Ford Variable Pressure Oil Pump

[ctechbob]

Doesn't really work how I thought it would.

Someone finally came up with the correct PID in order to pull oil pressure data from our trucks and today was the first time I've had a chance to log a short trip to work on the local highway.

Data here:
https://datazap.me/u/ctechbob/log-1719531096?log=0&data=4-8-9

It appears as though the pump only operates in 3 different modes. Low(30psi)/Medium(40psi)/and High(50psi), and for the most part, it looks like it is simply tracked off boost. No boost = Low, anything over a few psi it switches to medium, and full beans is 50psi+ (ish)

I guess I expected something different, such as lower idle pressures or the like. Don't know if the bigger EcoBoost motors operate the same way, although I suspect they probably do. Ford isn't exactly known for reinventing the wheel once they have a design.

Just thought it was something interesting people could look at.

 

[dnewton3]

That's good info.
Makes sense ... low boost should mean low bearing loads, and vice versa.

 

[ctechbob]

That's good info.
Makes sense ... low boost should mean low bearing loads, and vice versa.

Yep, although I was expecting the low to be lower, although thinking about it, the truck is supposed to shut off at idle, so that would be 0psi. I just have my Auto Stop/Start disabled and forget that is supposed to happen.

 

[twX]

It looks to be strongly correlated to boost, but normally these variable pumps would only be controlled based on rpm, which you're not logging. RPM would also be correlated to boost, especially with an automatic transmission, so it may just seem that it's the boost/engine load that controls the pump even if it's actually entirely based on engine speed.

The oil flow/pressure requirements of most engine components is related to engine speed. At low rpm, hydraulic components like those for the VVT system will often set the minimum oil pressure requirements, though sometimes it's the main bearings. At high rpm, the rod bearings will set the minimum oil pressure requirement.

Here are some examples from studies on variable pumps from Nissan and Toyota:

It could make sense for the system to take load into account as well though. It would allow for lower pressure at mid-rpm when the engine load isn't high enough for the piston squirters to be necessary. I'd take some more logs while logging rpm as well, at both high and low engine loads to see if engine load actually makes a difference.

 

[ctechbob]

I'll turn on the RPM logging for our travel trailer trip tomorrow. I usually just leave it off.

With the trailer and in tow/haul it will be telling since it will downshift a ton to try to use engine braking, so high rpm, but no load.

 

[SC Maintenance]

Very interesting. The new toyota engines specs something similar.

ECU controls the pressure. Engine has to be at operating temp, and its then based on calculated load and RPM. Starts at 16 PSI and as the engine either encounters more load or goes above 2500 RPM, it ramps up:

 

[21_6.7f450]

Bob, is that in Torque Pro? What app do you use? I want this bad for my 6.7 Powerstroke.

 

[ctechbob]

Bob, is that in Torque Pro? What app do you use? I want this bad for my 6.7 Powerstroke.

It is.

As for the graphing, you just export the log in a .csv and that Datazap site handles the rest.

 

[21_6.7f450]

I'm just looking for numbers. Do I need to update it? How do it find that PID?

 

[meep]

Chrysler’s variable oil pump in the venerable pentastar is a 2 stage (hi/lo) pump. They may have been the first, or at least were one of the first. They aren’t as sophisticated as they sound, but clearly they make enough of a difference. AND, demanding less from the pump not only reduces pumping losses, it also prolongs the life of the pump drive.

Great data @ctechbob!

 

[OneLonelyBurrito]

Mazda has a similar system for its newer 2.5L N/A engines. Most of the time is running about 20psi, but as the RPM and/or load go up. The ECU increases the oil pressure.

 

[Exhaustgases]

Nice how something simple and proven has been turned into a complicated mess.
The new saying if it isn't broke, then go and fix it till it does break.

Yeah now lets use 0W-8 oil.

 

[ripcord]

I'd be curious to see how oil temperature plays into the pressure strategy. Does it default to max pressure for a while after a cold start?

 

[OneLonelyBurrito]

I'd be curious to see how oil temperature plays into the pressure strategy. Does it default to max pressure for a while after a cold start?

At least on my Mazda, oil temperature doesn't seem to play a role. It only varies with RPM and load as far as I can tell. That's how I assume most others are programmed too.

 

[meep]

Mmm. Not sure what I’m trying to say.

All oil pressure will vary with RPM on an engine-driven pump? Pressures generally drop as the oil temps increase. The multi-stage (“variable”) pumps using stair-step their output which I would *think* would be notable steps in an rpm/pressure graph. I tried to make sense of the chart 4 posts above, but don’t quite understand it - looks like the pressure starts at 8 bar (112PSI), has a very mild step function around 1750 rpm, continues to increase in pressure (not a lot) with RPM to about 160PSI, then something really bad happens around 5000 rpm. Clearly I don’t know what I’m reading….

 

[OneLonelyBurrito]

Mmm. Not sure what I’m trying to say.

All oil pressure will vary with RPM on an engine-driven pump? Pressures generally drop as the oil temps increase. The multi-stage (“variable”) pumps using stair-step their output which I would *think* would be notable steps in an rpm/pressure graph. I tried to make sense of the chart 4 posts above, but don’t quite understand it - looks like the pressure starts at 8 bar (112PSI), has a very mild step function around 1750 rpm, continues to increase in pressure (not a lot) with RPM to about 160PSI, then something really bad happens around 5000 rpm. Clearly I don’t know what I’m reading….

So let start by saying, the graph is in absolute pressure and not the typical gauge pressure most people are use too. So to get gauge pressure subtract off 101.3 kpa, atmospheric pressure. The numbers will make more sense then.

In order to read the graph, you have to think of it as a bunch of circles, and when you see a number/color, everything on the outside of that circle is that pressure. So in the bottom left corner you see the 250kPaA (~150kPaG or 21psi), everything from that point is 21psi. Then when you hit the 350kPaA everything outside of that circle is 36psi. And that continues on with each circle.

That random 350kPaA in the middle of the 250kPaA zone is most likely an error of some kind, whether that be with the measurement or oil pump control.

 

[meep]

So let start by saying, the graph is in absolute pressure and not the typical gauge pressure most people are use too. So to get gauge pressure subtract off 101.3 kpa, atmospheric pressure. The numbers will make more sense then.

In order to read the graph, you have to think of it as a bunch of circles, and when you see a number/color, everything on the outside of that circle is that pressure. So in the bottom left corner you see the 250kPaA (~150kPaG or 21psi), everything from that point is 21psi. Then when you hit the 350kPaA everything outside of that circle is 36psi. And that continues on with each circle.

That random 350kPaA in the middle of the 250kPaA zone is most likely an error of some kind, whether that be with the measurement or oil pump control.

OHHHHHHHhhhhhhhhh! Goodness. That makes more sense.

 

[ctechbob]

It looks to be strongly correlated to boost, but normally these variable pumps would only be controlled based on rpm, which you're not logging. RPM would also be correlated to boost, especially with an automatic transmission, so it may just seem that it's the boost/engine load that controls the pump even if it's actually entirely based on engine speed.

The oil flow/pressure requirements of most engine components is related to engine speed. At low rpm, hydraulic components like those for the VVT system will often set the minimum oil pressure requirements, though sometimes it's the main bearings. At high rpm, the rod bearings will set the minimum oil pressure requirement.

Finally got a chance to sit down and post some more stuff on this.

Log from a drive home the other morning where I tried to get the truck to log a few very specific things.

https://datazap.me/u/ctechbob/log-1719657990?log=0&data=4-6-9-10&zoom=6329-7277

OP is also tied to RPM in the 2.3 Ecoboost (As @twX mentioned). On the way home a few days ago I manually shifted at redline, being careful not to roll into boost, and OP jumped up to ~50psi, as you would expect.

(I run Havoline LifeLong 5W30 in the truck if anyone cares)

High RPM, engine in vacuum, high oil pressure

Low RPM, rolling into boost, high oil pressure.

Full throttle, just before a redline shift. High RPM, High Boost, High oil pressure (As you would expect, although I would actually expect the pressure to be higher than 46psi)

 

[ctechbob]

I'd be curious to see how oil temperature plays into the pressure strategy. Does it default to max pressure for a while after a cold start?

Little hot to do a 'cold start' here in GA, but I grabbed this today. 96 degrees here today when I set off for work, truck hadn't run for a little over 24 hours.

HUGE spike right on startup, then regulated down to 40 psi and then to 30 where it stays.

You can see the RPM dip where I put the truck in reverse and then start backing out of the driveway, so this is maybe 10-15 seconds worth of data.

 

[ctechbob]

I was wondering if that huge pressure spike was an anomaly. Looks like, nope. Not sure why the thing spikes that high. First graph is leaving work this morning at about 74 deg ambient after a 13 hour off period. The second is leaving to come in to work tonight at about 90 degrees.

Things I find interesting.

You can clearly see the crank event at 595 RPM and the engine catching pretty much immediately. The oil pressure lags a bit but you can see it build quickly.

The Datazap website doesn't decode the device time properly, but here is the raw data. I'm kind of amazed at how short the crank period is: