Differential Pressure Data, Pure 1 & Motorcraft

Joined
Aug 12, 2005
Messages
4,563
Location
NW Ohio
Hopefully nobody got too blue in the face waiting for this. A lot of data to go through, a busy schedule and other priorities... what can I say?

Background:

I installed a differential pressure test setup on my F150HD (5.4L 3V Modular V8). Mine is a two gauge setup with senders on either side of the filter. The gauges are Isspro stepper motor types... very accurate... and mounted in a one-wire-daisy-chain using Isspro's Performax system. The heart of the system is a CU which Isspro calls the ESP (Electronic Signal Processor). You can install up to 17 gauges on one wire with this system but, more importantly, it has a datalogging system (albeit a primitive one) that you can connect to a Windows laptop. I soiled my All-Mac-All-the-Time household by bringing in a used Windows laptop, which remains hidden when not in use. ( : < )

The one thing this setup does not datalog (though it could if I installed a sensor in the right place) is Engine Oil Temperature (EOT). I can monitor that another way, and datalog it too, but only in 2-3 minute intervals via an EFI programmer. Ditto for RPM and that is why you see these datapoints listed separately and sparsely. I went nuts trying to synchronize the two datalogs and finally resorted to good, old fashioned note-taking. While driving and operating the PC, I have to take note of the line number as the datalogger is recording and jot down the RPM or EOT at that moment. RPM is easy usually as I need only record the peaks or the steady state. EOT is ALL-IMPORTANT but since it rises and falls relatively slowly, it's easy to keep track of.

The datalogger records four points per second, so even a short recording can yield dozen of pages of data... and it records all the data for which there is a sending unit. I cannot select which ones to record, so on a long record, I can have a hundred or so pages of data to sort thru because, 1) I can't post that much here, 2) you would go into a coma trying to sort thru it for the DP data, 3) you aren't interested in seeing what the rear diff temp or the fuel pressure at the time the DP is being tested.


Viewing the Data:

In reality, the P1 and MC Data comparison is not useful because of the viscosity disparity. That's why I did not get too anal over sorting out all that P1 Data. Things will get more interesting when I test other filters with the same oil. I am including the P1 stuff because it illustrates the effect oil viscosity has on DP. I plan to run another P1 with the 5W20 so we will see the differences.

As mentioned above, the unit records four data points per second. I left the line number on for my own sanity and ability to see where I snipped that particular batch of data. Ditto for the time, which is also important for interpretation. After some experimentation and reflection, I have decided the important data was Engine Coolant Temp (ECT.. pay attention to the abbreviations), EOP-1 ( the inlet side of the filter), EOT-2 (the outlet side of the filter), EOT, RPM and Differential Pressure (DP) If I knew more about how to work Excel, I could probably set it up to make those DP calcs, but I am Excel-challenged and too busy to learn what I need to know. I did some of the calcs (simply subtract EOT-2 from EOT-1 to get the DP) at what I thought were the crucial points. You can do the subtraction yourself on any line where I didn't. Along the same lines, I added RPM only where I deemed it important from what I recorded in my notes.

I picked what I thought was the most interesting parts of the data to present. Take it or leave it! ( : < )

The tests differ a little between these two tests here because I had not decided upon a test regiment for the first tests. They will match better from here on. The DP test regimen will be: 1) Cold start flare, 2) Cool oil WOT (Wide Open Throttle to about 5200 rpm with EOT~ 100F), 3) steady 55-60 mph cruise (1650-1750 rpm) at EOT=100F), 4) steady 55-60 mph cruise (1650-1750 rpm) at the highest oil temp that can be achieved reasonably (it will vary according to season), 5) Hot WOT at the highest oil temp achieved, 6) Hot idle at the highest oil temp that can be achieved.

Remember to note the the time when viewing the data. When you find the point where the filter is in bypass, take note of the short durations thereof. Also take note that the bypass is not a full-open-or-full-shut event. The valve is calibrated to crack at X-psi and be fully open at Y. So if cracking is 8 psi, full open is 16 and DP is 8.8 the valve ain't opened much. Unfortunately, only a few mfrs. presents the opening pressure in a range, eg 8-16, which reflect the cracking and full open pressures. The other Mfrs usually only list one pressure, most often the full open pressure. I will list whatever pressure I have available.


What is Differential Pressure?:

Quite simply, it's the restriction of the filter media. It increases with viscosity and oil flow, so you will see it highest with cold oil, thick oil or with high rpm/oil flow.

The Product Particulars:

You will see the results of two oil/filter combos. The first is a Pure One (P1) #PL24651 combined with 10W30 Motorcraft Super Duty HDEO oil. Yes, that is more than a grade heavier than spec'ed for the truck... don't ask, it was an experiment. The DP setup was installed about halfway thru a 15K interval , in the 8K mile range, and I took some recordings then and again later at about 12K and the last right before the OC at a bit over 15K.

The second set of Data came from a combo of an MC FL820s (generously donated by 2010FX4) and some stashed Castrol GTX SM 5W20 I had on hand. It had about a thousand miles on it for the tests.

Conclusions:

Few to make at this point. I can say that the Data from the early tests on the P1/10W30 combo were substantially the same as the last ones, so the suggestion is that filter loading had little effect on the tests. That makes sense since I have a bypass filter that is sharing the contaminant load. It's pretty easy to see the difference viscosity makes on cold starts. I won't pronounce on whether the filter efficiency (read restriction) or the oil viscosity has the most effect on DP until we see some Data on the more efficient filters with the 5W20, but my money is on the oil.



P1/10W30 Cold Start
Bypass Opening Pressure= 8-16 psi DP

This was done in March of 2013 on a 37F morning. When the engine starts, it "flares" up to about 1700 rpm to "lite off" the cats and then settles into a high cold idle, gradually inching down as the engine warms. This is where the 10W30 oil really killed things. You can see how high DP pressure got (42 psi ( : < o) for a fraction of a second. It took around 12 seconds for the DP to stabilize below the cracking pressure. Remember that this is an extreme mismatch of cold oil (34 EOT... I didn't record it on my chart) where the oil viscosity is is more than a grade thicker than spec'ed.

P1-10W30-37F-Cold Start Flare.jpg



MC/5W20
Bypass Opening Pressure: 12.5-18.5 DP
Efficiency: ~93%@20um

This is the specified grade for the engine and the specified filter. Note that this combo did not exceed the bypass cracking DP. This is a combination of a much thinner oil with better cold start viscometrics and a medium efficiency filter with very little inherent restriction.

MC-5W20,-37F-Cold Start Flare.jpg



P1/10W30 Cool WOT

This is where I took a cool engine with EOT ~ 100F and revved it to the redline. I did some with a much colder engine and saw EOP over 100 psi. After a couple of those I saw no point in subjecting my engine to more of that. Wide open at 100F is bad enough. Only a moron or someone in fear for his life revs a stone cold engine to redline and the pitfalls are not important to the driver in either case and neither case applies to me. (I hope).

Here is latest of the MC/10W30 at EOT=135F. You can see the DP only briefly rose above cracking pressure. I did another with EOT=64F and DP was 8.5-10 psi for about the same amount of time. Where is the MC/5W20 Cool WOT? Silly me! I lost that test somewhere in the bowels of my Windows laptop ( I despise XP). I will post it if and when I find it. I remember it though, and DP was well below cracking pressure the entire time. No drama.

P1-10W30-80F-Cool WOT-1.jpg



P1/10W30 Hot Wot

This was done in the summer and at about 92F ambient. I was running at 200+ EOT, so even with 30 grade oil the DP was manageable. For some reason the DP and final RPM got clipped off so I will leave DP to you and your calculator. RPM was the same as all the other WOTs, ~5200. You can see I bumped past the bypass a couple of times very briefly (a second or so). This is rare with the oil hot, even with the 10W30 but I decided to show this segment to show you that it does happen.

P1- 10W30- 92F-201=EOT-WOT-1.jpg



MC/5W20 Hot WOT

It was only about 38F ambient so it took about 20 miles to reach 170F. It might have gone higher but... In any case, no drama. The thin oil and "loose" filter combined to offer low DP. DOn't start crowing about the MC filters just yet. In later installments, I will be using very high efficiency filters so we will see whether the oil viscosity or the filter is the major contributor of these low DP readings.

MC-5W20-38F-171F=EOT-WOT.jpg



MC/5W20 Hot Cruise

Here is the MC and 5W20 in a hot cruise (1750 rpm) moment at EOT=171F. As you can see, no drama. Very low DP. I sorted similar data from the P1/10W30 Data but it comes thru here corrupted so I'll skip it for now. No drama there either. With roughly 15K on the oil and filter, in this situation DP is ~4.5 psi

MC-5W20-38F-171F=EOT-60 mph Cruise.jpg



MC/5W20 Hot Idle ~EOT=171F

No drama again. FYI, the P1/10W30 combo produced about 1+ psi more at idle.

MC-5W20-38F-171F=EOT, Idle.jpg
 
Thanks Jim for taking the time to record and lay all this out for us! Interesting indeed.
 
Thanks for the time and effort Jim! I am looking forward to the comparison between the high efficiency filters and these results.
 
Thanks Jim, that's a lot of work to go to for us.

I'm particularly impressed that the engine didn't grenade the second it went into filter bypass, given the common knowledge of how rare and devastating such an event it.
 
Originally Posted By: Shannow

I'm particularly impressed that the engine didn't grenade the second it went into filter bypass, given the common knowledge of how rare and devastating such an event it.


My sarcasm sense is tingling
wink.gif


Thank you, Jim! Good stuff to say the least.
 
YES! Real data! Mmmmmmmmm - I like it.

Thank you, Jim, for the expense and efforts; this is useful info.

As Jim would note, this test is not all encompassing; it's only one unique engine/gage set up. These results may not be relevant for every other motor/lube/filter combo out there. But ... It's WAY better than the conjecture and opinionated rhetoric we've seen to date.

Regardless of how many times the filter goes into bypass (which really is not often, and represents very small fractions of time overall relative to the event), Jim also has data to show that his engine and lube are in good shape (UOAs). Together, we can draw some inferences together which suggest that worrying over filter media and lube is a fools errand. Even with a non-spec'd lube, the bypass events were reasonably minor.

More data, please! Keep those experiments running.
 
That testing took time, effort and money to do. Thank you, Jim!

Nice to have this info for reference and I look forward to the high efficiency filter results, too. I'll also look into the Isspro gauges. Thanks!
 
Originally Posted By: Shannow
Thanks Jim, that's a lot of work to go to for us.

I'm particularly impressed that the engine didn't grenade the second it went into filter bypass, given the common knowledge of how rare and devastating such an event it.


..." he said, tongue firmly in cheek!

crackmeup2.gif


Yes, the BITOG tendency is to Hershey-squirt over the slightest element. As Dave said, and I should have, this is a unique combination that has applications mostly to owners of similar trucks. But the trends are clear. You can largely avoid bypass with the right oil viscosity in conjunction with a little care during warmup and WOT events. AND, the bypass events, if or when they occur, are measured in seconds or fractions of seconds and most often thru only a partly open bypass valve with only a small fraction of the total oil flow. Minor bypass is largely insignificant in my view.

FYI, some of the filters that are "on deck" include the Fram Ultra, Royal Purple, Bosch Distance Plus, Fram Tough Guard, another P1, Puro Classic.
 
Jim:

This is fantastic, thank you so much for doing this! What you observed on the cat-light-off cold start with the 10w-30 is exactly what myself and ZeeOSix were discussing in another thread and it appears to affirm what I had speculated (based on the Ford literature) about Ford's choice of bypass valve location on the FL-820S filter and the potential for the initial start surge to open the bypass, potentially washing any settled contaminant through the bypass and into the engine with the conventional dome-end style bypass filter.

Was this something that had popped into your mind when you saw this happening?
 
Thank you Jim. You've written this up so simply even I can understand it.
So if I'm reading that right the by-pass opened just a crack with cold oil that is much thicker at start up than ford specifies.
And using spec grade oil prevents the event happening for any significant(harmful) amount of time.
Jim
Since you're prepared to donate your time and effort into acquiring real world data I'm prepared to donate real world money to acquire oil,filters or whatever else you may need to keep the flow of data streaming in.
If you've got a PayPal account could you please inbox the addy to me so I can help out with the hard costs of data acquisition please.

Thank you for your effort. I cannot say thank you enough
 
Originally Posted By: Clevy
Thank you Jim. You've written this up so simply even I can understand it.
So if I'm reading that right the by-pass opened just a crack with cold oil that is much thicker at start up than ford specifies.
And using spec grade oil prevents the event happening for any significant(harmful) amount of time.
Jim
Since you're prepared to donate your time and effort into acquiring real world data I'm prepared to donate real world money to acquire oil,filters or whatever else you may need to keep the flow of data streaming in.
If you've got a PayPal account could you please inbox the addy to me so I can help out with the hard costs of data acquisition please.

Thank you for your effort. I cannot say thank you enough


Based on this quote:

Quote:
P1/10W30 Cold Start
Bypass Opening Pressure= 8-16 psi DP

This was done in March of 2013 on a 37F morning. When the engine starts, it "flares" up to about 1700 rpm to "lite off" the cats and then settles into a high cold idle, gradually inching down as the engine warms. This is where the 10W30 oil really killed things. You can see how high DP pressure got (42 psi ( : < o) for a fraction of a second. It took around 12 seconds for the DP to stabilize below the cracking pressure. Remember that this is an extreme mismatch of cold oil (34 EOT... I didn't record it on my chart) where the oil viscosity is is more than a grade thicker than spec'ed.


It looks like the bypass went WIDE open for a split second (42psi differential!
crazy2.gif
) and then stayed open for several seconds, gradually closing with the 10w-30 in the pan. This is displayed in the data he posted after making that statement.
 
Clevy- No need for any contributions, but thanks for the offer.

Overkill: This has been in the offing for a long, long time. I have long been curious about DP and most of that curiosity is based on the bypass and finding the means to prevent bypassing as much as possible. Secondarily, I wanted to verify the comments by several engineers over the past few years that bypass events are, "infrequent" and "momentary." I would say those are true statements. But we stall have a few filters to run thru yet.

The data I presented is more or less the worst of what I have recorded. It's strange how differently you and I focused on the 10W30 data. I saw it as evidence of oil too thick causing more bypassing than I prefer and you saw it as validation of the MC upper end bypass location. Hopefully you noted that the MC filter with 5W20 oil had narry a bypass and that the same would have been true, largely, with the P1 if I had been running 5W20.

As to the bypass location, if you recall, I argued that even if it occurs with frequency, it's still not that big of a deal. Also, Go back and look at the pics of all the autopsied filters that contained a lot of gunk, particularly those with base end bypass valves. If regular bypassing had a major washing effect that would blow the gunk thru into the clean lube oil flow, there would have been little to no gunk left to look at!
 
Originally Posted By: Jim Allen
Clevy- No need for any contributions, but thanks for the offer.

Overkill: This has been in the offing for a long, long time. I have long been curious about DP and most of that curiosity is based on the bypass and finding the means to prevent bypassing as much as possible. Secondarily, I wanted to verify the comments by several engineers over the past few years that bypass events are, "infrequent" and "momentary." I would say those are true statements. But we stall have a few filters to run thru yet.

The data I presented is more or less the worst of what I have recorded. It's strange how differently you and I focused on the 10W30 data. I saw it as evidence of oil too thick causing more bypassing than I prefer and you saw it as validation of the MC upper end bypass location. Hopefully you noted that the MC filter with 5W20 oil had narry a bypass and that the same would have been true, largely, with the P1 if I had been running 5W20.

As to the bypass location, if you recall, I argued that even if it occurs with frequency, it's still not that big of a deal. Also, Go back and look at the pics of all the autopsied filters that contained a lot of gunk, particularly those with base end bypass valves. If regular bypassing had a major washing effect that would blow the gunk thru into the clean lube oil flow, there would have been little to no gunk left to look at!


Jim:

Great thoughts
smile.gif


As per our last conversation on this, I believe we both agreed that the events weren't frequent. But there is the potential for them to occur, which I still believe validates Ford's position on the bypass location.

I'd also think that we may not have enough filter dissection data to really have any sort of firm grasp as to how much sediment washing has or hasn't occurred. I think it is a risk, that, in certain situations COULD lead to particulate ending up where it shouldn't, like in say a cam phaser
wink.gif


And allow me to extrapolate a bit on my observation and position:

Yes, your use of heavier than optimal oil certainly made the bypass event on start far more dramatic than it was with the spec'd weight of oil in the pan. However, I would argue that what you've really done here is emulated a much colder start with the spec weight. We both know viscosity correlates with temperature. Given the temperature that you witnessed the dramatic differential/bypass event with the 10w-30, one should be able to readily conclude that the same event would be reproducible at a lower temperature with the 5w-20; a temperature at which it mimics the viscosity characteristics of the 10w-30 when it yielded the differential/bypass event.

That is, your 10w-30 test essentially gave us the data from an extreme cold start with the spec oil by proxy
smile.gif


I would posit that if we were to use a 15w-40 that we we see the same event under more favourable ambient conditions.

Of course this also means that we are now reliant on the next stage of your testing
smile.gif
Finding whether it is the restriction in the media or the viscosity of the oil that triggered the bypass event. It is quite possible for it to be a combination of the two.

Either way, I think you've certainly made a case for running the spec viscosity in this application. I anxiously await what the effects a PureONE in place of the current Motorcraft are in terms of its effect on bypass frequency and magnitude.

You've done a tremendous job so far sir. Stepping up to the plate to provide data like this so that we can have these discussions is truly in-line with what the late Gary Allen was about and I'm sure he would have loved to partake in this discourse with us
cheers3.gif
 
Great work Jim, really appreciate the time and effort for this analysis.

These filters are doing exactly what they are designed to do, and are rarely seeing by-pass scenario under normal operating conditions. Hopefully this will help those filter-inclined folks to rest easier at night.
 
Jim - thanks for taking the time and effort to gather real time data and sharing the information. As your data shows, an oil filter really doesn't build up much delta-p across the media (2~4 PSI) when the oil is hot and thin - even at elevated RPM (high oil pump flow rates).

As usual, I will link the infamous thread where Purolator bench tested a PureOne for "Flow vs PSID" data. This data ties in well with your testing results.
http://www.bobistheoilguy.com/forums/ubb...451#Post1619451

The only data missing in your testing is the oil pump's output volume vs RPM, but based on your collected data, the engine in you truck is probably putting out 8~10 GPM at near max RPM.

In one of your tests (P1/10W30 Hot WOT), I saw that the filter's outlet pressure was higher than the inlet pressure, so not sure what was going on there (?). The outlet pressure then became less than the inlet pressure after a few seconds. Probably just a sensor/data glitch (?), as physically the outlet side of the filter should always be at a lower pressure than the inlet side because there is always some level of pressure drop across the filter whenever there is oil flowing through it.

The cold start and cold engine revs data is interesting, and shows that the bypass valve is way more likely to open under those conditions. I think that has been the consensus by many here, but it's always nice to see real data to prove the theory. Just goes to show that keeping the engine revs down until the oil warms up some is always a good move.
grin.gif
 
Originally Posted By: OVERKILL

Of course this also means that we are now reliant on the next stage of your testing
smile.gif
Finding whether it is the restriction in the media or the viscosity of the oil that triggered the bypass event. It is quite possible for it to be a combination of the two.


It's definitely a combination of the two, and also a 3rd factor that creates delta-p, which is the oil volume flowing through the filter. The filter's flow resistance is basically fixed, the oil viscosity and oil flow rate (due to changing the speed of the oil pump with engine revs) are obviously variables.

High engine revs with thick oil will result in the highest delta-p across the oil filter, and hence the easiest way to make the filter go into by-pass as Jim's data shows.
 
Originally Posted By: ZeeOSix
Originally Posted By: OVERKILL

Of course this also means that we are now reliant on the next stage of your testing
smile.gif
Finding whether it is the restriction in the media or the viscosity of the oil that triggered the bypass event. It is quite possible for it to be a combination of the two.


It's definitely a combination of the two, and also a 3rd factor that creates delta-p, which is the oil volume flowing through the filter. The filter's flow resistance is basically fixed, the oil viscosity and oil flow rate (due to changing the speed of the oil pump with engine revs) are obviously variables.

High engine revs with thick oil will result in the highest delta-p across the oil filter, and hence the easiest way to make the filter go into by-pass as Jim's data shows.


Certainly
thumbsup2.gif
What I was digging at was more to the overall effect being an even split between the two, both being significant contributors, rather than one taking the majority of the blame, which seems to be the theory that is currently being considered, that the 10w-30 is the biggest contributor to the event, not the PureONE.

I'm thinking specifically as to media efficiency (PureONE vs Motorcraft) with the same oil in the pan, what we are going to see as to the net effect on initial surge/bypass on cold start (as in the tests so performed, there was a big difference between the two viscosities and also two different filters).

Essentially, I'd love to see the start-up with the 5w-20/Motorcraft combo replicated with 5w-20/PureONE to see how close they are
smile.gif
Would be interesting if we find that there is a larger than expected differential with the PureONE and 5w-20. Not saying this will be the case or that I even expect it, but it is possible and that intrigues me
grin.gif
 
Originally Posted By: OVERKILL
Originally Posted By: Clevy
Thank you Jim. You've written this up so simply even I can understand it.
So if I'm reading that right the by-pass opened just a crack with cold oil that is much thicker at start up than ford specifies.
And using spec grade oil prevents the event happening for any significant(harmful) amount of time.
Jim
Since you're prepared to donate your time and effort into acquiring real world data I'm prepared to donate real world money to acquire oil,filters or whatever else you may need to keep the flow of data streaming in.
If you've got a PayPal account could you please inbox the addy to me so I can help out with the hard costs of data acquisition please.

Thank you for your effort. I cannot say thank you enough


Based on this quote:

Quote:
P1/10W30 Cold Start
Bypass Opening Pressure= 8-16 psi DP

This was done in March of 2013 on a 37F morning. When the engine starts, it "flares" up to about 1700 rpm to "lite off" the cats and then settles into a high cold idle, gradually inching down as the engine warms. This is where the 10W30 oil really killed things. You can see how high DP pressure got (42 psi ( : < o) for a fraction of a second. It took around 12 seconds for the DP to stabilize below the cracking pressure. Remember that this is an extreme mismatch of cold oil (34 EOT... I didn't record it on my chart) where the oil viscosity is is more than a grade thicker than spec'ed.


It looks like the bypass went WIDE open for a split second (42psi differential!
crazy2.gif
) and then stayed open for several seconds, gradually closing with the 10w-30 in the pan. This is displayed in the data he posted after making that statement.



Nice save. I missed the wide open part and thought it was only partially open.
Thank you Overkill. I'm my attempt to get through it quickly I obviously missed some very important info.
Thanks again.

So Jim. Are any of these tests going to address the loading of the filter media and how long an oil filter really can be used before requiring a new one.
And based on what I've read in this thread and only the grade of oil was changed yet there are so many points that need to be looked at.
I now completely understand why real world testing costs what it does and when Dnewton says that a half a million miles worth of used oil analysis using identical oil and filters is just scratching the surface when it comes to data acquisition and comparing wear metals from 2 used oil analysis using different oils is no way to accurately measure wear.
And as Doug Hillary and Overkill point out(just off the top of my head) that high wear metals are meaningless really,only teardowns and measuring the parts before and after will truly measure wear.
And I'm beginning to grasp just how much effort it takes to acquire all this data and changing a variable then doing it all again.
A person would truly need a few identical engines to test multiple variables and attain actual data since a single used oil analysis doesn't say much and hundreds of thousands of miles truly are required to get an accurate picture.
I now see the futility of comparing used oil analysis of different oils and comparing wear metals in the ppm,as if those data points really mean anything.

I do appreciate everyone's effort here. Dnewtons,Jim Allen,Shannon,Doug Hillary,overkill,garak,molekule,Jim in pa,dermapaint,mystic,and those who contribute actual data without absurd agenda or opinion.
You guys make this site worth coming to. I apologize to those I've missed
I appreciate everyone's effort.

Thank you gentlemen
 
Back
Top