Overcoming Dysmetria!

Well here I must express my apologies to true engineers- stop reading here, this bit isnt for you. This is where grapple with my inability to measure things accurately and the very tight limits of tolerance given by Lotus in the workshop manual. Since my problem is practically clinical I have coined the term dysmetria!  I am also concerned that the figures in the Lotus tech data section might  not be serviceable wear limits but tolerances in original manufacture so how worn is too worn?
Anyway this blog will be of no interest to the serious and experienced mechanic but it will at least document my struggles with measuring devices and the reasoning behind my decisions (be they right or wrong!)... how hard can it be?

Cylinder nip:
This is the height to which the cylinder liners project above the crankcase surface. You can measure it using a straight edge across the cylinder and a feeler gauge between that and the block. Or you can use a depth

Using a depth micrometer through a hole in the straight edge,
its tilted for photographic purposes- obviously use it vertically.

micrometer measuring through a block also positioned across the liner. I did this both ways, feeler gauges gave a higher value of 1 thou for cylinder 1 and 1.5 for the rest. The Micrometer gave 0.5 thou for no 1 and 1 thou for the rest.
Liner nips (if not the same) should show a smooth gradation from back to front, ups and downs will prevent the head gasket from seating, this appears to be fine. The cylinder liners are magnetic so these are presumably the original iron liners and for these the manual cites values of  1-5 thou for engines after 12478. If the feelers are right then I am in spec, if not then its tight on the low limit. I must check that I haven't depressed the liners with over tightening of the liner clamps. 

Mains and Big ends.
No pictures for these but here is my experience.
Direct micrometer reading on the journals was difficult, it tended to jam, it wasn't easy to see whether you were at right angles and truly across the diameter. This seemed just an easy way to scratch the nice polished surface! Its probably much easier if you have a way of mounting the heavy and cumbersome crank on the bench so it can be easily rotated and examined.
The solution I found was to use engineers' calipers/dividers which were light and easily set to the journal diameter, they were great for testing roundness. The calipers are then measured with a digital caliper gauge. This worked really well. All my journals came out at the expected diameters allowing for the regrinds inferred from the shell markings and there was no detectable out of round at all. Good news here then!

Bores- using the bore gauge.
Setting the gauge: 
Chose an extension anvil of a length close to that you wish to measure- I used 95mm. Insert the gauge at the top and depress it to give one full revolution and clamp it in place. Check that the combined anvil and gauge base is longer than the dimension you wish to measure. 
In order to compare cylinders it doesn't really matter what you set the gauge on, but for actual measurement the gauge must be set up. This is difficult without a setting ring (which is prohibitively expensive) so I have to use a micrometer. In this case set to 95 mm and clamped lightly in a vice for support (I have too few hands for this job). Set the gauge between the micrometer pins using the sprung pip between the wheels and the anvil tip. Even though the surface area of the pins is small, its still essential to rock the gauge to make sure you know where the needle hits its minimum. This is 95 mm so you can set the gauge ring to zero at that point and lock it off with the clamping screw. This is fiddly!

Use
You can then insert the gauge to the required depth in the bore, orient it in the required direction and rock it gently along the plane of the anvil whilst watching the gauge for the minimum reading. 0 represents  95 mm, so reading from zero to the needle maximum deflection gives the difference between the actual bore and 95 mm. Readings counted from zero to the left are oversize, those to the right undersize, provided the needle has passed through the zero (95 mm) point. If it hasn't then they are still oversize so read them from zero anticlockwise. Watch out for the full rev counter although if the gauge is set close to the measurement to be taken  this may not be needed. 

The bore gauge is very accurate and thus tends to generates different readings every time! For reproducibility of  readings its essential to put it in the same position for each measurement. This is tricky. The manual states manufacturer's values measured in the thrust plane (i.e. 90 deg to the crank and in the plane of conrod movement) at 50 mm depth. I drew markers on the gauge 50 and 100 mm up from the measuring tip and aligned these with the cylinder top to take the measurement. My gauge is marked to 0.01 mm per division, I estimated 0.005 by eye when the needle was between divisions.

Bore gauge measuring 90 deg to thrust direction, note depth marks on stem

I would be lying if I said that this was an unqualified success. However I did get a consistent picture. I measured at parallel to the crank, 90 deg to crank and 45 deg to crank in each orientation, i.e. 4 measurements at each point. In all cases the thrust dimension was the smallest at each level and the measurement at 50 mm down was smaller than that at 100 mm in all cases except cylinder 4 when there was virtually no difference at the two levels. Taking the largest recorded value for the bore in each cylinder thus means using the 50 mm, thrust plane measurements which were:
Cylinder 4, 95.355; Cylinder 3,  95.345; Cylinder 2, 95.34 and Cylinder 1, 95.34 mm. The manufacturers spec gives max values of  95.288 and 95.308 for a 912 LC motor depending on whether the liners are type "A" or "B". Pistons and liners come as a matched pair and I have "B" pistons (marked on crown), therefore "B" liners. This means my bores are worn by my "best guesstimate" by 0.047, 0.037, 0.032 and 0.032 mm or 1.9, 1.4, 1.3 and 1.3 thou. Lotus took the view that something was "in spec" or not, they didn't specify wear limits or  give regrind suggestions for the crank; it was right or it wasn't. However most UK motors of the time were more tolerant and a rough guide of 2 thou per inch of bore diameter was tolerable. In fact a typical first bore of say  an Austin A series motor would be 20 thou and by these criteria these cylinders are still serviceable. Remember, the aim here is to produce a serviceable road car not a racer or track car so I'm hoping this will be sufficient. 

Thrust washers.
Rear Thrust washer was clearly worn more than the forward washer.

Rear Thrust (copper side)

Front Thrust washer (Copper side)

Both were measured with a micrometer, front 2.33 mm and rear 2.29 mm. I don't currently know if they both had the same diameter when fitted but it seems likely as the extreme tip of the rear washer was thicker at 2.32 mm. If so then wear would account for an extra 0.04 mm (0.0015") end float. End float I have found  to be right on the max with these washers fitted, so it may be sufficient to simply replace them with new standard sized rather than oversized. Rapid wear of the  thrust washers was a problem with early motors and required a modification (see post: modifying main bearing panel). I will add a pair to my shopping list.

Piston to bore clearance.
This motor has type B pistons corresponding to the design used in 912 Std compression engines, compression ration 9.44:1. Since the pistons have to work in the bore the clearance between them is an important determinant of wear. I am taking this as the difference in bore as measured perpendicular to the crank at 50 mm depth and the piston diameter measured at the widest point which is the skirt. Pistons may vary in diameter at different points to allow for expansion and its not stated at what point the diameter is measured. I measured them both directly using a digital caliper and indirectly with engineers' calipers.  These were very useful to check the diameter around the circumference. I could then measure the calipers using a digital measuring caliper. The measurement was cross-checked with a 100mm micrometer measuring directly on the piston skirt app 10mm up from the lower edge.

Pistons measured at the skirt (mm); 
Piston No: 4, 95.14  ; Piston No 3, 95.14; Piston No 2, 95.18  ; Piston No 1, 95.19
Calculated clearance: Cylinder No 4, 0.215 ; Cylinder No 3, 0.205; Cylinder No 2, 0.16; Cylinder No 1, 0.15
The manual gives the values for piston-to-liner clearance of 0.11-0.15 mm. From these figures Cylinders 1 and 2 are almost within tolerance of 0.11-0.15 mm but 3 and 4 are over. However as with other places in the manual its not clear whether Lotus are providing figures for the tolerance in manufacturing or permitted wear before service is required, and again I'm up against the "as new or replace" doctrine which I just cant afford to do. Taking my cue from Brit cars of the same period, then wear of up to 3x the original spec was permissible and again mine are well within this. Piston measurements are hardly down from stated diameters and I think the existing liner/piston setup is reusable.

Piston Skirt 
Since the piston gets hotter at the top than the bottom, the crown expands more than the cooler bottom of the skirt. For this reason and to ensure that the piston is almost straight when at temperature, the piston is made to be tapered, being wider at the bottom because this will expand less than the top. This skirt flare is detected by an increase in piston diameter from pin level to skirt base. However, if a piston has suffered overheating then the skirt may expand and contact the cylinder wall. When this happens the  piston is forced into  a straighter configuration and does not contract back to its original shape when it cools: the larger diameter at the skirt is lost. Termed "skirt collapse" it can be eliminated by checking that the skirt diameter is greater than the diameter at the pin. In my case this was true in all 4 cylinders. It is possible to check the angle of this flare (grade) but  I was happy to find that skirt collapse wasn't a problem in this motor.

Piston Ring-to-Groove Gap
This is cited for a Std compression 912 as 1.5 - 3.5 thou measured by slipping a feeler gauge in between the ring and the groove land. All cylinders were similar and all gave a higher value of  6-7 thou for ring one and a more likely value of 4 thou for ring two which is within the specified range (just!). A larger than expected gap could be generated by a thinner ring or wear in the piston groove. I was expecting rings one and two to be the same thickness because only one gap is quoted in the manual for both compression rings and piston groves are the same width. However this difference in observed gap size was practically all accounted for by a reduced thickness in ring 1 since when each ring  was removed and measured with a micrometer it was clear that ring 1 (1.52 mm or 0.059 ") is  thinner than ring 2 (1.57 mm or 0.062") by 3 thou. When the actual thickness of each ring is added to its clearance measured by the feeler gauge it generates the groove width, which is practically the same in both cases, supporting the conclusions above that the pistons are probably not significantly worn and still usable if not perfect. Its not clear whether this difference in ring-to-groove gaps is due to similar wear in all 4 top rings, or simply the fitting of a replacement ring set that had thinner top rings to start with! I think these are unserviceable whether through wear or incorrect installation and so it seems likely that new rings would be a good idea.

So Conclusions to this measurement section:
Replace all rings
Fit new thrust washers (Std thickness)
Hone all cylinders.
Fit new big end and main shells- probably not strictly necessary but seems a shame not to do so whilst I am here!
Not perfect but hopefully all serviceable.