The best way to explain what I do with the data is with an example, but here are some thoughts.

Whether we use load or VE, there is a pretty wide range of good-bad, but I have found that by the time a code sets or there is a performance complaint due to a skewed MAF, both are usually (but not always) pretty low. A number of times I could pick this out without moving the vehicle (WOT with the brakes applied)

I'm going to interrupt myself here and suggest something... In between normal operation and a fail scenario, there is a good deal of change... In this "gray" area, performance is degraded and we can see it if we know what to look for. Ok, back to where I was going

I use load since it is already on my scan tool. I think the push for using VE came from the various definitions of engine load that have been used on the enhanced side.

On a normally functioning system, at WOT I expect to see the upstream O2 (or AFS) to either go rich or keep toggling. The latter because many systems will still be in CL at WOT. Regardless if the system stays in CL or goes into OL enrichment, it should never go lower than stoich and stay there.

If I don't see that, then I use load to help determine whether I have a measurement error (MAF - which includes the ducting) or a fueling error. Mechanical errors (in which I include intake and exhaust restrictions) can be sorted out using the data as well.

If I'm dealing with the gray area that I mentioned above, my road test data capture can help me a great deal. For example, excessive engine ping. I find it is most often caused by the MAF under-reporting air flow (timing advanced).

There is so much more to explore. For example, the MAF output curve. It is not linear and as the sensor degrades, the errors will most often show up in a particular region before codes set etc.

I can't read your data (too small), but let me describe how I would deal with what I assume was going on.

Ah yeah, sorry about that. :blush: It didn't have lean codes stored, but it should have.

Flatrater wrote: Many systems use LTFT as a modifier in power enrichment mode and by resetting trim, I get a better idea of how healthy the MAF is.

Something fun to think about - how do we get LTFT corrections for an area of operation that disables trim learn? Sorry, just had to throw that in.

Really? :blink: Are there any case studies you could link me to that'd expand on this?

I use load since it is already on my scan tool. I think the push for using VE came from the various definitions of engine load that have been used on the enhanced side.

Thank you! :woohoo: This filled in a lot of blanks for me, strange as it may sound.

I've been hearing teachers in various classes (ATG in particular) preaching VE. VE this, VE that. VE every car. And the whole time, I've been wondering why - doesn't Load tell me the same thing? Now it makes sense. VE isn't dependent on how the OE set up the Load calculation.

For any SD Premium subscribers following that may (not) be following along, the Lexus RX330 Low Power series has been bugging me for this exact reason.

www.scannerdanner.com/scannerdanner-prem...le-codes-part-1.html

It was gutless with skipped timing, but the Load went to 100% during a WOT snap in park. :blink: As far as I'm concerned, that's a useless Load calculation.

For example, excessive engine ping. I find it is most often caused by the MAF under-reporting air flow (timing advanced)

Right on the money here. The same Lexus I got those VE numbers from also had knock sensor codes for both banks. No knock and no codes after cleaning, but the service advisor gave me the most sideways look when I didn't want to sell knock sensors. :lol: :silly:

Really? :blink: Are there any case studies you could link me to that'd expand on this?

I think this capture from a ScanShare case study will cover it. I added the VE pid. In other words, I opened up the data file and created the PID using the VE formula on each data set. I find it helps when teaching the subject.

I think you'll see open loop and a LT correction. I selected a few pids for you, but you can also pull up the rest of the list.

www.scanshare.io/share/ad1_d3jx40e2IJDWJ...g#0,8,10,14,16,22,24

As for ATG and the VE thing... well... Let me put it this way. To me it seems that ATG is simply parroting the common view of the day and trying to get some traction with this "new -old- thing" VE.

I think it is counter productive. I would never "start" my diagnosis with it and I think doing so makes it harder for technicians to see what they need to see.

I buy most everybody's book and such. Knowing how hard it is to create accurate and useful material, I cut them all a lot of slack. But dang... some of the stuff is really out there...

My 2 cents anyway...

I'm really glad this thread had been revived

I didn't notice (or at least remember) the load pid on the Lexus reaching 100 when that series came out, but I'm reviewing it again in light of this conversation.

Thanks for your contribution Flatrater!

Certainly looking forward to more in the future.

Flatrater wrote:

Really? :blink: Are there any case studies you could link me to that'd expand on this?

I think this capture from a ScanShare case study will cover it. I added the VE pid. In other words, I opened up the data file and created the PID using the VE formula on each data set. I find it helps when teaching the subject.

I think you'll see open loop and a LT correction. I selected a few pids for you, but you can also pull up the rest of the list.

www.scanshare.io/share/ad1_d3jx40e2IJDWJ...g#0,8,10,14,16,22,24

Not trying to nag too much, but can you link me to the case study this was associated with? "From case study April 09 1" is what I get out of the notes attached this capture. I have questions! But I'm sure many of them would be answered by the case study. I searched the ScanShare Case Studies...

As for ATG and the VE thing... well... Let me put it this way. To me it seems that ATG is simply parroting the common view of the day and trying to get some traction with this "new -old- thing" VE.

I think it is counter productive. I would never "start" my diagnosis with it and I think doing so makes it harder for technicians to see what they need to see.

Agreed. FWIW, I have yet to run into a situation where VE would have told me something that Calc Load or Abs Load didn't. Is my faith in these PIDs misplaced?

Not trying to nag too much, but can you link me to the case study this was associated with?

Sorry for the delay, I don't visit here often

I'm pretty sure it is this one. You will have to login first.
www.scanshare.io/CaseStudy/Get/0113d92c-...ef-9ad9-9d3785a91025

Progress to remove Java and Flash requirements is in full swing for all of the older stuff. This came to mind because I just finished updating it.

FWIW, I have yet to run into a situation where VE would have told me something that Calc Load or Abs Load didn't. Is my faith in these PIDs misplaced?

I've found the same thing. My guess is that you are taking into account other pids etc. as I do for those gray areas. Which would be just as gray using VE IMHO.

Flatrater wrote: I'm pretty sure it is this one. You will have to login first.
www.scanshare.io/CaseStudy/Get/0113d92c-...ef-9ad9-9d3785a91025

Got it, thanks!

I think I'm misunderstanding your point with the long term corrections? :silly: I've seen the long term PID's show values during power enrichment on many vehicles, but never paid too much attention. Does the amount of correction have diagnostic value? Have you seen significantly different MAF test results after resetting the trims?

Progress to remove Java and Flash requirements is in full swing for all of the older stuff. This came to mind because I just finished updating it.

Awesome. While we're on the subject, how are you creating the VE PID (or custom PIDs in general) in some of those captures? Is this something that normal users can do, or only administrators?

I've found the same thing. My guess is that you are taking into account other pids etc. as I do for those gray areas. Which would be just as gray using VE IMHO.

Noah wrote: I didn't notice (or at least remember) the load pid on the Lexus reaching 100 when that series came out, but I'm reviewing it again in light of this conversation.

I think Paul noted it in the video? But didn't spend too much time discussing it. I'll have to review it, too.

Then he did a VE test and saw a big problem. :ohmy: One of the cases where doing a VE calculation was key. This is why I WANT to rely on Calc and Absolute Load, but not sure if I always can.

It's obviously too late now, but I would have been interested to see if the Global data would have shown a different Load value, compared to the OEM data.

I've seen the long term PID's show values during power enrichment on many vehicles, but never paid too much attention. Does the amount of correction have diagnostic value? Have you seen significantly different MAF test results after resetting the trims?

I've seen dramatic changes to the O2s at WOT when resetting the trims. So, if I understand your question, yes the LT correction (at least for most GMs - I suspect for all) remains during open loop.

While we're on the subject, how are you creating the VE PID (or custom PIDs in general) in some of those captures? Is this something that normal users can do, or only administrators?

I don't think it will be added. It is a lot of coding and pid selection validation Right now I do it with a utility program I wrote that I needed for some training I had developed.

Also, keep in mind that my goal is to convince technicians to work things out using what is already on their scan tool.

On that note, consider this. Experienced technicians can usually quickly differentiate between a restricted exhaust and a cam timing fault just by the driving the vehicle. So how do I teach a new guy that? To not need to play with calculators etc?

It is all about data. Cam timing does not get worse when trying to flow more air through a catalyst. A restricted catalyst does

Flatrater wrote: I don't think it will be added. It is a lot of coding and pid selection validation Right now I do it with a utility program I wrote that I needed for some training I had developed.

Also, keep in mind that my goal is to convince technicians to work things out using what is already on their scan tool.

That's cool, I was mostly interested in adding it for my own comparison purposes. Figuring out how different OE's Load calculations compare to each other, Global and VE. I know the behavior of a few different makes/models off the top of my head, but I'd like to quantify it. Nothing a VE calculator can't do, I was just hoping to simplify the process.

Posting up a truck that threw me for a loop recently. Seems like the Load PIDs kinda let me down...

'07 Ford F-150 has a noticeable lack of power from a stop, and at WOT. I took a drive with Global data, and saw these (reduced the PIDs down to the relevant parts):



Calculated Load says everything's good. 100%. But Absolute clearly isn't happy at 54%. I consult the ATG VE Calculator to break the tie:



Tough to argue with that. Calculated Load lied to me.

After the repair, I recheck the numbers:





Much better. Absolute Load = 1, Calculated Load = 0.

Hi Tyler

I noticed you've used global data. That's good, it means the load pids have very specific definitions.

Calculated Load says everything's good. 100%. But Absolute clearly isn't happy at 54%

Calculated load will be 100% at WOT, so if one is looking to use this to evaluate breathing or the MAF, it won't work. It is't meant to be used that way.

Absolute load tells you most of what you need to know. "Most", because it depends on what you're trying to determine.

Can you tell me what using the VE calculator added to your diagnosis?

Flatrater wrote: I noticed you've used global data. That's good, it means the load pids have very specific definitions.

Because of you! I started doing that regularly after our discussion earlier in this thread about varying Load calculations on the OE side.

Calculated load will be 100% at WOT, so if one is looking to use this to evaluate breathing or the MAF, it won't work. It is't meant to be used that way.

Respectfully, are you sure? :huh: Without digging up specifics, I can recall many known good vehicles that didn't go to 100% Calculated Load at WOT. I'll check out some of my saved captures.

So, in the theoretical case of a dirty MAF sensor, you're saying Calculated Load will not show a low percentage?

Can you tell me what using the VE calculator added to your diagnosis?

Confidence, mostly. Absolute Load was definitely the red flag here, but running the VE calculation made me certain. The issue with this truck was actually installed by another technician, so I needed all the evidence I could get before going to management.

Sorry this is late, been a little busy

Generic data has specific definitions. However, that didn't start until 2008.

I typically see 99% instead of 100%. This is, of course, assuming there isn't a problem. The PID is essentially giving you cylinder loading which is used for spark and EGR scheduling etc.

I have been buried in other things for some time, so I would be very interested if you would post (or send) captures that illustrate what you are seeing.

Here is a capture of calculated load using a generic scan tool, but it is a 2006 Envoy. It doesn't follow the later rules. It does peak at 99%, but it is way high with far less throttle.

This was made with an AE scan tool interface and their phone app (Gioto).

www.scanshare.io/share/k7NNFC7RGEGLcnOte6_byg#0,1,4,5