Hengst oil filter inflow holes changed, should I use this new filter??

[ZeeOSix]

To be clear, I'm calling you on an upfront blanket statement that you made.

I think we've established, it does matter, even if not an order of magnitude, or twofold, etc. Why not say something like, "it will matter little" or "the difference will be negligable" rather than standing beside a blanket and misleading statement about oft misunderstood PD pumps. Centrifugal pumps and gear pumps doing work on incompressible fluids both have curves that are...curves.

I further explained my viewpoint to clarify my statements as this discussion unfolded, but seems you're not quite following or grasping the details. This whole side discussion came up when I pointed out that a few more PSI of dP across an oil filter due to fewer or smaller base plate and/or center tube holes (within reason - very closed off louvers is another side discussion) won't really matter to a PD pump. I stand by that statement. Your definition of "matter" and my definition seem to be different. Tell me when it would "matter" to you. The PD pump is still going to supply the same basic oil flow volume per RPM when the total back pressure on the pump has only increased by 1-2 PSI. Do you really think that adding 1-2 PSI of dP to the oiling system is going to make any real difference to a healthy PD pump? Is it going to suddenly starve the engine of oil flow and blow-up the engine? Do you think you could actually measure the difference in total flow to the oiling system due to adding 1-2 PSI of dP accurately with instrumentation? I say no you couldn't because it's a very small impact on oil volume from the pump due to the difference in slip caused by 1-2 more PSI of back pressure on the pump (all other factors held constant of course). That is my "the difference will be negligible" clarification, and I already tried to make that clear before this post as the discussion progressed.

We have established that pump slip is a function of backpressure, which is additive for all components in the system, including the filter. The pump curve graph I shared shows this. Backpressure and viscosity determine slip. So yes, your curve shows slip increasing as pressure increases, but your data log graph is not one of pump flow vs engine rpm. It is pressure vs rpm, which is a related but different. To call the two the same would be making a big assumption about the fluid dynamic behavior of every oiling passage/component in the engine at various flow/pressure states. I don't remember many linear relationships in fluid dynamics.

Again, look at what the difference is to the overall dP of the entire oiling system and to the back pressrue on the pump when different oil filters are used. It's only a few PSI of dP difference. I ran 4 different oil filter brands on my Z06 and collected the same oil pressure vs RPM data with oil temperature at the same 200F, and there was no difference seen in the curve that I showed in post #17. I'm sure there must have been at least 1-2 PSI of dP difference across the filter between them, but I saw zero impact on the oil pump's performance. If any of those oil filters effected the pump's performance in anymore than just a "miniscule" or "negligible" manner, I would have seen it with the tests. That engine RPM vs oil pressure curve would have changed, but it didn't.

Please show me with any oil pump performance curve for a typical automotive oil pump, what the output volume flow difference would be with only 1-2 PSI more or less of back pressure on the pump output.

You recognized it. Then you tried to further proove your point and got it backwards. Pump slip is present at all RPM. It increases with pressure but becomes smaller relative the pump output at higher flow (RPM.) PD pumps are used across the world at constant RPM but variable output flow and/or pressure.

I said pump slip increases as the back pressure on the outlet of the pump increases, and also as the oil viscosity decreases. Yes, pump slip is present at all RPM, but the total volume that "slips past the pump" is less at lower RPM than at higher RPM because the back pressure on the pump increases with RPM. I've said nothing contrary to that. If you want to make the statement that "the % of slip volume relative to the total pump output volume is smaller at higher PRM", then knock yourself out. But the fact remains that there is more oil volume slipping past the pump as the RPM increases and the oil back pressure on the pump outlet increases. I'm just looking at the volume slipping by, and not comparing that volume "relative" to the overall output volume of the pump. That's the difference going on here.

Again, pump slip is not a function of RPM. It is a function of backpressure.

See my response above. Sure, but again as the RPM increases so does the back pressure, which means slip does increase with RPM indirectly - so it certainly is a function of RPM. My statement is still true that the pump slip increases with RPM, regardless of what the root cause is.

In your experiment, backpressure and RPM are not held independent. Slip increases at higher pressure. RPM was the innocent bystander, a variable your test setup did not allow to be held constant.

Yeah, slip increases with higher RPM, which causes higher oil pressure in an engine oiling system ... I've never claimed anyting otherwise. Each one of the data points in my experiment was at a constant engine RPM and at a constant oil temperature. How do you think I could plot data of engine RPM vs oil pressure without getting those measurments at the data points I've shown? Oil temperature was also constant. The graph is pretty self explanitory.

Can you please explain to me what the graph would look like if the oil pump had zero slip?

 

[johnmyster]

I follow the details. The side discussion originated because it was said that pressure head doesn't impact flow on PD pumps. I showed one typical industry pump chart that disproved the statement.

Can you please explain to me what the graph would look like if the oil pump had zero slip?

View attachment 205822

Your graph is of pressure vs rpm for a system: pump (supply) and engine (sink). You can't decouple the pumping flow rate (pump), slip flow rate (pump), and consumption flow rate (engine.) They are all pressure dependent.

That's not a pump curve. It's a pressure curve for a system. The curve only indirectly represents flow and the relationship need not be linear. Pump slip cannot be quantified from such a curve.

Here's the first graph I found via google of flow vs pressure for oil flowing through a single jet. It's not linear. Further increases in pressure achieve incrementally less flow through an orifice. If we take that lesson back to your graph we can't really learn anything from it. Three nonlinear terms in the equation and all we see is the sum.

But the point stands. Displacement of PD pumps is not independent of backpressure, and one can't deem the difference negligible without quantifying it. In this case we can't really quantify it. You're saying 1-2 psi is small enough to equal "no change." I'm saying it's more likely "negligible change" or "a change that doesn't matter in this application" but enough that we don't call PD pumps constant flow because words matter.

 

[johnmyster]

I ran 4 different oil filter brands on my Z06 and collected the same oil pressure vs RPM data with oil temperature at the same 200F, and there was no difference seen in the curve that I showed in post #17.

Also, correct me if I'm wrong, but the oil pressure sensor is usually on the oil passages after the filter. Is this true on your test stand? If so, that's another way you've comingled some assumptions about flow and pressure. You're measuring the output pressure of the filter and saying the system flow hasn't changed because...that's how positive displacement works.

 

[ZeeOSix]

I follow the details. The side discussion originated because it was said that pressure head doesn't impact flow on PD pumps. I showed one typical industry pump chart that disproved the statement.

No, you're obviously not following the details, because I've clearly said more more than once that back pressure on the pump does effect slip ... BUT in the case of adding a 1-2 more PSI of dP onto the back pressure it's not going to make any real difference - that's the part you seem to fail to not understand and gloss over. It's the context of, and the key to this whole conversation. Yet, you keep ignoring that context.

Again, tell me just how much more slip and how much oil pump volume reduction will occur when the output flow of a healthy automotive PD oil pump is against 75 PSI vs 77 PSI of oiling system back pressure (ie, using a filter with 2 more PSI of dP at high flow). It's going to be so small that it will not matter to the oil pump. That's the point I've always been making. I have never claimed that there is zero pump slip ... I'm claiming in the context of this discussion that it doesn't matter to the PD pump. You have missed that point ever since post #6.

Tell me how it actually matters in the context of this discussion. Exactly how is adding 1-2 PSI of dP of pump back pressure going to really matter. That is the focus and context of this discussion. We have already established that pump slip happens, and tell me how much of an impact on pump slit this added 1-2 PSI will cause. You think it's going to make or break the level of lubrication to the engine. It won't ... that's why it doesn't really matter.

Your graph is of pressure vs rpm for a system: pump (supply) and engine (sink). You can't decouple the pumping flow rate (pump), slip flow rate (pump), and consumption flow rate (engine.) They are all pressure dependent.

That's not a pump curve. It's a pressure curve for a system. The curve only indirectly represents flow and the relationship need not be linear. Pump slip cannot be quantified from such a curve.

It doesn't matter if it's a pump curve or not. The graph of my engine RPM vs oil pressure still represents the performance of the oil pump - including any pump slip that may be going on. Tell me how it doesn't. And as I asked you before, please tell me what that curve would look like if there was zero pump slip going on in that case. Better yet, what do you think that graph would look like if I removed the oil filter and put a different filter on the engine that added 2 PSI more of back pressure to the pump? You wouldn't even be able to see the difference.

The engine oiling system is essentially a fixed flow resistor comprised of various components. As you force more and more oil volume through it, the oil pressure does indeed correlate to the oil flow volume. If the oil temperature & viscosity is constant, and the oil flow volume at the oil pressure sensor is exactly the same regardless of what the pump is doing, then the oil pressure will be the same at that location in the oiling system. In other words, in that scenario the oil pressure directly correlates to the oil flow at that location. This is the key to understanding why the graph data is a measurement of the oil pump performance (include the pump slip factor) while supplying oil to a fixed system.

If a very restrictive filter was installed (say it added 15 PSI more back pressure) and that caused a noticeable increase in pump slip volume loss (and the pump wasn't in pressure relief of course), then the flow at the pressure sensor after the filter would show a slight decrease because the flow was decreased from the added pump slip. That's how the pressure and flow are connected when looking at the same location in the oiling system (the oil pressure sensor in this case) when the viscosity is held constant.

The only variable in my test was the engine & pump RPM, the oil temperature and viscosity were consant. The graph of my test data would obviously include any pump slip. Again, what do you think that graph would look like if there was zero pump slip? I've asked you this previously. Do you have any idea what it would look like?

Here's the first graph I found via google of flow vs pressure for oil flowing through a single jet. It's not linear. Further increases in pressure achieve incrementally less flow through an orifice. If we take that lesson back to your graph we can't really learn anything from it. Three nonlinear terms in the equation and all we see is the sum.

So are you claiming that the roll-over in my RPM vs oil pressure graph is due to the flow behavior like this orifice example? The oiling system as a whole is not really like a single orifice - it's a mix of all kinds of different flow regimes. If this orifice behaviour is your claim, then how much of the roll over in my graph is due to pump slip and how much is due to this kind of flow behavior example? If there is some flow loss (and also a corresponding pressure loss) due to this behavior, then the pump slip is even less if it's only part of the reason for pressure roll-over as RPM increases. And yes, the pressure rolling over in my graph does indicate that the flow volume per RPM rate has changed becasue it's no longer on a near linear track above the 2000-2500 RPM point.

But the point stands. Displacement of PD pumps is not independent of backpressure, and one can't deem the difference negligible without quantifying it. In this case we can't really quantify it. You're saying 1-2 psi is small enough to equal "no change." I'm saying it's more likely "negligible change" or "a change that doesn't matter in this application" but enough that we don't call PD pumps constant flow because words matter.

I never claimed there was "no change" and there is never any pump slip - that's something you've fabricated to keep arguing about it all. I said in post #6: "The filter with a little less flow area in the base plate inlet holes might produce 1-2 PSI more dP at high RPM oil flow, but that will not matter to the positive displacement oil pump." This statement seemed to have triggered you for some reason. I never claimed there as "no change" ... I said it won't matter, because the change is miniscule as expalned later due to your triggering. I qualified that it doesn't matter in the context of adding 1-2 PSI of back pressure to the pump due to a slightly more flow restrictive filter. Yet, you ingnore the qualification.

Adding 1-2 PSi of more back pressure to the pump is a nothing burger ... it's "negligible" ... "doesn't really matter to the pump" ... "a minuscule difference" ... etc, etc.

Tell me how much reduction in pump volume will result by adding 1-2 more PSI of back pressure to the pump output if you think it's so large that it would actually matter. But your very statement of: "I'm saying it's more likely "negligible change" or "a change that doesn't matter in this application" but enough that we don't call PD pumps constant flow because words matter." basically now agrees with what I've been saying all along. So you're just going in circles now, for the sake of just arguing and trolling it seems. I never claimed that PD pumps are "constant flow". Show me exactly where I made that claim.

 

[ZeeOSix]

Also, correct me if I'm wrong, but the oil pressure sensor is usually on the oil passages after the filter. Is this true on your test stand? If so, that's another way you've comingled some assumptions about flow and pressure. You're measuring the output pressure of the filter and saying the system flow hasn't changed because...that's how positive displacement works.

I covered that in my previous post above this one. You don't seem to understand the direct relationship between flow and pressure at the same location when the viscosity and the flow resistance of the system is constant. With constant viscosity, there is a direct correlation between the oil pressure and flow at the location of the oil pressure sensor.

You will not see the effect of the oil filter's flow resistance (dP) when the oil pressure sensor is after the filter, unless the pump is in pressure relief or the difference in filter dP is huge and caused lots of pump slip - what we've been hashing over here. But the oil pressure sensor located after the filter will give you the pressure vs flow correlation, without the effect of the filter, unless the filter adds so much more resistance that the pump exhibits way more slip and/or the pump hits pressure relief. That's what half of this debate has been about. This is also the reason that my RPM vs oil pressure test is a valid measurement of the oil pump's performance. Like I said before, with 4 different oil filters on the engine (the only change in the testing) I saw exactly the same RPM vs oil pressure curve - so there was no added pump slip caused by different oil filters that had slight differences in total dP vs flow, and the resulting back pressure on the pump. The only time I would have seen a different curve is if the filter was so restrictive that it cause way more pump slip, and/or the pump output pressure was high enough to make the pump hit pressure relief. The pump was not in pressure relief in my testing.

 

[johnmyster]

I am following the details. You make assumptions that support your augment and then type pages of contradictory words to wash away the details that don't align. Verbatim quote:

Do some study on positive displacement oil pumps used in engine oiling systems. The only time any oil flow is reduced in the oiling system is when the PD oil pump is in pressure relief, which is a pretty hard condition to achieve unless you rev the engine real high with cold thick oil. When the oil is at full operating temperature, it's really hard to get the oil pump to go into pressure relief, even near or at engine redline.

Only is an absolute word. It means exclusively. You said only and then named a single condition. Out of the gate I showed a set of pump curves for a PD pump that shows flow is reduced as backpressure is increased. So it's not only. Holding pump speed and viscosity constant, pressure and pressure relief (bypass) are two flow reduction contributions.

The engine oiling system is essentially a fixed flow resistor comprised of various components. As you force more and more oil volume through it, the oil pressure does indeed correlate to the oil flow volume. If the oil temperature & viscosity is constant, and the oil flow volume at the oil pressure sensor is exactly the same regardless of what the pump is doing, then the oil pressure will be the same at that location in the oiling system. In other words, in that scenario the oil pressure directly correlates to the oil flow at that location. This is the key to understanding why the graph data is a measurement of the oil pump performance (include the pump slip factor) while supplying oil to a fixed system.

...

So are you claiming that the roll-over in my RPM vs oil pressure graph is due to the flow behavior like this orifice example? The oiling system as a whole is not really like a single orifice - it's a mix of all kinds of different flow regimes. If this orifice behaviour is your claim, then how much of the roll over in my graph is due to pump slip and how much is due to this kind of flow behavior example? If there is some flow loss (and also a corresponding pressure loss) due to this behavior, then the pump slip is even less if it's only part of the reason for pressure roll-over as RPM increases. And yes, the pressure rolling over in my graph does indicate that the flow volume per RPM rate has changed becasue it's no longer on a near linear track above the 2000-2500 RPM point.

Here you've confused a few things. The curve I showed was for a single orifice. It's nonlinear. That's the only point I needed from that graph. Which contradicts your resistor theory.

First, resistors are only linear as long as they aren't heated relative to the power they absorb, which is always. Linearity is only an handy approximation made within limits.

So first you said the oiling system was a [singular] fixed resistance value...linear...like ohms law. Then later you said couldn't do that because the system is an additive system made of many components and can't be approximated by a single value. So you contradicted yourself to make an argument. If either is true then the other half of your approximative argument falls apart.

If you take electrical engineering 101 (circuits) you'll see equations where banks of resistors can be summed into one single value. If you take mechanical engineering xxx (fluid dynamics) you'll see equations that do the same, even for nonlinear components of a system. Take a look at Bernoulli's equation. There's a nonlinear term in there. (Fluid has to speed up to move through an orifice.) Summing a bunch of nonlinear components is harder but it can be done. That said, you can't take the sum and then extract back the components. You can't weigh the bag of apples and then tell me how much each apple contributes. You showed a nonlinear curve and then said it was nonlinear because of pump slip. I argue that it is also nonlinear because the oiling system moves oil (flow rate) in a nonlinear way vs pressure. (You said it too.) So you can't then go on to say one drives the shape of the graph. They both do.

Flow can't be inferred but not quantitatively extracted from the pressure vs rpm data you harvested from your engine, because it balls together multiple nonlinear relationships into one nonlinear output. You've make my argument even if you can't stomach to admit it.

I never claimed there was "no change" and there is never any pump slip - that's something you've fabricated to keep arguing about it all. I said in post #6: "The filter with a little less flow area in the base plate inlet holes might produce 1-2 PSI more dP at high RPM oil flow, but that will not matter to the positive displacement oil pump." This statement seemed to have triggered you for some reason. I never claimed there as "no change" ... I said it won't matter, because the change is miniscule as expalned later due to your triggering. I qualified that it doesn't matter in the context of adding 1-2 PSI of back pressure to the pump due to a slightly more flow restrictive filter. Yet, you ingnore the qualification.

Because you didn't add the qualification until you were called on it.

I never claimed that PD pumps are "constant flow". Show me exactly where I made that claim.

Okay.

Do some study on positive displacement oil pumps used in engine oiling systems. The only time any oil flow is reduced in the oiling system is when the PD oil pump is in pressure relief, which is a pretty hard condition to achieve unless you rev the engine real high with cold thick oil. When the oil is at full operating temperature, it's really hard to get the oil pump to go into pressure relief, even near or at engine redline.

That's what you said. I didn't have to say it. Flow not being reduced means constant.

A key to technical writing is to say less and stick to first principles and applicable basis behaviors. The more you try to bring in other information as "examples" the more you run the risk of confusing the topic with irrelevant information that undermines your own argument.

 

[ZeeOSix]

I am following the details. You make assumptions that support your augment and then type pages of contradictory words to wash away the details that don't align. Verbatim quote:

Only is an absolute word. It means exclusively. You said only and then named a single condition. Out of the gate I showed a set of pump curves for a PD pump that shows flow is reduced as backpressure is increased. So it's not only. Holding pump speed and viscosity constant, pressure and pressure relief (bypass) are two flow reduction contributions.

You're not really following along with much, because as this discussion has progressed I clarified my viewpoint with respect to the context of the discussion, which was that only adding 1-2 PSI of dP to the pump's back pressure is not going to reduce any measurable oil flow - ie, it won't matter. You certainly are not following, because if you were this debate (them Mods will see it as "bickering") should have been over a long time ago.

I've asked you many times how much you think a healthy automotive PD pump it going to reduce flow volume when the back pressure goes from 75 to 77 PSI. The answer is it will not be noticeable to matter. I said that many time, yet you ignore that and just want to try and call me out an argue about things out of context ... essentially went from a technical discussion to a troller.

Here you've confused a few things. The curve I showed was for a single orifice. It's nonlinear. That's the only point I needed from that graph. Which contradicts your resistor theory.

I never once claimed a "resistor theory" ... you're making up assumptions again. When I said: "The engine oiling system is essentially a fixed flow resistor comprised of various components." all I meant was it's a flow "resistor" (a resistance to flow) that doesn't change. Should have used the word "resistance" so you didn't trigger, lol. I didn't say or claim is acts like a linear resistor in an electrical circuit - that's something you assumed. You're reaching and assuming way too much in this discussion it seems. What I'm saying is that the flow resistance (not saying it behaves like a "resistor" in an electrical circuit) of the system is fixed. In other words, it you changed the clearance for instance of all the journal bearings, the flow resistance of the system will change, and you would see a different RPM vs oil pressure curve, mostly shifted up or down as the flow resistance increases or decreases for changed clearances between components in the oiling system. Or if you had a PD pump what was pretty worn out, you would see the RPM vs oil pressure curve change due to a meaningful change in pump slip.

And also, you showing the flow curve for a single, small orifice doesn't mean the entire engine oiling system as a whole flows like that orifice does - so no, it's not "all you needed". You have a similar flow vs oil pressure for an actual engine?

First, resistors are only linear as long as they aren't heated relative to the power they absorb, which is always. Linearity is only an handy approximation made within limits.

So first you said the oiling system was a [singular] fixed resistance value...linear...like ohms law. Then later you said couldn't do that because the system is an additive system made of many components and can't be approximated by a single value. So you contradicted yourself to make an argument. If either is true then the other half of your approximative argument falls apart.

See my previous response above that clarifies all the "resistor" misconception you latched onto. I never claimed it acts like a linear resistor in an electrical circuit - I never even once brought up any kind of electrical circuit analogy for anything - you did by assuming that. You saw the word "resistor" and went off in the weeds. It's a fixed "resistance" made up all many different flow components, that's what I meant. The word "resistor" doesn't automatically mean it's an actual "resistor" in an electrical circuit ... in general terms without the context, it can mean anything that causes a resistance to something. So is this a debate on the meaning and context of words, or some other kind of technical discussion, lol?

If you take electrical engineering 101 (circuits) you'll see equations where banks of resistors can be summed into one single value. If you take mechanical engineering xxx (fluid dynamics) you'll see equations that do the same, even for nonlinear components of a system. Take a look at Bernoulli's equation. There's a nonlinear term in there. (Fluid has to speed up to move through an orifice.) Summing a bunch of nonlinear components is harder but it can be done. That said, you can't take the sum and then extract back the components. You can't weigh the bag of apples and then tell me how much each apple contributes. You showed a nonlinear curve and then said it was nonlinear because of pump slip. I argue that it is also nonlinear because the oiling system moves oil (flow rate) in a nonlinear way vs pressure. (You said it too.) So you can't then go on to say one drives the shape of the graph. They both do.

Re: Bold part. Yes, and that further makes my point that I didn't claim it acts like a "resistor" in an electrical circuit - never brought up any kind of "electrical circuit analogy". I agree the shape of the RPM vs oil pressure curve is due to multiple things going on, including some slight pump slip. But here's the clincher ... if the pump slip is only part of that curve roll-off, then the pump slip has to be even a smaller factor than if the roll-over was all pump slip. That just further supports my comment that adding a few more PSI of dP to the pump back pressure from a slightly more restrictive oil filter - that's the main context of this discussion. That means it's even less of an impact to a healthy PD pump. Like I've said all along, using a filter with a few more PSI of dP added to the system does not matter to a PD oil pump. That's been the whole focus of what this discussion is about. You think it would have some big impact, without any real info (just theory) to make that claim. I proved it by running 4 different brand filtes on the same engine and saw absolutely zero change in the PRM vs oil pressure curve. And you never answered what you think that curve would look like with adding a few more PSI of back pressure. I know what it looks like. Adding a little more back pressure has zero effect, or an effect so small I couldn't measure it and therefore it doesn't matter. Go back to post #6.

Flow can't be inferred but not quantitatively extracted from the pressure vs rpm data you harvested from your engine, because it balls together multiple nonlinear relationships into one nonlinear output. You've make my argument even if you can't stomach to admit it.

I haven't made your argument at all, you just think I did from more assumptions and failure to actually follow the discussion as it's unfolded.

Because you didn't add the qualification until you were called on it.

This is why you're not really following along. It takes explaining things 3 times before you get my viewpoints.

Okay.

That's what you said. I didn't have to say it. Flow not being reduced means constant.

A key to technical writing is to say less and stick to first principles and applicable basis behaviors. The more you try to bring in other information as "examples" the more you run the risk of confusing the topic with irrelevant information that undermines your own argument.

Again, it's all about context in the discussion. Like I said from the beginning in post #6:
"The filter with a little less flow area in the base plate inlet holes might produce 1-2 PSI more dP at high RPM oil flow, but that will not matter to the positive displacement oil pump."

Not following along also means taking the discussion out of context. And you have already agreed with me on that context, so you're now just going in circles trying to find more things to argue about. If you can proved to me that adding a few more PSI of back pressure makes a real measurable difference in an automotive PD oil pump flow volume output to the engine's oiling system, then show it with real data, not with "theory" of "feelings".

 

[johnmyster]

Okay, ignore electrical. My bad. But flow "resistance" isn't constant. Because curves. I've shown you curves. Bernoulli's equation is velocity squared. It's not a constant response to flow. It's squared. Pressure loss follows flow rate squared.

I agree the shape of the RPM vs oil pressure curve is due to multiple things going on, including some slight pump slip. But here's the clincher ... if the pump slip is only part of that curve roll-off, then the pump slip has to be even a smaller factor than if the roll-over was all pump slip.

This is only true if you the underlying, summed curves are curved in the same direction. They they actually aren't.

So once again, a quick assumption to favor a point, but not really on a sound footing. The pressure vs rpm curve you presented cannot be used as a quantitative basis for the flow within the system without looking at the component responses individually.

This isn't a discussion. It's you with your thumbs in your ears, refusing to state that your initial post (which I have shown you three times) much like your posts on the topic for the past ten years state an absolute that is untrue - that PD pump flow is only impacted by pump bypass. However negligible it might be in the grand scheme of a particular application, flow rate is impacted by system pressure.

 

[ZeeOSix]

Okay, ignore electrical. My bad. But flow "resistance" isn't constant. Because curves. I've shown you curves. Bernoulli's equation is velocity squared. It's not a constant response to flow. It's squared.

This is only true if you assume the curves are curved in the same direction. They aren't.

If you look at the orifice flow curve I gave you, what happens? As pressure increases, flow increases, but decreasingly so. (Second derivative is negative.)

If you look at the PD flow curve I gave you, what happens? As pressure increases, slip losses increase, but decreasingly so.

But wait, the pump is a positive pressure device. (Pressure and flow are in the same direction.) The orifice is a negative pressure device. (Pressure loss in the direction of flow.) The same turbulent effect that explains a decreasing increase of flow through the engine (with pressure) is increasing the output of the pump (with pressure) by decreasing slip. The two curves are in the opposite direction when summed (taking a mass balance approach to the system.)

The two contributions are in the opposite direction. Which one dominates first as we continue to increase RPM? Well, we could assume.

So once again, a quick assumption to favor a point, but not really on a sound footing. The pressure vs rpm curve you presented cannot be used as a quantitative basis for the flow within the system without looking at the component responses individually.

The bottom line is the RPM vs oil pressure curve I produced is a measure of the pump feeding oil volume to that oiling system at those operating conditions - regardless of all the flow factors going on simultaneously inside the engine as RPM increases, or the level of pump slip going on. That oil pressure curve is a fingerprint of that oiling system in operation under those operating conditions. Changing anything in the oiling system (like bearing clearances, oil pump worn badly, even oil filters, etc), that would have any real impact on the system would change the shape of that curve. Changing oil filters had absolutely zero effect, and the curve was identical in all cases.

This isn't a discussion. It's you with your thumbs in your ears, refusing to state that your initial post (which I have shown you three times) much like your posts on the topic for the past ten years state an absolute that is untrue - that PD pump flow is only impacted by pump bypass. However negligible it might be in the grand scheme of a particular application, flow rate is impacted by system pressure.

Again, you're not following along because context matters. I think this discussion on your part is trying to twist things up by not staying on context. What I said in my first post #6 is absolutely true. If you think it's not, then tell me just how a 1-2 more PSI of back pressure on a healthy PD pump actually matters - how would it actually matter in this case of oil filters? It doesn't is the bottom line.

My first post in this thread (post #6) says exactly this (quote below), which I've pointed out many times already and I will always stand by ... yet you really don't want to listen or understand it. Instead, you latched on to it and twisted it all up out of context. You think it will actually matter? If so, how much? You already basically agreed that it wouldn't really matter being so negligible. If 1-2 PSI of added back pressure caused the PD to reduce flow output by a noticeable amount, then I'd say that pump is junk and/or pretty worn out.

"The filter with a little less flow area in the base plate inlet holes might produce 1-2 PSI more dP at high RPM oil flow, but that will not matter to the positive displacement oil pump."

So, in the context of the difference in oil filter dP vs flow curves, it will not really matter to a PD oil pump on an engine. Like I mentioned earlier, if the filter was so clogged and resistive to flow where its dP was huge (like 20+ PSI) then you might see an actual impact on the oil pressure and volume supplied to the oiling system with the pump still out of pressure relief. Once the pump hits pressure relief, then that's a whole different scenario.

My claim all along is that a small change like an oil filter with only a few more PSI of dP is meaningless to a mechanically healthy PD oil pump, and it doesn't make any difference to the oiling system ... it doesn't really matter, my original claim in post #6. I stand by that claim, and I'll keep making that claim for another 10 years. You have absolutely nothing that proves otherwise, whereas I do have measurements that show using many different oil filters had zero impact on the RPM vs oil pressure curve - specifically meaning the impact on the pump slip was not noticeable. If any of those filters had a massive difference in flow restriction, it would have been seen as a change in the curve. Yet, it was not seen so the impact of different oil filters flow resistance on pump slip is a non-issue on an automotive engine ... it doesn't matter. It's the context of the whole debate, and that context will never change.

 

[johnmyster]

What I'm saying is that the flow resistance (not saying it behaves like a "resistor" in an electrical circuit) of the system is fixed. In other words, it you changed the clearance for instance of all the journal bearings, the flow resistance of the system will change, and you would see a different RPM vs oil pressure curve, mostly shifted up or down as the flow resistance increases or decreases for changed clearances between components in the oiling system. Or if you had a PD pump what was pretty worn out, you would see the RPM vs oil pressure curve change due to a meaningful change in pump slip.

And also, you showing the flow curve for a single, small orifice doesn't mean the entire engine oiling system as a whole flows like that orifice does - so no, it's not "all you needed". You have a similar flow vs oil pressure for an actual engine?

Okay then, thought experiment time. As you say, assume fixed oiling system geometry. The "context" (the hypothesis) is that pump slip is not relevant in the domain of a typical passenger car internal combustion oil pump. No more blanket statements about how positive displacement pump flow is not a function of backpressure. (I've shown it is.) We will get specific to application.

So everywhere I turn (including fluid dynamics models) for incompressible fluid pressure loss vs flow measurements across a fixed geometry, I see this relationship:

So pressure drop follows the square of flow rate. Or, pressure increase fourfold is required to double flow. That makes sense. If we crack our fluid dynamics textbooks we see velocity squared, flow squared, etc., related to driving pressure requirement. I never claimed any engine oiling system is a single orifice. It's many. So we would expect the sum of many to exhibit a similar behavior, where the largest orifices dominate the consumption (flow) relationship. Sum up many quadratics and you still have a quadratic. The coefficients sum.

So follow me here. Assume your Z06 oil pump has zero slip. Clearances are zero. Perfect pump. It is a perfectly efficient PD pump. In that case the flow rate and RPM would be directly coupled, independent of backpressure. No flow dependence on pressure or viscosity relationship exists. Only speed. Spin it twice as fast, get twice the flow rate. No loss terms in the model are necessary. In that case, the graph you've presented, the horizontal axis could be switched from RPM into flow. The two are directly related. (RPM directly represents flow through some scalar value, but it's not ^2, ^3, exponential, log, ^0.5, etc.)

In that case, wouldn't we expect to see the quadratic form I've shown above? Double RPM results in double the flow. Double the flow into a fixed "resistance" of flow requires 4x the pressure delivered by our hypothetical perfect pump. In that case, you'd see the graph curl up. Positive second derivative. The graph you've shown isn't quadratic. It isn't even linear. It's log form. What contribution could possibly cause this?

Is motor oil non-Newtonian in the "context" of interest? (No.) Is the oiling system geometry changing? (No. Fixed orifices and passages everywhere. No turbines, regulators, reliefs, check valves, etc. We've assumed the pressure relief valve isn't contributing in this "context.") Is oil temperature changing over the course of the experiment? (No.) I just can't think of any explainable mechanism outside of pump performance (slip?) that would explain the shift of a textbook quadratic relationship into a log/asymptotic form. (The PD pump performance slip curves I've shown even have this form.)

What other pump/fluid principle could be at play here?

I've looked at pressure drop of an incompressible fluid through an orifice. Quadratic. I've looked at pressure drop around an elbow. Quadratic. Pipe of constant diameter. Quadratic. Pipe of varying diameter. Quadratic. Across filter media. Quadratic. I'm at my wits end.

I'm all ears here. If not pump slip, what mechanism dominates all these quadratics so much that the empirical data you've shown takes the log/asymptotic form?

 

[ZeeOSix]

Okay then, thought experiment time. As you say, assume fixed oiling system geometry. The "context" (the hypothesis) is that pump slip is not relevant in the domain of a typical passenger car internal combustion oil pump. No more blanket statements about how positive displacement pump flow is not a function of backpressure. (I've shown it is.) We will get specific to application.

Again ... I never made any such "blanket statement". I said that PD pumps do have slip from back pressure - go back and read again from post #6. I made a specific statement with specific context, that an oil filter with a few more PSI of dP back pressure on a healthy automotive PD pump will not make any real difference to the amount of pump slip to effect the flow to the oiling system enough to actually matter. No engine oiling system is going to be impacted in any meaningful way by the use of oil filters with slightly different flow characteristics (ie, dP vs flow). My test data with using different oil filters proves that claim - no difference in the RPM vs oil pressure curve with the only variable being the oil filter. You need to get off this merry-go-round, because you've been wrong all along on what you "think" I've said due to taking things out of context. Context matters a lot in this discussion. So get off that merry-go-round and let's have an actual technical discussion like I've been trying to do.

So everywhere I turn (including fluid dynamics models) for incompressible fluid pressure loss vs flow measurements across a fixed geometry, I see this relationship:

So pressure drop follows the square of flow rate. Or, pressure increase fourfold is required to double flow. That makes sense. If we crack our fluid dynamics textbooks we see velocity squared, flow squared, etc., related to driving pressure requirement. I never claimed any engine oiling system is a single orifice. It's many. So we would expect the sum of many to exhibit a similar behavior, where the largest orifices dominate the consumption (flow) relationship. Sum up many quadratics and you still have a quadratic. The coefficients sum.

I agree so far. I have never claimed anything otherwise. And agree that the flow vs pressure curve would look similar to the graph - ie, pressure increasing in a non-linear fashion as flow rate increases linearly.

So follow me here. Assume your Z06 oil pump has zero slip. Clearances are zero. Perfect pump. It is a perfectly efficient PD pump. In that case the flow rate and RPM would be directly coupled, independent of backpressure. No flow dependence on pressure or viscosity relationship exists. Only speed. Spin it twice as fast, get twice the flow rate. No loss terms in the model are necessary. In that case, the graph you've presented, the horizontal axis could be switched from RPM into flow. The two are directly related. (RPM directly represents flow through some scalar value, but it's not ^2, ^3, exponential, log, ^0.5, etc.)

I agree with this, with your assumption of zero pump slip.

In that case, wouldn't we expect to see the quadratic form I've shown above? Double RPM results in double the flow. Double the flow into a fixed "resistance" of flow requires 4x the pressure delivered by our hypothetical perfect pump. In that case, you'd see the graph curl up. Positive second derivative. The graph you've shown isn't quadratic. It isn't even linear. It's log form. What contribution could possibly cause this?

I was going to bring up yesterday why I think my RPM vs oil prssure curve is the shape it is, but wanted to do some calculations to verify the theory. I think if I ran the same test I did on any engine oiling system that I would see a similar shaped curve - ie, it rolls off like it does as RPM increases - for two main reasons: 1) The flow dynamics of all the journal bearings, and 2) Pump slip.

I think the journal bearing flow dynamics is the main cause of the curve roll-off, and pump slip on top of that is a much weaker secondary factor. You have to understand the way journal bearings work and flow as a function of their RPM to understand this theory. As the engine RPM increases, all the journal bearings essentially "self pump" oil volume through themselves (that also increases with RPM), which then causes the overall back pressure of the oiling system to decrease more as RPM increases. The closest analogy I can think of is as RPM increases, the journal bearings essentially act like a variable orifice would in a non dymanic system to slighlty increase the flow area as RPM increases. I did some calulation models using a dP calcuation tool to simulate that aspect (assuming a slight orifice increase as RPM increases), and then adding some slight pump slip on top of that results in curve with the same basic shape as my test data graph.

Is motor oil non-Newtonian in the "context" of interest? (No.) Is the oiling system geometry changing? (No. Fixed orifices and passages everywhere. No turbines, regulators, reliefs, check valves, etc. We've assumed the pressure relief valve isn't contributing in this "context.") Is oil temperature changing over the course of the experiment? (No.) I just can't think of any explainable mechanism outside of pump performance (slip?) that would explain the shift of a textbook quadratic relationship into a log/asymptotic form. (The PD pump performance slip curves I've shown even have this form.)

Re: Bold part. I think in essence, the overall oiling system resistance to flow (which causes the back pressure on the pump) is changing (becoming less restictive to flow) from the dynamic "self pumping" aspect of the journal bearings as the engine RPM increases, as explaned above. And I think it's the main reason, along with a much smaller pump slip factor, that would explain the shape of my test data curve. And the reason using differernt oil filters (with slight dP vs flow characteristics) made no difference in the curve during my tests - because the pump slip is not a major factor going on when the back pressure change on the pump is minor. Now if a large amount of back pressure was added (like 20+ PSI), the pump slip factor would most likely be noticable, and the RPM vs oil pressure curve would reflect that up to the point of the pump hitting pressure relief.

What other pump/fluid principle could be at play here?

I've looked at pressure drop of an incompressible fluid through an orifice. Quadratic. I've looked at pressure drop around an elbow. Quadratic. Pipe of constant diameter. Quadratic. Pipe of varying diameter. Quadratic. Across filter media. Quadratic. I'm at my wits end.

I'm all ears here. If not pump slip, what mechanism dominates all these quadratics so much that the empirical data you've shown takes the log/asymptotic form?

See above. Try to model it to account for some assumed slight oiling system back pressure reduction factor as the RPM increases due to the "self pumping" dynamics of the journal bearings. It doesn't take much of a change to see a big effect on the pressure vs RPM curve. And when you add in some slight pump slip as the back pressure increases on top of that, the curve has the same shape as my test data graph. There are lots of simultanious factors going on at the same time in an engine that's going through the RPM range.

If you could run the PD pump through its same RPM range on the engine when it's not running, I think you would see the pressure increase with flow as you would expect. And in that test, you would see only the pump slip factor as the pump RPM increased. That would be an interesting experiment.

I will still claim that the PD pump slip is not enough in a mechanically healthy pump feeding an engine's oiling system (regardless of what's going on down stream of the pump) to make any difference that matters when a small back pressure change occurs from using different oil filters on an engine. That was my orginal claim made in post #6 with that context, and I will always make that claim based on my testing of different filters on the same engine under the same exact operating conditions.

 

[ZeeOSix]

... deleted, repeat info.

 

[johnmyster]

Your assertion that the oiling system is a fixed behavior system doesn't fit the data being presented unless pump slip is enough to govern the observed behavior (concave up hyperbolic vs concave down.)

The best I can come up with this, although the RPM range isn't in the domain of our interest. Very little isobaric data is out there. I haven't yet read the study to see how much aligns or doesn't with our use case. I did come up with a few older papers modeling lubricant flow for journal bearings and can investigate further at some later time.

 

[ZeeOSix]

Your assertion that the oiling system is a fixed behavior system doesn't fit the data being presented unless pump slip is enough to govern the observed behavior (concave up hyperbolic vs concave down.)

My initial assertion that the oiling system's back pressure was fixed was wrong. Like I said in post #31, I updated that assertion after digging into this further. In that post I said: "I was going to bring up yesterday why I think my RPM vs oil pressure curve is the shape it is, but wanted to do some calculations to verify the theory."

I've had lots of journal bearing discussions over the years on this board, and after more thinking I recalled how journal bearings "self pump" oil volume through themselves, and that's why I'm saying that's the main cause of the "RPM vs oil pressure" curve shape. That curve will not roll over unless the oiling system flow resistance is decreasing some with increased RPM.

I came up with some curves that show what's going on after doing some modeling using a dP calculation tool where I tried to account for slight decreasing oiling system back pressure due to the journal bearing flow dynamics, and also accounting for some pump slip on top of that, Both factors have a totally different effect on the curve.

The key in this discussion is that the faster a journal bearing rotates, the more oil it will "self pump". If you study journal bearings more, you will see that they typically flow more oil as they spin faster, regardless of the supply pressure to them. If the supply pressure is held constant, the bearing will flow more oil as the bearing RPM increases - this is a key point. Your example above may not be applicable to bearings in an engine that only revs to 6000-6500 RPM. You can actually see a slight increase in flow in that graph's red line, so hard to say what that data would look like from 1000 to 6500 RPM. You might want to do more searching for journal bearing flow vs RPM info.

As the engine RPM increases, the journal bearings essentially become less flow restrictive because they are like "mini oil pumps" feeding off of the oil supply reaching them from the PD pump. This causes the overall back pressure on the pump to decrease as RPM increases, which is seen as an oil pressure decrease as engine RPM increases.

From my journal bearing info archives. Journal bearing oil flow is caused by two factors: 1) The natural "self pumping" action of the rotating bearing, and 2) The oil supply pressure. If the bearings flow more oil as their speed increases, then the PD pump back pressure is going to decrease. This is why the oil pressure curve rolls over more with increased engine RPM as shown in my model plots. The self pumping flow factor of the journal bearings essentially reduces the flow restriction of the oiling system as engine RPM increases.

Here's the resulting curves I came up with - took a lot of effort to get this info. The annotations in graph explains what each curve is showing. Pretty clear to me that the curve shape from my Z06 testing is caused by all the journal bearings slightly reducing the overall flow restriction of the oiling system as the RPM increases. Note that the red curve is with a fixed system flow restriction and only pump slip added (which increased from 5% to 20% with RPM), and the basic shape of that curve doesn't change, it just shows increased deviation from the orange curve due to non-constant slip percentage. So the roll-over in oil pressure I saw can't just be caused by only pump slip. The overall flow restriction of the system needs to change lightly (it's pretty sensitive) to make the curve roll over like I saw in my test data. Any small change in system flow restriction makes a big difference. Pump slip in a healthy automotive PD pump may not even be as much at 20% since the pump I used to obtain that was pumping water at 1 cP viscosity, and I used a viscosity of 12.75 cP in my model which is closer to hot motor oil. Thicker oil will cause less pump slip. "Pump Output Pressure" in the graph is essentially the engine oil pressure after the oil filter (pressure sensor location).

Anyway, that's what I've come up with. If you can show it's not really caused by the self pumping action of the journal bearings, then I'm all ears.

 

[ZeeOSix]

Some Info on journal bearings - tip of the iceberg in terms of journal bearing design and operation. The Sommerfeld number is key to understanding how journal bearings work in many aspects, including how they flow oil as a function of their rotational speed. Note the bearing speed factor (N) in the Sommerfeld number equation.

Bearing flow is a function of the Sommerfeld number. As the bearing speed increases and the other factors remain constant, the Sommerfeld number increases. Bearing side leakage increases with bearing speed. This makes them act like mini oil pumps on their own. If a journal bearing was simply fed with an oil reservist at atmospheric pressure (like shown below), it would pump an increasing amount of oil volume as the bearing speed increased. Rod and crankshaft journal bearings are doing the same thing in an engine, even when pressure fed by the PD pump. ... they still act like mini oil pumps. To see their effect on the back pressure with an extreme example, assume they self pumped so much oil that it was equal to the amount of oil the PD pump supplied. In that case, the back pressure on the pump would be near zero as the bearings acted like mini scavenger pumps on the oil volume supplied by the PD pump.

 

[Reloader59]

The "artistic" hole arrangement will have no effect on flow. Made in Ukraine is also fine unless Putin actually conquers them. In that case we should avoid them at all costs.

Putin conquered them a long time ago.