Announcement

Collapse
No announcement yet.

M42 oil filter housing passage questions

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    M42 oil filter housing passage questions

    I am working on adding an oil cooler to my turbocharged M42, and the only off-the-shelf option I've ever seen (Mocal takeoff plate) is no longer on the market. Before anyone asks if this is actually necessary, I see 280+F oil pan temperatures on track before I back off and run a couple laps in 5th. 2 laps on, 2 laps off is no fun.

    So the goal is to take a stock early oil filter housing and chop it up to divert flow. I don't see an easy way to divert flow and use the stock filter without adding a bunch of restriction, so I plan to plumb an external oil filter and thermostat to reduce any additional load on the oil pump.

    Problem is, I can't find any good documentation on how flow through the housing works, and there are more cavities and passages than I was expecting to find. Can anyone tell me if I'm right?




    Starting from the top and working downwards.
    • The round port at the top with the oring seal and drainback valve. Oil flow out of the filter housing, post-filter, into the main block oil rifle.
    • The bung to the left of the round oring-ed port is the oil pressure switch, also after the filter
    • Small rectangular port. This is on the unfiltered side...I have no idea where this goes into the engine, especially since it isn't filtered.
    • Large casing void. This should not have any oil in it, and is sealed by the little oring on the end of the filter housing cap bolt.
    • Upper square port, at the bottom left of the filter housing. Oil feed from the pump.
    • Lower square port. dry and not used? Seems like wouldn't be a part of the casting if it didn't have a purpose.

    So, do I have the flow path right? I guess I could have it backwards, but am basing this on the oil pump being at the bottom of the engine closer to those square ports, and most pressurized oiling and the oil rifles being up higher.



    And what is that little rectangular port for? I can't find any reasonable explanation for it, nor have I seen any good block pics with it disassembled. I should have taken better pics when I replaced the gasket 6ish years ago...

    Project M42 Turbo

    #2
    Reading this kicking myself for not knowing the answer... paging simon. If he doesn't know he sure does have detailed photos to help figure it out

    Turbo M42 Build Thread :Here
    Ig:ryno_pzk
    I like the tuna here.
    Originally posted by lambo
    Buttchug. The official poster child of r3v.

    Comment


      #3
      The oil pump flow paths are mostly contained in the front cover. There is one on Ebay with some decent pictures of the flow paths. If you need more pictures I can probably find some from when I rebuilt my engine. I tool the pump and cover apart and cleaned it all.

      http://m.ebay.com/itm/BMW-E30-318i-3...%257Ciid%253A2
      -Nick

      M42 on VEMS

      Comment


        #4
        I have no idea how, all this time, I didn't realize that the M42 used a front cover/gerotor pump. I've even had the pans off of mine and never noticed. Huh.

        I've pilfered these pics from other sources, including that ebay ad, and rehosted them. Hopefully the next person to google M42 oil coolers will end up at this thread and be able to get something from it. I'm going to pedantically step my way through the oil system in hopes that if I start explaining it to myself, the answers to my questions may become clear. Or, at the very least, one of you lot will better be able to explain it to me, haha.

        Starting at the inlet of the system, this pic clearly shows how the upper pan interfaces with the pickup tube. The hole to the left is the dipstick hole, which on my setup doubles as the turbo drainback vent:



        The next interface is between the upper oil pan and the bottom of the front cover. The hole on the right is the pickup for the oil pump. The integrity of this interface and the pickup tube interface are critical for lube system performance - if these gaskets fail, the pump will suck in air. The hole to the left, with the visible snap ring, is the pressure relief valve. Behind it is a spring and a plunger that regulate peak oil pressure when pump flow/pressure has exceeded the requirements of the engine at higher rpm.



        This diagram clearly shows the spring and plunger. The relief cut plunger design is relatively unusual, so for now I'll assume that this was done to either reduce the mass of the valve or to reduce return flow restriction when the valve is open to prevent pressure creep.



        It's not clear to me how the excess flow from the pressure relief valve makes it back into the pan. Typically, there is a bypass port that directly dumps this flow back - the shorter (or rather, the lower the restriction in) this return flow path, the less parasitic loss you'll have in the lube system.

        It does not recirculate to the bottom part of the pickup:



        However, the bottom of that port definitely ties into the pickup on the top side. Maybe just a return path for leakage past the piston, which is just a hydrodynamic seal?:



        Apparently the relief valve design was updated at some point, with the plastic one being the new design. I'd guess that there were some issues with scoring in the early design, causing some loss of oil pressure. Just a guess though.



        You can see the bottom of the valve in the lowest (or leftmost, in this pic) square hole. So that hole may have something to do with the relief valve return flow, because the valve moves downwards when it opens. May just be a machining access hole though.




        So, at this point, what do I know? The bottom-most square hole sees oil flow (or more accurately, pressure since it doesn't go anywhere) when the relief valve is open. The upper square hole is oil flow into the filter housing. Because that's now decided, the round port at the top and the rectangular port below it appear to be clean oil back into the engine, though I still don't know what that rectangular port is for. That last picture also makes it look like the large cavity is actually full of oil - there is a round hole below the two square ones that connects to the spring passage below the relief valve. This may just be a pressure relief for that cavity, but either way, I will need to make sure I don't accidentally block that off. I also don't like setups where 100% of the oil flow has to go past the check valve (more heat, more parasitic load), but that appears to be how this system works. I can't tell without taking a front cover apart and looking at it.

        It looks like it isn't going to be practical to just machine a flat piece of aluminum to do the job then. There's enough going on in there that I will likely have to modify the existing housing to minimize my impact to the lube system, for fear if screwing it up. I am doing this because the car sees a decent amount of track time, after all, and that's a bad place to have a marginal lube system...
        Last edited by gearheadE30; 05-30-2017, 04:10 PM.

        Project M42 Turbo

        Comment


          #5
          Referring back to a previous explanation of the oil flow that I made, since I've looked into this myself, back when I had an M42:

          Originally posted by varg View Post
          The M42 oil pump suction side is the square hole on the bottom of the timing case, and the discharge is the top of the two stacked square holes visible when the oil filter housing is off. The pressure relief valve is visible through these ports, the lower of the two is completely blocked when the relief valve is in the normal position. When the relief valve opens the bottom port which is connected through to pump suction is opened and pressurized oil is allowed to recirculate through the pump.
          My understanding hast been that the round port feeds the block and head, since it leads directly into a round hole on the front of the block and nowhere else, and the rectangular port just beneath it feeds the timing case itself.

          There is no easy answer to it, but machined sandwich plate that goes between the housing and the timing case, diverting flow from the round port through a filter then back to the round port in the case, is the only one I've come up with. Problem is, it would have to be a rather thick plate and this would complicate alternator mounting.

          An easier solution would be to run a scavenge pump based system which draws from the sump, possibly from the same port as turbo return, feeds through an oil cooler, then discharges back into the sump near the oil pump pickup. I would use baffles to minimize cross flow, sloshing and prevent the oil pump pickup from taking potentially aerated oil in directly from the oil cooler pump discharge.
          @turbovarg
          '91 318is, M20B25, T3/T04E 60 trim (15psi), megasquirt, coilovers, Z3 rack, cold AC
          [b u i l d]
          [Car of the month: April 2018]

          0c8b7c9527af628a346878feb14bf757

          Comment


            #6
            Awesome, that's super helpful. One thing potentially doesn't make sense though...let me know what you think. This is the inside of the filter housing:



            The oil feed from the pump comes up through the broken check valve closest to the camera, so the outside of the filter is the dirty side. The broken valve on the right is the filter pressure bypass valve (probably responsible for killing the engine this came off of - the oil wasn't being filtered). The clean oil flows down the center tube axially around the cap bolt, and there are two rifles drilled perpendicular to this tube that connects to the port to the cylinder head, the oil pressure sensor, and the back side of that filter pressure bypass valve. All that stuff makes sense.

            See that little triangular port on the dirty side of the filter? That's what feeds the rectangular port that would feed the timing case with oil. It just seems strange that the timing case would be fed unfiltered oil when it would be a trivial machining/casting change to connect that to filtered oil instead. That's what led to the creation of this thread in the first place - I would expect one inlet feed, and all the outlet feeds to branch off of that, post-filter (or bypass valve). This port doesn't follow that logic.

            I also looked briefly at the sandwich plate idea, but you're right - there's just no room with the alternator there. My plan at this point is to cut the filter housing off the top and , provided there is enough space, have ports on the top of the housing.

            Project M42 Turbo

            Comment


              #7
              Good question. Since the timing case is all splash lubrication and there are no bearing surfaces (other than the ball bearing in the sprocket), they must have decided filtering isn't strictly necessary.

              It's clear in these images that the round hole only feeds the block, and there is a long passage above it in the first image (the case being upside down) which has a hole at each end; the hole at the right lines up with the rectangular passage.



              @turbovarg
              '91 318is, M20B25, T3/T04E 60 trim (15psi), megasquirt, coilovers, Z3 rack, cold AC
              [b u i l d]
              [Car of the month: April 2018]

              0c8b7c9527af628a346878feb14bf757

              Comment


                #8
                Awesome thread! Thank you for taking all this time to document the M42 oil supply system...finally there are detailed photos all in one place! I had never realized that the timing case was being fed dirty oil...while I suppose that it is not that big of a deal since there are no really tight interfaces to lube, I think that it is still sort of like playing with fire since the tensioner rail squirter could possibly get gummed up / plugged with RTV.

                Last time I fixed a leak on the oil filter housing, I noodled hard about gasket sealing and how to avoid crap in the oil passages. In the end I went with some Permatex high tack SPray-A-Gasket since it would mean a thin, uniform coating with minimal squeeze-off. I have used it other places, and it will lead to small ribbons of material forming and potentially ending up in the working fluid, but it seems like it should not clog up anything as large as the timing case squirters. Obviously you want to avoid any sort of RTV or spray sealant in the round hole for the main oil galley...I used some silicone assembly lube on the O-rings on the round port of the filter housing to prevent binding, and that seems to have stopped all leaks. As long as the walls of the round port are clean, leaks should not be an issue.

                Permatex spray stuff:
                https://www.permatex.com/products/ga...asket-sealant/

                Also, consider an M44 oil filter housing. It is a direct bolt on IIRC, and its overall diameter is a lot smaller than the M42's housing. That could buy you enough room for a sandwich plate. I am also running an M44 timing case since it has thicker gear pump rotors (more oil flow capacity), and you may want to consider one. As long as you get the M44 timing case gasket, it should bolt right up. There is one problem though, which is that there is no crank position sensor mount, so you would either need to machine or buy an adapter. Metric Mechanic should be able to sell you one since they provided one to me with my 2.1L M42.

                Good luck, keep the info coming! This is super useful stuff! I am also paranoid as shit now about the Spray-A-Gasket stuff being loose in the timing case lube holes lol.

                Comment


                  #9
                  I believe the later M42 timing case had the deflector rail instead of the bearing socket that fails. It should have the crank sensor mount.
                  -Nick

                  M42 on VEMS

                  Comment


                    #10
                    Yup, the 1994+ M42 had the updated timing case. But, the M44's oil pump rotors are larger (2mm thicker IIRC) which makes for higher oil flow.

                    Comment


                      #11
                      For my oil filter housing seal I used copper spray-a-gasket to stick the paper gasket to the filter housing and prevent any RTV from entering the oiling system. Seems to have worked fine for the past 15k miles!

                      Excited to see where this thread goes!

                      -NICK

                      Comment


                        #12
                        Originally posted by bmwman91 View Post
                        Yup, the 1994+ M42 had the updated timing case. But, the M44's oil pump rotors are larger (2mm thicker IIRC) which makes for higher oil flow.
                        Ah I didnt realise the gerotor size was different between the two covers! Thanks.
                        -Nick

                        M42 on VEMS

                        Comment


                          #13
                          Originally posted by varg View Post
                          Good question. Since the timing case is all splash lubrication and there are no bearing surfaces (other than the ball bearing in the sprocket), they must have decided filtering isn't strictly necessary.

                          It's clear in these images that the round hole only feeds the block, and there is a long passage above it in the first image (the case being upside down) which has a hole at each end; the hole at the right lines up with the rectangular passage.
                          Thanks for the pics! That's very informative. I have started cutting up my spare oil filter housing and believe I have found a way to drill a hole through the casting to the clean side to get clean oil flow to the timing cover. I'll post again once I figure out the orifice size and actually drill the hole. Then, I will just weld the triangular hole shut and that should be all set.

                          Originally posted by bmwman91 View Post
                          Last time I fixed a leak on the oil filter housing, I noodled hard about gasket sealing and how to avoid crap in the oil passages. In the end I went with some Permatex high tack SPray-A-Gasket since it would mean a thin, uniform coating with minimal squeeze-off. I have used it other places, and it will lead to small ribbons of material forming and potentially ending up in the working fluid, but it seems like it should not clog up anything as large as the timing case squirters. Obviously you want to avoid any sort of RTV or spray sealant in the round hole for the main oil galley...I used some silicone assembly lube on the O-rings on the round port of the filter housing to prevent binding, and that seems to have stopped all leaks. As long as the walls of the round port are clean, leaks should not be an issue.

                          Also, consider an M44 oil filter housing. It is a direct bolt on IIRC, and its overall diameter is a lot smaller than the M42's housing. That could buy you enough room for a sandwich plate. I am also running an M44 timing case since it has thicker gear pump rotors (more oil flow capacity), and you may want to consider one. As long as you get the M44 timing case gasket, it should bolt right up. There is one problem though, which is that there is no crank position sensor mount, so you would either need to machine or buy an adapter. Metric Mechanic should be able to sell you one since they provided one to me with my 2.1L M42.
                          When I fixed my oil filter housing a few years ago, I did the same thing. Spray-on gasket sealer and some silicone lube on the oring. Worked great for me as well.

                          M44 oil filter housing would also work, but at this point I've abandoned the sandwich plate idea and will be pursuing direct modification of the housing itself. The M42 housing should work fine for that.

                          I didn't realize the M44 had a larger pump. I guess it must have something to do with the different valvetrain; maybe it had different oiling requirements. Definitely good to know. If/when I have it apart again, I will do that, but I rebuilt the whole timing case about 50k miles ago so there is very little motivation for me to take it all apart again right now. In any case, my oil pressure is still more than sufficient, even when hot after pulling off the track.


                          I've got a work trip coming up that's taking some priority over this project, so there won't be much progress for a week or two until I get back and can get bungs and fittings ordered and in the mail. One way or another though, I'll come up with something. Necessity is the mother of invention, and the high oil temps have definitely made this a necessity.

                          Project M42 Turbo

                          Comment


                            #14
                            Well, this pretty much has to happen now. I'm officially into the 4 digits of $$ making this work....but here's to hoping I only have to do it once.

                            Improved Racing suckered me in with some of their badass parts:



                            Integrated remote oil filter and oil thermostat. I went with the 215F option, since the car sees enough street use still that it does still need to get warm.

                            Cooler will be a Mocal 34 row, 115 matrix cooler that is narrow enough to offer low pressure drop but tall enough to still have good surface area. It will mount between the intercooler and the radiator core, where the AC evaporator core normally would be.

                            Add 8 AN fittings, a bunch of line, and some adapters...this stuff adds up fast. But hopefully will have some progress worth taking pictures of soon here. And hopefully it will all go together well and be leak-free. I have high hopes, after measuring everything and mocking some stuff up.
                            Last edited by gearheadE30; 06-20-2017, 04:49 PM.

                            Project M42 Turbo

                            Comment


                              #15
                              I've always used -10 lines, on the theory that bigger can't hurt,
                              whereas smaller can't help. There's no practical difference in
                              cost- it IS harder to route them, though.
                              This is on M10s and M20s...

                              That IS a very pretty solution to a difficult problem.
                              Like you, I've found that done right, oil coolers are ex-pen-sive...

                              And, even on track, the thermostat's nice- the oil comes
                              up to temp a bit quicker, so you can get on it harder, earlier...

                              t
                              now, sometimes I just mess with people. It's more entertaining that way. george graves

                              Comment

                              Working...
                              X