No announcement yet.

e30 M3 minor rust repair.

This is a sticky topic.
  • Filter
  • Time
  • Show
Clear All
new posts

    I've read this restoration thread on s14 at least a dozen times... I did not think to look for it here on r3v, but I'm glad it's here, as well. As many have said prior to this post, this is absolutely brilliant! Probably the single most informative and thorough thread ever written and your writing style is as real as it gets. I read and imagine standing next to you as you re-tell the story of having a friend's hand to use instead of your own hand while re-assembling the diff, so as not to smash your own, Lol! Well done my friend and I look forward to reading more, as soon as you are ready to inform us some more! Too bad we live an impossible distance apart, I'd love to drink a pint with you and see your work in person!!!




      Thank you for the kind words, much appreciated.
      I'm currently sorting through the pictures and uploading
      them to the interweb for the next update, should be done
      within a week if my typing finger holds up. ;)


        I'm as excited as an 18 year old HS chick the day before prom


          Originally posted by CorvallisBMW View Post
          I'm as excited as an 18 year old HS chick the day before prom
          I am as well... as Thomas Dolby wrote, "little rivers of anticipation are running down my inseam", Lol!



            Is she back together yet?
            Much wow
            I hate 4 doors


              Well, how are ya? Good? Excellent! It's been a while since we last talked
              and I've a pile of shite here to bore you with. Believe it or not I still haven't
              managed to bolt together this bloody engine. However, I have just finished
              all the prep work and need to empty all this crap out of the camera before
              it bursts at the seams. So...... how the hell have I managed to spend the last
              six months at this and still not built a simple engine? Well, it would appear I
              possess a rare talent, the ability to drag out each and every task to its absolute
              maximum length, however, in my defence, as this story unfolds you'll see
              this river hasn't run smoothly.....

              If you can remember all the way back to the time when Jesus was a young boy and I started
              the bodywork on this restoration, you may remember that the preparation phase of the
              bodywork took a lot longer than the actual spraying part, well, engine building is much
              the same. While you could probably assemble an engine in a fairly short space of time
              if you have all the parts in front of you, checking, measuring, cleaning and preping the parts
              beforehand takes a great deal longer. But, just like the bodywork, if you don't spend
              the time on the prep work the end result will almost always suffer.

              One small point before we go any further, I'm not going to go into the minute details of
              engine building in this thread for two simple reasons.....
              a) theres a wealth of knowledge and fantasticly detailed build threads out there of how
              to choose the right cams, pistons, valves etc. for your specific engine build and I'm far too dumb
              to try and add to that knowledge.
              b) what does seem to be scarce out there is some explanations of the simpler
              stuff for people who may never have built an engine before,
              like what to check to see if parts like the block, crankshafts, conrods etc are fit
              to be reused again. If you've built your share of engines in the past then it's unlikely
              you'll pick up anything useful here, however, if you've never built an engine before
              then hopefully........ you'll fail to realise that most of whats to follow is,
              as usual,
              probably wrong!

              First up, a quick word on measuring. When your looking up spec's for wear tolerances
              or shopping for measuring tools then you'll most likley come across the two different
              units of measurement common in the trade. Metric and Imperial.
              Things usually tend to be a little less confusing if you choose one type and stick to it
              for all your measuring. In this regard the various overhaul manuals from BMW for our
              cars are well laid out and almost always give all important measurements and
              tolerances in both millimetres and thousands of an inch.

              Below you can see a little diagram of how each of the units of measure relate to each other.
              The large blue circle on the left represents 1 thousandth of an Inch (1 thou) or 0.001" as you'll
              usually see it written. To try and give some perspective to this measurement, the average human
              hair is 2 thou thick. For most of the tighter tolerances in this engine build we'll be measuring
              down to ten thousandths of an inch, 0.0001", thats 20 times smaller than that hair you just plucked
              and it's represented below by the smaller of the two blue circles.........


                The other two red circles above represent the similar metric sizes, a hundredth
                of a mm and a thousandth of a millimetre.
                In the following waffle I tend to favour Imperial measurements using Thou and
                Ten Thousandths of an inch, however, it would appear I also switch over to
                millimetres at times as well, because, just between you and me, my brain is
                The reason I'm telling you this, is if your going to be using any of this info to
                check your own engine parts (very brave) then you need to be careful of
                what units are being used.
                The following link is to an online converter to change between thou and
                mm and can be quite handy to have....
                One other final link before we get started and thats to BMW's
                Tolerance and Torque manual, all the data you'll need for figuring out
                working clearances and so on can be found here.....
                Click "Contents" then "engine" and then choose your weapon of choice.

                Right, all that shite done with, it's on to the actual engine bits, starting with
                the block. After the engine was stripped, the block was "hot tanked" and
                steam cleaned in every orifice to remove all the gunk that had built up over
                the previous 20 odd years........

                next up was to check the flatness of the deck face. This is the surface the
                headgasket has to seal against when it's sandwiched between the block and the
                cylinderhead, and as such it needs to be perfectly flat and smooth.
                To check this we use an engineers straight edge which is basically a very straight
                piece of tool steel. This is placed across the deck and you check to see if a 1 thou
                feeler blade will fit in anywhere underneath indicating that the surface isn't quite flat enough.......

                The block needs to be checked in all the following directions......

                As you can see above my deck face looks suspiciously clean and flat and
                thats because when disassembled at the very start of this restoration the
                surface was found to be a bit uneven and as a result I had the local machine shop
                surface grind the deck 0.002" to flatten it out. Unfortunately this was so long ago
                that cameras weren't invented and I don't have any before pictures to show you.


                  Another job that was carried out at the machine shop was the block was "crack tested".
                  Basically this involves the use of a strong electro magnet, some coloured metal dust and an
                  ultraviolet light. This process should show up any stress cracks in the block and
                  stop you spending any more money on preping it because it's most likely just
                  been relegated to boat anchor status. If your curious to see how the process works
                  look up "Magnaflux" on youtube.

                  Next up is the cylinders, the holes in which the pistons are to travel in. As could probably be
                  expected the bores on my block were fairly well worn from over a hundred
                  and seventy thousand miles of one careful owner and five wanabe Arton Senna's.
                  Usually on a high mileage rebuild like this the block will need
                  to be "re-bored", the reason being the bores are so worn that to refit the
                  same size pistons again would leave them wobbling around like a stick in a
                  welly boot. There are oversize piston sizes available from the main dealer to deal with
                  this situation
                  standard size is = 93.355mm/3.675"
                  first oversize is = 93.555mm/3.683"
                  second oversize is = 93.755mm/3.691"

                  Your job is to decide which piston will you need. Common logic dictates that you should only
                  go as large as you need and therefore try and leave one more oversize piston to "bore to"
                  down the line should the block need another rebore at a future rebuild. However, you may not
                  have the luxury of this choice if your cylinders are badly worn. A good machine shop should be able
                  to advise which size the block needs to be bored to, to remove all traces of wear and return the
                  cylinder to a factory fresh finish ready for new pistons. As mentioned previously, I had my block at the
                  machine shop a long time ago (in an effort to speed up the rebuild, seems funny now looking back)
                  and the resulting wear meant that I had to get my block bored out quite heavily to return it to a usable state.
                  However, there is one slight difference with my specific build, and that is I am intending to
                  use custom pistons. These would be made to my own stated sizes and so all I had to ask of
                  the machine shop was to rebore all 4 cylinders to a common size of there choice.
                  In an ideal world you would collect your block from the machine shop and take his word as to what the
                  new cylinder size was, however past experiences has taught me never to accept someone else's word
                  on important things like this. You've got to measure for yourself !

                  So, next step was to measure the cylinder bore so I could start to calculate what size pistons would be needed,
                  and also to check the quality of the rebore. In the pic below you can see the two axis that measurments were taken,
                  X and Y, and these pair of measurements were also taken at 3 different depths in each cylinder.
                  Depth A = about half an inch down from the top
                  Depth B = halfway down the cylinder
                  Depth C = about half an inch up from the bottom of the cylinder

                  To take these measurements we use whats called a bore gauge, which looks like

                  The gauge above comes with a range of different attachments to allow it to be
                  used in measuring a large range of different hole sizes. This specific one reads in
                  0.0001" increments and is fairly accurate. When it comes to buying tools for measuring
                  engine parts it really makes sense to try and buy the best you can afford, this is one item
                  where buying cheaply usually ends up costing you more money down the line as your
                  engine shits itself due to bad measurements made by cheap tools. (the down side is if
                  you have decent tools you've less things to blame it on when it all go's pear shaped.)

                  So, how does it work?

                  have a look at the picture of the gauge shown below. At the business end of the tool you
                  can see theres two ends (arrowed). Whats a little harder to see is that each of these ends has
                  a ball tip on it. The end marked purple is rigid and doesn't move. The end marked red does
                  move however, and as it is pushed inwards the needle on the dial gauge reads how much it's


                    The gauge on top has only a very small range of movement, this specific one only reads
                    50 thou from fully out to fully retracted, so you've got to choose the right
                    attachment (anvil) to put you in the ball park for the hole your measuring.
                    The cylinders on my block were all bored to 3.6950", so the correct anvil is
                    chosen and a micrometer is set up in the bench vise with a gap of 3.6950"
                    like so.......

                    and then the tips of the bore gauge is placed between the jaws
                    of the micrometer and the dial gauge is moved up or down in the tool
                    to get the needle zero'd on this measurment........

                    there's more appropriate setting block's out there to make it easier to
                    set the bore gauge up to a certain measurement and it's on the long list
                    of things to buy, right after a better f**king camera (don't hold your breath).
                    So, with the bore gauge zero'd on 3.6950" it's placed into the bore and rocked
                    left and right like so.........

                    Basically, you keep an eye on the needle and read off the lowest number it
                    shows while your rocking it at that point. The lowest number will be when the
                    gauge is straight and perpendicular in the bore, and thats the measurement of the
                    bore at that point. If the gauge read's 0, then the bore is bang on 3.6950", if it
                    will only drop as low as +05 then the bore is 3.6955" (3.6950"+ 0.0005") or if the
                    needle drops to the other side of zero, like -05 then your bore at that point is
                    3.6945" (3.6950" - 0.0005").
                    I know, I know, clear as mud isn't it? The only thing I'll say is when you have the tool
                    in your hand it's a damn sight easier to understand than what you've just read.

                    When your done you should have 6 measurements for each cylinder (3 X's and 3 Y's)
                    giving you a total of 24 measurements for the block, which should look a little something
                    like this......

                    So what the hell do I now do what all this crap you may ask?
                    Well I'm glad you did ask, because I just spent the last 30 minutes fighting this bloody computer
                    to get that Excel graph small enough so you don't have to be an orbiting astronaut to read it.
                    The answer is your now going to check these figures for cylinder bore "out of round" and "taper".

                    When you start to assemble your nice clean engine your going to fit a
                    nice round piston into each cylinder, and that piston will need to have a clearance between it
                    and the cylinder walls, too tight and when the aluminium piston starts to expand with heat it'll
                    scuff the bore or worse seize, too loose and the rings will struggle to seal against the walls of the
                    bore and you end up loosing compression and with it, power.
                    In an ideal world, you'd leave your block in and have it bored to 3.6950" and no matter where
                    you measured it you'd get 3.6950". Well, take a look at that list of measurements up there and you'll
                    realise thats not how the real world works.

                    So, your given tolerances and if your figures falls within these tolerances then everything should work
                    out ok. The first tolerance is for "out of round". Basically this is how oval your bore can be at any given
                    height and still be acceptable.
                    So, remember those X and Y measurements we took.......

                    If the cylinder was perfectly round both X and Y would be the same, however
                    the max "out of round" tolerance figure I'm using for this engine build is 0.0004".
                    That means the biggest difference between any of the 12 X and Y readings listed
                    above can only be 0.0004". I've a couple that are right on that limit of 0.0004"
                    but none over it, so I'm good to go.


                      Next up "taper". What we're after here is that the walls of the cylinder are "straight"
                      and not tapered like in the diagram below........

                      Again I'm using a maximum tolerance of 0.0004", so any of the X measurements
                      or Y measurements in the same cylinder can't have a difference of greater then 0.0004".
                      Again looking at the figures I'm good to go, the largest taper figure I have is in cylinder 4
                      on the Y measurements, Top Y = 3.6950" Bottom Y = 3.6953" a taper of 0.0003".

                      There was one other thing I wanted to check at this stage and that was the viability
                      of a "torque plate". In the picture below you can see the ten bolt holes that the
                      head bolts screw down into to secure the cylinder head to the block. These head
                      bolts take a fairly good squeeze to fully tighten them and what can happen on some
                      engines is that the metal around these bolt holes in the block can distort a little.
                      Depending on how close those bolt holes are to the cylinder bores this can sometimes
                      lead to the cylinders getting a little distorted. After you've just seen the tolerances we've
                      measured the bores to, you can see how this could be a problem......

                      so, how to check, well it mightn't be the most accurate way but it'll do for me.
                      Old head gasket fitted.......

                      Old cylinder head fitted and torqued down to the right torque........

                      and then flip the block over stick the bore gauge back in again to see if it made
                      much difference to the figures taken earlier........

                      The answer is I couldn't measure a big enough change in the figures taken earlier
                      to justify the cost of a torque plate. A torque plate by the way is a big slab of
                      aluminium that bolts down onto the block just like the cylinder head. Its job is to
                      distort the block just like the cylinder head might, the only difference being that the
                      torque plate has four big round holes in it allowing you to bore out the block while
                      it's fitted.......

                      typing finger sore, more to follow as the week goes on..........


                        Sweet baby Jeebus. Amazing thread that I can't stop revisiting.


                          F5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5 f5
                          Originally posted by george graves
                          If people keep quoting me in their sig, I'm going to burn this motherfucker down.





                              I heart you


                                Next up was the crankshaft. After it was fully cleaned and crack tested like the
                                block we could move on to measuring the bearing journals. The crank has 9 bearing
                                journals in total. 5 main journals which the crank spins on in the block (blue arrows)
                                and 4 conrod journals which the conrods spin on funnily enough (yellow arrows).......

                                First up is a visual inspection, any pitting, scratches or scoring on these journals
                                and they are going to have to be ground down to the next size. These need
                                to be as smooth as a new born baby's ass. Thankfully the old bearing shells
                                and the journals were in amazingly good condition upon strip down considering
                                the mileage this engine had on it. She must have had fairly frequent oil and filter
                                changes during her life........

                                Happy that there were no obvious signs of wear the next step was to measure each
                                journal carefully to make absolutely sure there was no machine work needed. For this we used
                                the next tool in the inventory, a set of micrometers.......

                                Again these read down to 0.0001" and thats the scale you'll need if your going to pick
                                up a set to do similar checks. First check is for "out of round" or "ovality" as it's
                                sometimes called. Again it's fairly similar to what we done in the cylinder bores, a measurement
                                is taken across the journal like so........

                                and then another measurement is taken at 90 degrees to this like shown in the
                                diagram below. What your looking for is the two measurements to be the same
                                indicating that the journal is round like the circle on the left, and not worn oval like the circle
                                on the right.......

                                Again just like in the bore measurements previously there is a tolerance your allowed.
                                For journal ovality I'm working to a max tolerance of 0.0004" between the A and B
                                readings shown above.
                                Last edited by xworks; 04-18-2011, 10:01 AM.