DPF fuel consumtion

[JimT]

(03-18-2024 )Boehmer Wrote:  I know that there are a lot of variables that need to be considered and I understand that a lot of this data can be downloaded from the trucks ECM but I was hoping to get more support or info on this in dealing with the IRS. I work on fuel tax related issued with about 100 trucking companies. There are fuel tax credits that can be pursued based on the diesel used during these DPF regenerations (nontaxable use). It is too difficult and cost prohibitive for these companies to bring their trucks in to have this data downloaded every quarter. We have been using the data given to us from Cummins on this from about 5 years ago. Its nothing more than a response to an E-mail I sent to them. The issue is coming under more scrutiny lately and I was hoping to get some type of supporting information that would back up our 10% diesel to DEF ratio. Not looking for specifics just something that would give some industry average. Regardless if that average applies to my clients or not. I need to find something to use.

Ah, I see your reasoning now. You should have mentioned this at the start as it changes the entire conversation. You're ultimate goal is to estimate how much fuel is used solely for active/parked regens (ie: in the DPF only and not in the engine) as accurately as possible and with supporting evidence. You've been using a diesel to DEF ratio to estimate the amount of fuel burned solely inside the DPF and now you're looking for more evidence to support your numbers. That also explains the desire for an industry average.

Well, good news and bad, i guess. Bad news, even if we on this forum could give you a better average estimate it probably wouldn't be considered legit for accounting purposes. Good news is, the data that you can pull from the ECM's should be able to provide you with what you're looking for. The frequency you collect that data is not as important as the documentation of the data, since you're still only going to use it as the basis to create an average. The larger the sample the more accurate the data. I agree that trying to calculate an accurate number for each truck every quarter, while possible, is not practical. Also, doing that would result is significant swings quarter to quarter and truck to truck that would probably cause more headaches then the data collection would solve in the first place.

Here's an idea to consider. As Rawze mentioned here the ECM stores the "- Total Doser fuel used. - the total fuel that has been consumed by the Aftertreatment system over the lifetime of the engine." I believe this is the actual information you are seeking. If you were to collect this info for each truck and/or fleet then you could aggregate and average it out to estimate what a single truck might use for a given interval. Of course you'll need to develop a calculation using engine hours, miles, and probably age, but that's just simple math. Once you have an average you can then apply it to all clients citing the data you collected along with your calculations as proof. Then you continue to update your data yearly (while unlikely to significantly change the result it shows that your using the most accurate/up to date data to make your claim). Always use the lifetime average as your basis since this will account for highs or lows do to external factors (weather, driving patterns, engine/vehicle maintenance, etc...).

The tricky part is that engine hours would be the most accurate time metric for the fuel used calculation since time spent idling isn't accounted for in miles traveled but directly contributes to soot load in the DPF which in turn determines regen needs. But of course engine hours are probably not something you or your customers track for accounting purposes, so you'll need to find a way to convert to miles for the purpose of the quarterly calculations.

As I continue to think about this I can't help but wonder if, after you compile this data, you find that there's a correlation between truck fuel mileage and fuel used for the dpf doser. IF a correlation like that can be established then estimating quarterly numbers for DPF usage would be as simple as calculating the average fuel mileage for each unit and applying the transformation function. But I'm probably getting ahead of myself here.

Anyway, it all hinges on collecting and compiling the data from your customers. The good news is that because your using lifetime averages across fleets of trucks, the timing and actual frequency that they provide the info isn't as important. That gives your customers the option of collecting this data while doing other work. That is, say it costs $100 each time a truck is hooked up for computer diagnostics at a shop. If they collect this data at the same time other diagnostic work is being done then it won't add to the cost burden for the customer.

If you can't tell from the above you have my inner nerd intrigued. If you can get access to the required data I'd absolutely love to help you develop calculations. Possibly even help write a small program to do the calculations as well, as long as any programs developed can be freely shared and/or distributed.

Question: since you're going through all this trouble to figure out fuel that's not being burned by the engine, what do you currently do to calculate or estimate fuel that is burned by a diesel powered APU, bunk heater or coolant heater which draws off of the same fuel tanks as the engine? In the case of the bunk or coolant heater the quantity used is probably so small that it's not going to result in more than a few dollars difference per year. But a diesel APU might use enough to justify calculation.

[JimT]

Why the DEF usage to DPF fuel usage calculation is inherently flawed:

DEF usage is based on SCR conversion efficiency. This is determined by data provided by the outlet NOx sesnor. There are specific levels of outlet NOx that the truck must try to achieve and DEF is consumed in the SCR canister as a means to achieve this goal. If everything is working perfectly then on paper the DEF used would be proportional to the fuel burned, but that's never going to happen in the real world. In reality there are so many variables that go into the amount of NOx produced (and therefor the amount of DEF needed to reduce that NOx) that equating them is never going to be accurate. Driving patterns alone will cause differences, not to mention the operation of the EGR valve, EGR cooler, turbo, multiple engine sensors and even exhaust leaks.

So if the DEF usage under normal engine operation fluctuates then there's no way to accurately determine the % of DEF used over-all to the amount of fuel burned exclusively by the DPF. This is probably why you can't get any good answer from Cummins as they most certainly understand the situation. While they might talk about averages and %'s for sales purposes (based on static testing?) they absolutely wouldn't want to provide those numbers as definitive proof for calculations based on real world operation.

Edit:
Also, the DEF doser valves, at least in some of the older trucks, are known to leak sometimes. And, the DPF doser can also leak fuel into the exhaust too. These factors can further skew the numbers you're trying to use as the basis for calculation.

[Papa_Buck]

If you are thinking about adjusting fuel consumption numbers down, are you also thinking about adjusting F&R tax reports that for audit purposes need to match purchase receipts. Adjusting your numbers down will require adjustments everywhere.