Is it only when the filter is clogged to the point of preventing flow through the filter media? TIA
When does the bypass valve psi matter?
- Thread starter Dave_in_Frisco
- Start date
- Status
- Joined
- Sep 28, 2002
- Messages
- 39,798
Not really ..it would add to the situation. It's mostly when the oil pump is in relief. That's typically when the PSID is most elevated. You're at peak pressure on one side of the media ..with reduced flow passing through it. Reduced flow means reduced back pressure.
Doug Hillary finally confirmed why some Euro-alloy have 30+ PSI rated filters. It's since they may be in pressure relief for a good bit of their lives given the pump volumes, peak pressures, and viscosity that they spec.
Doug Hillary finally confirmed why some Euro-alloy have 30+ PSI rated filters. It's since they may be in pressure relief for a good bit of their lives given the pump volumes, peak pressures, and viscosity that they spec.
Here's an interesting story about bypass valves.
Three times this has happened to me.
I used to be an Amsoil dealer back in the mid 90's.
So I used Amsoil products for everything.
My car was old and worn out, a 1987 VW Golf with 140K miles on it. I mistakenly thought it was a good idea to run the Amsoil Series 2000 racing formula to compensate for how loose the engine was.
Three different times, I went out and started the car only to hear the oil pressure alarm begin screaming. I'd shut off the car, get out, and discover a gigantic puddle of cold oil under the car.
The oil was too think at ambient temperature to pass through the Amsoil filter, and apparently the bypass valve failed (at least these three times), and the oil would squeeze out around the filter gasket. The gasket would give way like it was burping, and I'd lose one to two quarts of very expensive oil, and make a huge mess.
I quit running that thick oil, and switched to the Mobil 1 synthetic filter, all was fine.
Moral of this story?
I dunno.
Three times this has happened to me.
I used to be an Amsoil dealer back in the mid 90's.
So I used Amsoil products for everything.
My car was old and worn out, a 1987 VW Golf with 140K miles on it. I mistakenly thought it was a good idea to run the Amsoil Series 2000 racing formula to compensate for how loose the engine was.
Three different times, I went out and started the car only to hear the oil pressure alarm begin screaming. I'd shut off the car, get out, and discover a gigantic puddle of cold oil under the car.
The oil was too think at ambient temperature to pass through the Amsoil filter, and apparently the bypass valve failed (at least these three times), and the oil would squeeze out around the filter gasket. The gasket would give way like it was burping, and I'd lose one to two quarts of very expensive oil, and make a huge mess.
I quit running that thick oil, and switched to the Mobil 1 synthetic filter, all was fine.
Moral of this story?
I dunno.
140k is worn out? Mine's still breaking in!
I think too low bypass is ok, but too high is a problem. Now, maybe too much flow "in bypass" could be bad on the backside that needs to build pressure to pump up hydrolic lifters and actuate intake runners and such....I dunno. otoh, I can see how a lowered BPV could be an advantage if you were using the engine in an performance application. I'll take 30% more flow over 30% more filtering.
Gary, Please provide more details.
- Joined
- Sep 28, 2002
- Messages
- 39,798
I hope this is what you're looking for...
Okay... unless your oil pump is in relief, the oil has no choice but to accelerate when encountering ..well, anything. Now if your oil passages are empty, you may see some PSID at start up ..and if the volume pushed ..and the viscosity is high enough, you may actually have some bypass activity. Once the engine is fully enveloped at the filter (ponder this for a moment) you're in a kinetic type situation. The filter does have a resistance ..but the engine is far greater ..as long as the pump isn't relieving flow ..the filter can't trump the engine for restriction.
It's like going from 4 lanes to 3 ..to 2 ..to 18 (whatever) ..does it really matter if your exit is 1 lane wide. The BIG choke is the engine. The traffic jam of the single outlet is what's giving you "back up" or "back pressure" that you see on your gauge.
Now when the pump reaches it's relief limit ..then the dynamics change up a bit. There is no dedicated flow ..now there's a fixed pressure ..with divergent flows. The pressure from the engine side is still a product of (now reduced) flow ..but the pressure on the pump side is at the regulated limit. With certain rough edges ..the PSID is related to how the pump output is out of phase with the realized flow to the engine. It probably doesn't work out exactly like this ..but
Let's say you had a 100lb limit ..and you happen to have a differential pressure gauge across the filter (this I have done) ..suppose you see 10PSID. So now you're only producing 90lb on the engine side of the filter. You're regulated to 100lb on the pump side. Let us assume a 1gpm flow at whatever visc the fluid is ...then you've got 10% shunted/recirculated flow.
To stop this inter-relationship from tipping too far into the shunted/recirculated end of things, the filter's bypass valve limits the PSID. Since the PSID is limited ..then the shunted/recirculated flow is limited.
That's the "in the vacuum" view.
Now when you get to the Porsche/Audi/VW/etc. that have 30lb+ bypass valves (Thanks to Doug Hillary for giving me the rationale behind this)..it's where the volume/visc/pressure limits are going to assure routine high differential on the filter. The filters are specially reinforced to withstand the added pressure ...but the engines, if fitted with normally rated bypass valves, may spend a decent amount of time circulating unfiltered oil. In these engine's cases, even the pump relief limit may not limit the oil pressure. The relief port may be too small for the visc of the fluid to pass enough to keep pressure at limit levels.
I also recommend looking here https://bobistheoilguy.com/forums/posts/1204648/
Okay... unless your oil pump is in relief, the oil has no choice but to accelerate when encountering ..well, anything. Now if your oil passages are empty, you may see some PSID at start up ..and if the volume pushed ..and the viscosity is high enough, you may actually have some bypass activity. Once the engine is fully enveloped at the filter (ponder this for a moment) you're in a kinetic type situation. The filter does have a resistance ..but the engine is far greater ..as long as the pump isn't relieving flow ..the filter can't trump the engine for restriction.
It's like going from 4 lanes to 3 ..to 2 ..to 18 (whatever) ..does it really matter if your exit is 1 lane wide. The BIG choke is the engine. The traffic jam of the single outlet is what's giving you "back up" or "back pressure" that you see on your gauge.
Now when the pump reaches it's relief limit ..then the dynamics change up a bit. There is no dedicated flow ..now there's a fixed pressure ..with divergent flows. The pressure from the engine side is still a product of (now reduced) flow ..but the pressure on the pump side is at the regulated limit. With certain rough edges ..the PSID is related to how the pump output is out of phase with the realized flow to the engine. It probably doesn't work out exactly like this ..but
Let's say you had a 100lb limit ..and you happen to have a differential pressure gauge across the filter (this I have done) ..suppose you see 10PSID. So now you're only producing 90lb on the engine side of the filter. You're regulated to 100lb on the pump side. Let us assume a 1gpm flow at whatever visc the fluid is ...then you've got 10% shunted/recirculated flow.
To stop this inter-relationship from tipping too far into the shunted/recirculated end of things, the filter's bypass valve limits the PSID. Since the PSID is limited ..then the shunted/recirculated flow is limited.
That's the "in the vacuum" view.
Now when you get to the Porsche/Audi/VW/etc. that have 30lb+ bypass valves (Thanks to Doug Hillary for giving me the rationale behind this)..it's where the volume/visc/pressure limits are going to assure routine high differential on the filter. The filters are specially reinforced to withstand the added pressure ...but the engines, if fitted with normally rated bypass valves, may spend a decent amount of time circulating unfiltered oil. In these engine's cases, even the pump relief limit may not limit the oil pressure. The relief port may be too small for the visc of the fluid to pass enough to keep pressure at limit levels.
I also recommend looking here https://bobistheoilguy.com/forums/posts/1204648/
I like this discussion. Gary's explanations are excellent. He assumes, however, that pumps don't leak and the pump bypass has no limits. Neither of those is the case. Most, if not all pumps leak, even those touted as positive displacement. The variation in flow with this leakage is minimal but present, none the less. Additionally, bypasses are designed not to bypass. What I mean by that is that, though the valve might open at 50 PSI or so; the galleries in the bypass are not large enough to pass the total volume of the pump at the bypass pressure. The bypass, therefore, still acts as a major restriction to flow and the system pressure can get to a much greater pressure than the relief pressure.
I don't know what combination of flow, viscosity, contamination, element restriction, element area etc. will cause the 8 to 10 psia differential across the filter required for the filter to bypass but it can happen regardless of system pressure and it will reduce the pressure downstream of the filter by that amount. So what will that do to flow to the engine. With the pump bypass closed nothing. However, even assuming no pump leakage, how can volume not be reduced when the pressure is reduced? Simple, the 8 to 10 psi restriction simply does what any restriction does downstream of a positive displacement pump does. Pressure ahead of the filter increases by 8 to 10 psig and the restriction of the engine continues to restrict the flow to keep the same pressure as before bypass.
Put any number of restrictions in a constant flow incompressible stream and they will each will increase pressure at the pump in accordance with their restriction even if the final orifice is so small that it's contribution is far greater than any of the rest.
My interest in this actually comes from discussions I have on occasion, as to the importance of where the filter bypass is located. It is my contention that bypass happens and, when it does, it is a major advantage for the filter to bypass oil at the open end. If the bypass is at the open end; oil is allowed, no, forced, to flow over the dirty element on its way to the engine. Sounds a little less than optimum to me.... has that factor been discussed here?
I don't know what combination of flow, viscosity, contamination, element restriction, element area etc. will cause the 8 to 10 psia differential across the filter required for the filter to bypass but it can happen regardless of system pressure and it will reduce the pressure downstream of the filter by that amount. So what will that do to flow to the engine. With the pump bypass closed nothing. However, even assuming no pump leakage, how can volume not be reduced when the pressure is reduced? Simple, the 8 to 10 psi restriction simply does what any restriction does downstream of a positive displacement pump does. Pressure ahead of the filter increases by 8 to 10 psig and the restriction of the engine continues to restrict the flow to keep the same pressure as before bypass.
Put any number of restrictions in a constant flow incompressible stream and they will each will increase pressure at the pump in accordance with their restriction even if the final orifice is so small that it's contribution is far greater than any of the rest.
My interest in this actually comes from discussions I have on occasion, as to the importance of where the filter bypass is located. It is my contention that bypass happens and, when it does, it is a major advantage for the filter to bypass oil at the open end. If the bypass is at the open end; oil is allowed, no, forced, to flow over the dirty element on its way to the engine. Sounds a little less than optimum to me.... has that factor been discussed here?
Last edited:
- Joined
- Sep 28, 2002
- Messages
- 39,798
Quote:
He assumes, however, that pumps don't leak and the pump bypass has no limits. Neither of those is the case.
True. In the vast multitude of posts that I've had on this matter, I usually acknowledge those things ..and put them to the side for the sake of discussion. That's sorta why I qualified it with "That's the "in a vaccum" view. It shortens an already lengthy description without the minor nuances that aren't all that germane to the "conceptual view" that's being communicated. Those tend to fall into the realm of "fringe". "When I'm at 9000 rpm's..." or whatever
There are surely exceptions and extensions that one can enter into that alter some things ..but you can do that with any topic.
Quote:
What I mean by that is that, though the valve might open at 50 PSI or so; the galleries in the bypass are not large enough to pass the total volume of the pump at the bypass pressure. The bypass, therefore, still acts as a major restriction to flow and the system pressure can get to a much greater pressure than the relief pressure.
I'm sure that you could create conditions that make this occur ..but I would say that they aren't present in the vast commonality of incidents of bypass operation.
Quote:
I don't know what combination of flow, viscosity, contamination, element restriction, element area etc. will cause the 8 to 10 psia differential across the filter required for the filter to bypass but it can happen regardless of system pressure and it will reduce the pressure downstream of the filter by that amount.
This, if I'm reading it correctly, I don't agree with. All of the sub-elements mentioned have no impact on pump output (outside of the minor losses that have marginal impact on our view in the vast majority of cases). If you can agree that pump output is not substantially effected below the relief level ..then 100% of the sensible pump output goes to the engine. That means that regardless of visc/flow rate/whatever ..the "produced back pressure" will be identical as viewed on the engine side of the media. Any "added" pressure being contributed by the media HAS to fall into a subordinate relationship to the VASTLY GREATER resistance that the engine downstream creates. Outside of the minor pump losses ..this is a series circuit as viewed by the pump and filter ..any complexity to it beyond the oil gallery is irrelevant. The whole engine is one BIG resistance/restriction and generates the most back pressure at the given flow. The filter has to fall into a proportion of that massive restriction ..commensurate with how it elevates the pressure in addition to the engine.
Loading (contamination) can surely effect PSID in a "reactive" manner that falls outside of a static "vacuum" (linear) view. It's mostly transient in nature ...but for the most part you're looking at the long end of a long curve in a filter's life.
There's no two ways about it ..a filter has both a resistance and an impedence to flow. But when you're not in pump relief ..or don't have a worn leaky pump or one of low efficiency (remember- we're talking the vast majority of users and aren't accounting for every "what if" here) ..it's TRUMPED so big time by the engine in "relative resistance" ...making it a very small player in pressure influence.
Quote:
However, even assuming no pump leakage, how can volume not be reduced when the pressure is reduced?
Without any leakage ..how can volume be altered in any way
1gpm through a 5 FOOT pipe
1gpm through the eye of a needle
Where is the flow altered? It's not. 1gpm is 1gpm ..what is altered is limited to pressure generate ..or "back pressure" generated at that flow.
Quote:
Simple, the 8 to 10 psi restriction simply does what any restriction does downstream of a positive displacement pump does. Pressure ahead of the filter increases by 8 to 10 psig and the restriction of the engine continues to restrict the flow to keep the same
Sorta ..but your presumption of magnitude is off. The filter, without any flow alterations from the pump, cannot get out of proportion with the engine in terms of relative pressure generated (aka inverted view of "resistance") It will always be small with a fully enveloped engine below pump relief.
Again this is exclusive of transient and exceptional conditions like accounting for loading.
A filter does have a given PSID spec at a given flow at a given viscosity.
Let's say it's 10PSID @ 10gpm with a ISO 150 fluid. Fine. When compared to one that shows 5 PSID at the same flow rate at the same viscosity ..it's "twice" as restrictive ..or the other one is twice as "free flowing".
But seeing as (again, in the vast majority of engines on the road) ...you're max volume will be 5 gpm ..and you'll be in relief if the visc is too high ...just how much does it really mean anything
That is, only a select few circumstances ...rolling fleet wide globally ..will ever fall into the "spec's spec". Since you'll most likely never see 10 gpm of flow ..just what can it mean in terms of free flowing or restrictive ..when the REAL choke point is the engine ..by a HUGE margin.
Take my high volume pump on my 4.0. The max that I can process through the engine, in terms of "normal" throughput is about 5gpm. I've got 9 gpm output @ 3500. I have to go to lighter oil to not slam up against the relief off idle. So, IF I'm generating 9 gpm of output and IF the visc at the moment falls into the filter spec's spec ...I MAY generate a given PSID that will still have to be in proportion to the generated back pressure from the engine. If the visc is too high or the volume too high ..the pressure will reach a level that opens the relief valve ..and then you're limited in (NOW) a pressure "drop" across the filter ..which will be the difference of a reduced volume to the engine ..and the subsequent reduced generated pressure on that side of the media ..the upper side is now (again, typically) "pressure regulated".
Quote:
Put any number of restrictions in a constant flow incompressible stream and they will each will increase pressure at the pump in accordance with their restriction even if the final orifice is so small that it's contribution is far greater than any of the rest.
I have no dispute with this assertion. It's a fundamental rule of my whole gig here. What I would offer you is to view all of them as velocity changes that are evidenced in terms of pressure ..where flow is not altered (in our perfect pump). 1gpm is 1 gpm
Essentially the pump relief and the filter bypass valve are "cushions" or "slack" in an otherwise (mostly) solid fluid transmission mechanism. The filter bypass valve rating is to protect the media from collapse. The bypass valve is to protect the engine from oil starvation where you may indeed, due to viscosity and/or volume challenge the media's ability to pass the oil ..and the pump's relief's limits for shunting it.
Quote:
My interest in this actually comes from discussions I have on occasion, as to the importance of where the filter bypass is located. It is my contention that bypass happens and, when it does, it is a major advantage for the filter to bypass oil at the open end. If the bypass is at the open end; oil is allowed, no, forced, to flow over the dirty element on its way to the engine. Sounds a little less than optimum to me.... has that factor been discussed here?
It's come up a few times. First ..it's not like you're opening a flood gate when the bypass valve opens. It's only going to pass the excess required amount to maintain the differential. It should (again, in our "vaccum of perfection" view) merely maintain the differential. That is, the same flow potential across the media will still be where it was just before the valve opened. Now I'm sure that we can pull some scenarios out of our behinds with envisioning pulsing and whatnot ..but if you look at it from a variable applied flow aspect ..the media was passing 100% of the flow at a slightly lower volume at a slightly lower PSID ..there's no reason to think that ALL of that flow ..or any of it for that matter decides to take a hike to the bypass outlet ..since it offers no advantage in potential (see what I'm saying?). The bypass valve should only be passing any excess volume required to maintain the differential.
Again, I'm sure that there are exceptions that can be constructed if desired.
The topic of open end vs. closed end has come up in the past. Ford's engine division insists that the bypass be on the open end. It's not due to any "washing effect" (which I can't really see occuring - see above) ..but due to "silt" from debris that may fall out of suspension (on the bottom or exterior wall if horizontal) which can be entrained in the bypass flow upon cold start. I've only heard of one member reporting finding such accumulations in a filter ..and had mostly considered it bunk ..but then I cut open a M1 filter and found a good accumulation of goo. This was probably some film formation that was liberated "en mass' and not indicative of normal conditions ...but it did demonstrate the potential for this to occur.
This was a horizontally mounted filter and this is the first time that I've ever actually seen this type of thing.
He assumes, however, that pumps don't leak and the pump bypass has no limits. Neither of those is the case.
True. In the vast multitude of posts that I've had on this matter, I usually acknowledge those things ..and put them to the side for the sake of discussion. That's sorta why I qualified it with "That's the "in a vaccum" view. It shortens an already lengthy description without the minor nuances that aren't all that germane to the "conceptual view" that's being communicated. Those tend to fall into the realm of "fringe". "When I'm at 9000 rpm's..." or whatever
Quote:
What I mean by that is that, though the valve might open at 50 PSI or so; the galleries in the bypass are not large enough to pass the total volume of the pump at the bypass pressure. The bypass, therefore, still acts as a major restriction to flow and the system pressure can get to a much greater pressure than the relief pressure.
I'm sure that you could create conditions that make this occur ..but I would say that they aren't present in the vast commonality of incidents of bypass operation.
Quote:
I don't know what combination of flow, viscosity, contamination, element restriction, element area etc. will cause the 8 to 10 psia differential across the filter required for the filter to bypass but it can happen regardless of system pressure and it will reduce the pressure downstream of the filter by that amount.
This, if I'm reading it correctly, I don't agree with. All of the sub-elements mentioned have no impact on pump output (outside of the minor losses that have marginal impact on our view in the vast majority of cases). If you can agree that pump output is not substantially effected below the relief level ..then 100% of the sensible pump output goes to the engine. That means that regardless of visc/flow rate/whatever ..the "produced back pressure" will be identical as viewed on the engine side of the media. Any "added" pressure being contributed by the media HAS to fall into a subordinate relationship to the VASTLY GREATER resistance that the engine downstream creates. Outside of the minor pump losses ..this is a series circuit as viewed by the pump and filter ..any complexity to it beyond the oil gallery is irrelevant. The whole engine is one BIG resistance/restriction and generates the most back pressure at the given flow. The filter has to fall into a proportion of that massive restriction ..commensurate with how it elevates the pressure in addition to the engine.
Loading (contamination) can surely effect PSID in a "reactive" manner that falls outside of a static "vacuum" (linear) view. It's mostly transient in nature ...but for the most part you're looking at the long end of a long curve in a filter's life.
There's no two ways about it ..a filter has both a resistance and an impedence to flow. But when you're not in pump relief ..or don't have a worn leaky pump or one of low efficiency (remember- we're talking the vast majority of users and aren't accounting for every "what if" here) ..it's TRUMPED so big time by the engine in "relative resistance" ...making it a very small player in pressure influence.
Quote:
However, even assuming no pump leakage, how can volume not be reduced when the pressure is reduced?
Without any leakage ..how can volume be altered in any way
1gpm through a 5 FOOT pipe
1gpm through the eye of a needle
Where is the flow altered? It's not. 1gpm is 1gpm ..what is altered is limited to pressure generate ..or "back pressure" generated at that flow.
Quote:
Simple, the 8 to 10 psi restriction simply does what any restriction does downstream of a positive displacement pump does. Pressure ahead of the filter increases by 8 to 10 psig and the restriction of the engine continues to restrict the flow to keep the same
Sorta ..but your presumption of magnitude is off. The filter, without any flow alterations from the pump, cannot get out of proportion with the engine in terms of relative pressure generated (aka inverted view of "resistance") It will always be small with a fully enveloped engine below pump relief.
Again this is exclusive of transient and exceptional conditions like accounting for loading.
A filter does have a given PSID spec at a given flow at a given viscosity.
Let's say it's 10PSID @ 10gpm with a ISO 150 fluid. Fine. When compared to one that shows 5 PSID at the same flow rate at the same viscosity ..it's "twice" as restrictive ..or the other one is twice as "free flowing".
But seeing as (again, in the vast majority of engines on the road) ...you're max volume will be 5 gpm ..and you'll be in relief if the visc is too high ...just how much does it really mean anything
Take my high volume pump on my 4.0. The max that I can process through the engine, in terms of "normal" throughput is about 5gpm. I've got 9 gpm output @ 3500. I have to go to lighter oil to not slam up against the relief off idle. So, IF I'm generating 9 gpm of output and IF the visc at the moment falls into the filter spec's spec ...I MAY generate a given PSID that will still have to be in proportion to the generated back pressure from the engine. If the visc is too high or the volume too high ..the pressure will reach a level that opens the relief valve ..and then you're limited in (NOW) a pressure "drop" across the filter ..which will be the difference of a reduced volume to the engine ..and the subsequent reduced generated pressure on that side of the media ..the upper side is now (again, typically) "pressure regulated".
Quote:
Put any number of restrictions in a constant flow incompressible stream and they will each will increase pressure at the pump in accordance with their restriction even if the final orifice is so small that it's contribution is far greater than any of the rest.
I have no dispute with this assertion. It's a fundamental rule of my whole gig here. What I would offer you is to view all of them as velocity changes that are evidenced in terms of pressure ..where flow is not altered (in our perfect pump). 1gpm is 1 gpm
Essentially the pump relief and the filter bypass valve are "cushions" or "slack" in an otherwise (mostly) solid fluid transmission mechanism. The filter bypass valve rating is to protect the media from collapse. The bypass valve is to protect the engine from oil starvation where you may indeed, due to viscosity and/or volume challenge the media's ability to pass the oil ..and the pump's relief's limits for shunting it.
Quote:
My interest in this actually comes from discussions I have on occasion, as to the importance of where the filter bypass is located. It is my contention that bypass happens and, when it does, it is a major advantage for the filter to bypass oil at the open end. If the bypass is at the open end; oil is allowed, no, forced, to flow over the dirty element on its way to the engine. Sounds a little less than optimum to me.... has that factor been discussed here?
It's come up a few times. First ..it's not like you're opening a flood gate when the bypass valve opens. It's only going to pass the excess required amount to maintain the differential. It should (again, in our "vaccum of perfection" view) merely maintain the differential. That is, the same flow potential across the media will still be where it was just before the valve opened. Now I'm sure that we can pull some scenarios out of our behinds with envisioning pulsing and whatnot ..but if you look at it from a variable applied flow aspect ..the media was passing 100% of the flow at a slightly lower volume at a slightly lower PSID ..there's no reason to think that ALL of that flow ..or any of it for that matter decides to take a hike to the bypass outlet ..since it offers no advantage in potential (see what I'm saying?). The bypass valve should only be passing any excess volume required to maintain the differential.
Again, I'm sure that there are exceptions that can be constructed if desired.
The topic of open end vs. closed end has come up in the past. Ford's engine division insists that the bypass be on the open end. It's not due to any "washing effect" (which I can't really see occuring - see above) ..but due to "silt" from debris that may fall out of suspension (on the bottom or exterior wall if horizontal) which can be entrained in the bypass flow upon cold start. I've only heard of one member reporting finding such accumulations in a filter ..and had mostly considered it bunk ..but then I cut open a M1 filter and found a good accumulation of goo. This was probably some film formation that was liberated "en mass' and not indicative of normal conditions ...but it did demonstrate the potential for this to occur.
This was a horizontally mounted filter and this is the first time that I've ever actually seen this type of thing.
- Joined
- Sep 28, 2002
- Messages
- 39,798
Ooops
(click on the image after you're rerouted for a real close view)
(click on the image after you're rerouted for a real close view)
- Status