Nick vs the Engine, Part 1

Got all the bits back for the 3.6 bottom end from Dan the engine man yesterday and I was eager to get going on it. Since it's a 6 cylinder, I'm really keen to get the engine as close to perfect balance as possible, so I've had it balanced to 0.5g (much more than that is a waste of money, according to Dan, as the oil on the rotating assembly is random and may vary by up to 0.5g). I was still determined to do better, though!

The balancer had brought everything into balance individually (matched the lightest of each) so nothing was marked for a particular cylinder. I figured that if I chose each component carefully I could get an overall balance better than the tolerance of each piece. See what I mean? If one piece was 1g heavier than I needed and another was 1g lighter, the average would be spot on. It might (probably) wouldn't make a difference in practice, but I'd feel better having done the best job I could, and that's really all that matters, isn't it?

Jo got me a set of scales last Christmas with an accuracy of +/- 0.1g, max 2000g, just for this sort of work, so I could do the final component selection at home. It's bloody cold here for California, so I brought the scales and parts indoors and let them warm up for 12 hours or so before trying to measure them. I needed a solid and flat surface to work on, so I chose one of the kitchen worksurfaces as we have granite tile and that's perfect for the job.

If you ever want to do this, here's how to go about it...

Weigh everything individually and set them out so that you have the weight of the part written next to it and therefore you can easily pick up a part and know it's weight. You're going to get inaccuracies because of the accuracy of the scales you're using, but it will give you a start. I used sheets of shop towel laid on the worksurface and wrote the weight next to the part as I measured it. I piled parts of the same weight together and ordered the weights from left to right so I could pick the parts later without searching.

Use Excel to make a table of all the parts and weights. I had columns of piston, pin, clips, top ring, middle ring, oil control ring and expander, and rows of cylinder 1, 2, 3, 4, 5, 6. Sum up all the weights on each row so you get a total mass for an assembled piston. Add the lot together and divide by the number of cylinders and that will give you an average mass for an assembled piston, this is your target weight. Enter the weights in order, with the lowest weight at the top of the table and the highest weight at the bottom.

The table with all the weights. The "Set" column is purely arbitrary at this point, it doesn't refer to a cylinder number

If you've set up the table the way I did above, you'll now have a column that has the total mass of cylinders 1..6, with cylinder 1 being the lightest and 6 being the heaviest.

To get me started, I paired the piston and pin that came assembled together by moving the pin weights between rows to match the piston weight and pin. That mixed things up a little, but the total mass column recalculated itself so I could see my weight spread from the lowest to the heaviest. In my case, this was 2.0g.

Now comes the fun. "Swap" weights of bits around on the table to try to get the total masses column with the lowest difference. In my case I got it down to 0.5g after matching the light weight rings with the heavy pistons.

Once you've got it sorted on paper, make up sets of parts, based on the table you've been working on, and then weigh them. If you've got a nice set of scales, all your weights will match your calculations. If you've got a set of scales like me, the inaccuracies will add up and the measured weights won't match the calculations too much. My spread was at least down to 0.6g, but it wasn't as good as the 0.2g I had calcuated!

The next step is to use the table to work out which individual parts to swap between your sets to try to get the weights closer. I didn't have much I could do, but I did get it down to 0.5g.

Top left - swapping parts to get the closest measured weights. Below left - pairing cylinders to get the closest pairs, gives us cylinder allocations for the sets (middle upper column). Right column - adding rods into the equation and lower right is the final pairings for full rod + piston sets

We're on the home stretch. My next reasoning was that on a 6 cylinder (or a 4, come to that) pairs of cylinders move at the same time, so you want to match the weights of those pairs as close as possible to keep things in balance. Your target weight for this is 2x your calculated average weight per cylinder, of course. This is where things started getting better for me. When I paired cylinders up I got my calculated difference down to 0.2g (calculated by adding pairs of measured weights) and my measured differences (I weighed those pairs of sets of parts) down to a mere 0.1g.

The rods had to enter the equation at some point. They had all been balanced with the bit and little ends balanced separately and I had no reliable way to measure each end. I tried using a dial gauge stand to hold one end of the rod while I weighed the other, but when I added both ends together I didn't get the total rod weight, so I wasn't going to do anything reasonable there.

What I decided to do in the end was to weigh the whole rod and then match the rod up with the pistons to again get the closest spread possible. This was pretty easy as my first choice was the best - I matched the lightest rod with the heaviest piston set, then the next lightest rod with the next heaviest piston set and so on. The spread was a mere 0.4g, and that's with a total rod + piston set weight of 1255,0 - 1255,4g. Not bad. And even better was the pairing of cylinders - that was only 0.3g (1+6 was 2510.2g,2+5 was 2510.5g, 3+4 was 2510.3g).

That's as far as I'm going to go. I've bagged all the sets up together now to keep them together and I'll get on with measuring the crank clearances.