LSJr tests oil filter BP function ... "OIL FILTERS: What most people DON'T know"

[ZeeOSix]

Why not put these questions as comments on the YT video to see what they say?

I get enough "technical bickering" here.

 

[Hose Pipe]

I personally enjoyed the video.
If there was cursing in it, I missed it.

Okay I am going to be brave and give it a watch..

 

[TiGeo]

I get enough "technical bickering" here.

Clearly at 43.5K post count. I think it's silly that folks here go on about these videos and the technical innaccuracies they feel are there but won't just put the questions/concerns that they have in his comment section to get it sorted out - in my experience he responds as I've put some mundane comments on several and gotten responses from him. Would be interesting to see what he says.

 

[Pablo]

I didn’t pick up on the cursing. He did keep it shorter than usual.

There are no absolutes - but the implication really is use a proper viscosity for your situation and no bypass during cold start

 

[ZeeOSix]

There is a good explanation for why the dP across the oil filter might increase as viscosity is reduced. The dP would decrease if the flow rate were kept constant, but the flow rate will increase as the viscosity decreases, as long as the oil pump is in pressure relief.

Oil filter dP is more sensitive to flow rate than it is to viscosity. Most of the dP from the filter canister and the filter head will be from turbulent flow, and this dP will increase with the square of the flow rate. The lower viscosity will tend to counteract this increase in dP from the increased flow rate, but viscosity on has a small effect on dP in turbulent flow conditions.

How much the oil filter dP changes with viscosity depends on how much the oil flow rate changes with viscosity, which is largely dependent on the oil pump PRV characteristics and will be different for every engine.

The additional dP that's measured with the warmer oil should be coming from the filter canister and the filter head. The dP across the actual filter media should not be increasing, since filter media should have both a very linear flow-dP curve and a very linear viscosity-dP curve. So, this increased dP that's measured is not relevant in terms of filter bypassing, but it is still relevant in terms of its effect on main gallery pressure.

Yes, the only way it could happen is if the flow rate going through the filter (and the whole oiling system) was increasing as the oil was getting hotter and thinner because the oil pump was slowly reducing its relief and sending more oil through the system. Yes, it mostly depends on the performance characteristics of the oil pump on that specific engine. Obviously there was no oil flow meter on the setup, only an upstream and down stream pressure sensor, so the test was blind to flow rate data.

Wish they would have did a pull to full RPM when the oil was at room temperature, but they probably didn't want to possibly hurt the engine. In a run like that, the filter would most likely go into bypass before the pump had enough decrease in relief to cut the flow back enough to prevent filter bypass. A case like that can happen at the drag strip races when the racer's cars sit in the pits before they go to the line and do a big burn out at redline, then hammer it WOT down the 1/4 mile. Racers also like to do the 1/4 mile with a relatively cold engine because they don't want the intake too hot. But they most likely don't realize that the oil filter is bypassing during those cold oil redline episodes.

 

[Propflux01]

Wonder if there is a neutered version out there

 

[ZeeOSix]

The graph they show at time 10:10 doesn't represent the oil filter going into bypass because if the dP is large enough to open the bypass valve then the dP across the filter will remain, and even might increase from that point if the oil flow is increasing. The two curves are not going to converge and become essentially the same pressure as that graph shows. Don't think that graph is from any data they took, as it has flow rate on the x-axis.

 

[TimeFlies]

Seems like Ned Flanders is the oil guy...

Ned probably says that...

 

[bullwinkle]

I think the reason the dP increases with RPM is because the main & rod bearings actually “pull” or pump oil themselves with increasing RPM, and the pump hits bypass, so the dP increases somewhat. Unless I’m misremembering an old BITOG lesson somewhere?

 

[ZeeOSix]

I think the reason the dP increases with RPM is because the main & rod bearings actually “pull” or pump oil themselves with increasing RPM, and the pump hits bypass, so the dP increases somewhat. Unless I’m misremembering an old BITOG lesson somewhere?

The oil pump puts more oil volume through the filter and on to the engine oiling system as RPM increases. On an old fashioned PD oil pump with a spring loaded pressure relief valve like used on that engine, when the pump is in pressure relief, the oil volume output can still increase with more RPM. The "self pumping" aspect of the journal bearings would actually help slightly reduce the pressure on the inlet of the system. Think of the self-pumping journal bearings like miniature scavenging pumps on the system outlet.

 

[dnewton3]

Interesting ... 12 deleted posts in three pages.
Some of you seem WAY more interested in discussing the topic of profanity rather than the oil filter bypass.
Fine by me; we'll just lock this.

I have reopened the thread.
Let's see if we can stay on topic and keep the sidebar profanity talk out of this, shall we?

 

[Faawrenchbndr]

The graph they show at time 10:10 doesn't represent the oil filter going into bypass because if the dP is large enough to open the bypass valve then the dP across the filter will remain, and even might increase from that point if the oil flow is increasing. The two curves are not going to converge and become essentially the same pressure as that graph shows. Don't think that graph is from any data they took, as it has flow rate on the x-axis.

Did it ever go into bypass? If it did I missed it.

 

[ZeeOSix]

Did it ever go into bypass? If it did I missed it.

It wasn't seen in the data from their test run data. The graph shown at time 10:10 wasn't their test data, but used to try and say that what it would look like if the filter went into bypass, which is a wrong assessment. If they would have done some dyno pulls with much colder oil the filter bypass would most likely have opened.

 

[Faawrenchbndr]

It wasn't seen in the data from their test run data. The graph shown at time 10:10 wasn't their test data, but used to try and say that what it would look like if the filter went into bypass, which is a wrong assessment. If they would have done some dyno pulls with much colder oil the filter bypass would most likely have opened.

Got it……thanks that’s what I thought

 

[johnmyster]

It just showed that if the oil is warm enough on an engine with a pretty low volume oil pump that the dP across the filter won't be enough to cause it to bypass. Do that same max RPM dyno pull with the oil at 20F and it would go into bypass for sure. They don't really touch on that much, and leave the audience thinking (and ready to go start spreading misinformation, lol) that oil filters never go into bypass unless they are nearly totally clogged ... which isn't a true statement. If the oil is cold enough, the bypass valve setting low enough and the engine RPM high enough the filter will bypass.

A plausible explanation would be an argument that you and I had a year ago about pump slip. If one oversimplifies their understanding of an oil pump as absolutely 100% positive displacement (or that pump slip is trivial), then yes, it makes no sense. However, if one accepts that the specific flow rate of the pump decreases as it pumps against greater pressure, then it all falls into place. Pump slip being non trivial supports the observation. But if there's one think we can count on on Bitog, it's that Z06 will argue flow is proportionate to RPM with no other contributing factors worth considering.

Or I guess we could blame pump bypass instead.

In any case, the popular bitoger hypothesis about filter bypass flushing debris through the engine during startup seems to be un supported by this one dyno example. I guess if you hit redline at the apex of your driveway on a zero degree morning, perhaps.

 

[ZeeOSix]

A plausible explanation would be an argument that you and I had a year ago about pump slip. If one oversimplifies their understanding of an oil pump as absolutely 100% positive displacement (or that pump slip is trivial), then yes, it makes no sense. However, if one accepts that the specific flow rate of the pump decreases as it pumps against greater pressure, then it all falls into place. Pump slip being non trivial supports the observation.

In this case, there isn't enough pump slip to cause what was seen in the video. Only time pump slip really matters is when the pump is basically worn out and junk.

But if there's one think we can count on on Bitog, it's that Z06 will argue flow is proportionate to RPM with no other contributing factors worth considering.

Not true ... you obviously have not been reading all the threads where PM pumps have been discussed. We've had the pump slip discussions, and unless the pump is very worn out there is little pump slip in a gerotor style oil pump. And pump slip actually goes down as the oil warms up and thins down. I even posted pressure vs RPM and flow volume vs RPM of LS Melling oil pumps showing they start going into pressure relief around 3000 RPM, and explain how a simple spring loaded pressure relief valve on a PD oil pump will still allow the pressure and volume to keep increasing with increase RPM. All of these these factors involved with PD oil pumps have been discussed in various threads.

Or I guess we could blame pump bypass instead.

So you think the filter didn't go into bypass because of pump slip? The filter didn't go into bypass in that video because there simply wasn't enough dP across the media to make the bypass valve open. As already mentioned, if they would have did a run to redline even with room temperature oil (instead of letting is fully warm-up instead), the filter would have most likely gone into bypass at the higher RPM of the run, even with the pump in pressure relief.

In any case, the popular bitoger hypothesis about filter bypass flushing debris through the engine during startup seems to be un supported by this one dyno example. I guess if you hit redline at the apex of your driveway on a zero degree morning, perhaps.

It certainly could if the start-up was with cold enough oil and the engine RPM high enough. As mentioned in this thread, the on-vehicle filter bypass testing done by Jim Allen showed that happens often in cold weather climates. What was done in the JSJr video tells you nothing on how the filter would behave in other conditions with colder and thicker oil.

 

[johnmyster]

In this case, there isn't enough pump slip to cause what was seen in the video. Only time pump slip really matters is when the pump is basically worn out and junk.


Not true ... you obviously have not been reading all the threads where PM pumps have been discussed. We've had the pump slip discussions, and unless the pump is very worn out there is little pump slip in a gerotor style oil pump. And pump slip actually goes down as the oil warms up and thins down. I even posted pressure vs RPM and flow volume vs RPM of LS Melling oil pumps showing they start going into pressure relief around 3000 RPM, and explain how a simple spring loaded pressure relief valve on a PD oil pump will still allow the pressure and volume to keep increasing with increase RPM. All of these these factors involved with PD oil pumps have been discussed in various threads.


So you think the filter didn't go into bypass because of pump slip? The filter didn't go into bypass in that video because there simply wasn't enough dP across the media to make the bypass valve open. As already mentioned, if they would have did a run to redline even with room temperature oil (instead of letting is fully warm-up instead), the filter would have most likely gone into bypass at the higher RPM of the run, even with the pump in pressure relief.


It certainly could if the start-up was with cold enough oil and the engine RPM high enough. As mentioned in this thread, the on-vehicle filter bypass testing done by Jim Allen showed that happens often in cold weather climates. What was done in the JSJr video tells you nothing on how the filter would behave in other conditions with colder and thicker oil.

I don’t do hot laps of the neighborhood straight off a cold start, so those “conditions” are not concerning to me.

You’ve once again run your theory into a self supporting circle. Pump slip is the reason why the drop in pressure across the filter seems to indicate higher flow at lower galley presssure (higher temp, lower viscosity.) It’s large enough of an effect to overcome the higher slip rates caused by the lowering of viscosity. You can’t write that off by saying it only applies to worn out pumps. New pumps have clearances too.

 

[ZeeOSix]

You’ve once again run your theory into a self supporting circle. Pump slip is the reason why the drop in pressure across the filter seems to indicate higher flow at lower galley presssure (higher temp, lower viscosity.) It’s large enough of an effect to overcome the higher slip rates caused by the lowering of viscosity.

The dP across the oil filter is directly correlated to the oil pressure in the gallery where the oil pressure sensor is. At a constant oil temp, if the oil pressure sensor shows higher pressure, that means there's more flow rate and it also means there is more dP across the oil filter - regardless of what the oil pump is doing. Can't have one without the other. A higher flow rate with thinner oil can result in less dP across the filter. The pump was in some level of relief as the RPM climbed higher even with hot oil because that's how those PD pumps operate, and the oil was hot enough to not produce enough dP across the filter for it to hit bypass. Do the dyno run with colder oil and the filter would most likely hit bypass even with the pump in relief because spring loaded pressure relief valve can't control the pump output pressure and volume to a perfect constant level with increasing RPM. Most engines aren't going to cause an oil filter to hit bypass dP even with the engine at redline just like seen in the LSJr video (and it's not mainly due to "pump slip") ... except those crazy high volume oil pumped Subarus, lol). Do the high RPM with thick cold oil and you can bet the filter will hit bypass, just as Jim Allen's on-vehicle testing showed.

You can’t write that off by saying it only applies to worn out pumps. New pumps have clearances too.

How much pump slip do you think a healthy gerotor style PD pump in an engine has. I've never claimed they have zero slip. In the discussions in the past, it was pointed out they would have around 10-15%, which isn't enough to cause them to fall on their face at the oil pressures seen in an engine. You do realize that if you disabled the pump's pressure relief valve it would literally build so much pressure (like 300+ PSI) that it would blow something up in the oiling system - whatever the weak link is would blow. Try it on your car if you don't believe it. If a gerotor style oil pump had as much pump slip as you think, it wouldn't have the ability to do that. I don't think you really understand PD oil pumps if you think pump slip was the main reason why the oil filter in the LSJr video didn't hit bypass dP across the filter.

 

[johnmyster]

The dP across the oil filter is directly correlated to the oil pressure in the gallery where the oil pressure sensor is.

The dp across the filter is more directly a function of flow, viscosity, and filter loading. I guess you could indirectly characterize the first two using gallery pressure if you had a complete model of oil consumption (volume flow rate) by the engine as a function of viscosity, rpm, temperature, etc. But to state it this way is an oversimplification that undermines any following statements.

At a constant oil temp, if the oil pressure sensor shows higher pressure, that means there's more flow rate and it also means there is more dP across the oil filter - regardless of what the oil pump is doing. Can't have one without the other.

But this ignores the actual hypothesis proposed by LSJ - that holding other things constant, colder oil has less filter dp because overall flow rates in the system (which has not reached pump bypass) are lower despite higher gallery pressures. The engine "consumes" less (volume flow rate) cold oil at high pressure than it does hot oil at low pressure. They observed this on 1800 rpm data - not the full RPM sweeps.

The pump was in some level of relief as the RPM climbed higher even with hot oil because that's how those PD pumps operate, and the oil was hot enough to not produce enough dP across the filter for it to hit bypass.

Yes, pump slip looks alot like pump bypass if you'd rather ignore pump slip when it's contrary to your position.

because spring loaded pressure relief valve can't control the pump output pressure and volume to a perfect constant level with increasing RPM

Correct, nozzles/orifices are not isobaric. There's a nonlinear flow vs pressure relationship that also depends on viscosity.

Most engines aren't going to cause an oil filter to hit bypass dP even with the engine at redline just like seen in the LSJr video (and it's not mainly due to "pump slip")

Again, the hypothesis at hand isn't about filter bypass opening. It's about why the observed cold dp across the element is lower than hot.

How much pump slip do you think a healthy gerotor style PD pump in an engine has. I've never claimed they have zero slip. In the discussions in the past, it was pointed out they would have around 10-15%, which isn't enough to cause them to fall on their face at the oil pressures seen in an engine.

Well, it depends on viscosity, pump speed, and back pressure. PD pumps of interest are more "leaky" at low RPMs, with thin fluid, and with high back pressure. In past discussions we also found data to say none of these relationships are linear. So at one extreme or the other, different factors may dominate. To peg a figure as 10-15% (make it a constant) is misleading.

You do realize that if you disabled the pump's pressure relief valve it would literally build so much pressure (like 300+ PSI) that it would blow something up in the oiling system - whatever the weak link is would blow. Try it on your car if you don't believe it.

I do. My duramax even at 40f would idle at 80 psi (pump bypass) until it began to warm the 15w-40. Hot it was magically only 17 psi. So a difference of 5x...

If a gerotor style oil pump had as much pump slip as you think, it wouldn't have the ability to do that. I don't think you really understand PD oil pumps if you think pump slip was the main reason why the oil filter in the LSJr video didn't hit bypass dP across the filter.

It would if the engine itself was consuming far less of the thicker oil. 5w-30 is only what, two orders of magnitude (100x) different across the range of temperatures experienced on my morning commute. Certainly 100x viscosity change makes as much of a difference to flow as 5x pressure change, if not more.

if you think pump slip was the main reason why the oil filter in the LSJr video didn't hit bypass dP across the filter.

I never said that. He didn't see high enough pressures across the element to trigger filter bypass because flow, viscosity, and filter loading were not high enough to get there. Again, wrong hypothesis.

 

[ZeeOSix]

The dp across the filter is more directly a function of flow, viscosity, and filter loading. I guess you could indirectly characterize the first two using gallery pressure if you had a complete model of oil consumption (volume flow rate) by the engine as a function of viscosity, rpm, temperature, etc. But to state it this way is an oversimplification that undermines any following statements.

Of course the dP across the oil filter is a function of flow, viscosity and the flow resistance of the filter ... been saying that for 10+ years here. And it's completely true that that with a constant viscosity, if the pressure at the location of the pressure sensor in a main gallery feeding the system goes up, so does the flow rate at that point, and also at every point in the oiling system, including the flow through the filter. It's that simple without any simplification. It's a key basic understanding of an engine oiling system, and doesn't "undermine" anything. The dP setup on the engine in the LSJr video shows the two pressure points they were using, and the slight dP due to flow between those to points in the gallery doesn't matter, it's negligible in measuring the dP across the filter.

But this ignores the actual hypothesis proposed by LSJ - that holding other things constant, colder oil has less filter dp because overall flow rates in the system (which has not reached pump bypass) are lower despite higher gallery pressures. The engine "consumes" less (volume flow rate) cold oil at high pressure than it does hot oil at low pressure. They observed this on 1800 rpm data - not the full RPM sweeps.

The dP across the filter with room temperature oil at 1700 RPM (time 6:50 in the video, if that's what you're referring to) was only 0.5-0.6 PSI because the flow volume was low - no surprise because filters are not very flow restrictive. They never revved the engine above 1700 RPM with room temperature ("cold") oil. They should have.

The oiling system basically behaves like a fixed flow resistance that the pump is forcing oil through. As mention in the other debates, there is a small factor due to the self-pumping action of the bearings on the system which can slightly decrease the source pressure - ie, think of the journal bearings as mini scavenging pumps on the oil source. That factor complicates the discussion, but it can be ignored due to the small impact on the oiling system's behavior.

When they claim at time 5:25 in the video that the only time a filter bypass valve can open, even with cold oil, is when the filter is plugged up with debris that should make you think twice if they really know what they are talking about and how they interpret their testing later in the video. If the engine was started cold and never revved up, that might be true but with some RPM with cold oil the filter is most likely going to hit bypass dP. Again, this was clearly showed by Jim Allen's on-vehicle filter dP testing at different oil temps and engine RPM.

Yes, pump slip looks alot like pump bypass if you'd rather ignore pump slip when it's contrary to your position.

I'm saying pump slip doesn't really matter in this case because the pump will still put out way more flow than the engine can flow before the pump has to go into pressure relief to prevent oiling system damage.

Again, the hypothesis at hand isn't about filter bypass opening. It's about why the observed cold dp across the element is lower than hot.

Because they never revved the engine over 1700 RPM, and the "cold" oil was at room temperature (call it 68-70F) which really isn't that cold. If they would have did a full RPM sweep with room temperature oil you would have seen a whole different result than doing it with 200+ F oil.

Well, it depends on viscosity, pump speed, and back pressure. PD pumps of interest are more "leaky" at low RPMs, with thin fluid, and with high back pressure. In past discussions we also found data to say none of these relationships are linear. So at one extreme or the other, different factors may dominate. To peg a figure as 10-15% (make it a constant) is misleading.

Well, the Melling LS pumps have ~15% slip with hot 5W-30 oil. That can be determined by looking at the flow vs RPM curve of the pump and compare the ideal pump volume output vs the actual volume output. I'd say that pretty much all gerotor style PD pumps used on ICE that are in good shape will have ~15% slip. But in reality, the slip doesn't really matter much because the pumps put out enough volume per revolution that they have to go into pressure relief pretty early depending on the oil viscosity and RPM. Look at these pump performance curves and you can see exactly where the pressure relief valve starts to open. Even though there is 15% pump slip, the flow curve is still linear before the pump starts pressure relief. All the pump slip does is lower the flow vs RPM curve, and if the pump didn't go into pressure relief that linear flow output curve would still climb at that slope. If the pump output wasn't regulated, it would literally destroy something from over pressure ... because that's what PD pumps do, even if there's some pump slip.

I do. My duramax even at 40f would idle at 80 psi (pump bypass) until it began to warm the 15w-40. Hot it was magically only 17 psi. So a difference of 5x...

See graphs above ... of course the oil pressure is lower with hot oil. Hopefully it's 17 PSI at idle and not at elevated RPM, lol. My LS6 had 30 PSI at idle with 5W-30 at 200F.

It would if the engine itself was consuming far less of the thicker oil. 5w-30 is only what, two orders of magnitude (100x) different across the range of temperatures experienced on my morning commute. Certainly 100x viscosity change makes as much of a difference to flow as 5x pressure change, if not more.

The engine can only flow (you call it "consume") the volume of oil that the fixed flow resistance will allow the PD pump to force through it. If that supply force from the pump (the oil pressure) becomes more than the pump relief valve setting, then the pump will start going into relief. But a spring loaded pressure relief valve will still allow the pump output pressure and therefor the output flow to increase some with increased RPM. See the Melling graphs above. Pump slip doesn't matter when the flow vs RPM curve is on a steep increasing slope before the pressure relief starts opening, and even after the pump goes into pressure relief. Pump slip is invisible to the system when the pump is in pressure relief.

I never said that. He didn't see high enough pressures across the element to trigger filter bypass because flow, viscosity, and filter loading were not high enough to get there. Again, wrong hypothesis.

And that's exactly what I already said in post 31. They needed to do some full RPM runs with cold oil. It was an incomplete test effort.