(Image: Header Graphic)

Wednesday, October 18, 2017

RSS Feed

Doug's Domain

Doug Vetter, ATP/CFI

dvatp.com has been updated!

June 22, 2016: Donations to dvatp.com are now processed via Stripe. Like this site? It's easier than ever to show your appreciation.

Saturday, April 15, 2017

More Engine Overhaul Research

Piston and Rod Options

It turns out the manufacturer of the MaxSil pistons, Bavarian Engine Exchange, went out of business at some point in the last six months. You'd never know it as their websites are still live at this writing, but when I called the Bavarian number (which is one off the MaxSil number), I was connected to an auto salvage company of some sort that told me of Bavarian's fate and the fact that they have nothing to do with the MaxSil piston line. So the MaxSil pistons are no longer an option.

After yet more research I've concluded I have exactly two piston options for a more or less stock build:

4032 is a high silicon alloy (~12% silicon). The coefficient of expansion of 4032 exceeds that of a cast piston like the BMW/Mahle or the defunct MaxSil units but 4032 allows tighter piston to cylinder bore clearances than the 2618 (0.2% silicon) alloy used by CP-Carillo, among others. For a street car, tighter clearances translate into less bottom end noise (piston slap when cold) and generally longer service life. Unfortunately, no one stocks JE's M52 4032 product but I did find one supplier (racetep.com) that will custom order them for $1150. As that is half the price of the OE pistons and they are made in the US I am now quite likely to go the JE route. The downside? A three week lead time.

Minimum clearances on the JE 4032 pistons are 0.002 to 0.0025 inches (2 to 2.5 thou). Metric Mechanic builds their engines with a minimum clearance of 0.001" (1 thou) over the stock cast piston clearance numbers. If I use the MaxSil numbers (0.0015", 1.5 thou) as a starting point, that puts the minimum clearance at 0.0025" (2.5 thou). Metric Mechanic also recommends honing the innermost cylinders (3 & 4) to the minimum spec and then adding 0.0002" (2 ten thou) for cylinders 2 & 5, and another 0.0003" (3 ten thou) over that for the two outer cylinders (1 & 6) because the outer cylinders run cooler than the inner cylinders. At this point I'm not convinced I'll add the extra clearance mostly because I know of only one company that does this (MM) and because I imagine it will add to the machining time and cost, but it's under consideration.

While looking at the ETK I discovered that BMW sells rod bushings for about $20/ea, which means I should be able to recondition my rods by having the machine shop check them to ensure they are straight, and then press in new bushings. Then I'll have the shop check and hone the rod bushing to the piston pin as needed. In a related note, I also noticed BMW sells undersize rod bearings in addition to the stock units, so if my crank has to be polished I can maintain the ideal 0.001 to 0.0015 bearing clearances by installing one of the two available sizes.

Should I find my rods bent I could justify buying up to two new rods ($350 each) from BMW before justifying the purchase of an entire set of aftermarket rods ($1350). If I am forced to go this route I will buy Pauter rods and call it a day. Pauter rods are 30 grams lighter (520) than the factory pieces (550) and have a lower windage design. Both characteristics should translate into slightly more horsepower, however I'm under no illusion the performance gains will justify the price in and of itself. In any case, this is a fallback option only, so power increases are not the objective. The bonus is, like the JE pistons, Pauter rods are entirely made in the US.

Torque Wrenches

On the M52 the main and rod bolts are torqued in a sequence of 20 Nm and then 50 degrees. This means I'll need a way to measure angle in addition to torque. When I bought my set of mechanical SK torque wrenches there were digital options available but I decided to keep it simple, mostly to keep costs down, but also to avoid hassles with fragile new technology that I knew would be obsolete tomorrow.

Watching 50sKid's rebuild I was reminded that mechanical click-type torque wrenches are either not functional in the reverse direction or significantly less accurate. The end result is that they cannot be depended upon for applications involving reversed threads like the oil pump nut. I also learned the Eastwood electronic torque wrench he used, despite having a torque angle feature, uses a mechanism to measure the angle that is highly dependent on operator technique. So I'm not inclined to spend $100 only to get something that has about the same accuracy and (more importantly) consistency in measurement as a $20 torque angle meter.

The alternative, of course, is the "BMW of torque wrenches" -- the SnapOn units. By all accounts these things work perfectly. In addition to standard torque and angle measurements, these units:

Much like my 3/8" drive SK mechanical unit the SnapOn equivalent will do 5-100 ft*lbs. This means it will safely do the rod and main bolts as well as the oil pump nut.

The issue, as always, is price. SnapOn is in love with their products and this is no exception: it's a bit over $500. That's over twice as expensive as my most expensive SK mechanical click type torque wrench. Considering that I never again plan to do a job requiring torque angle measurement, my options are to go with the torque angle meter and accept the added time this will require, or buy the SnapOn product and resell it when I'm done. I am leaning toward the latter scenario because I'll have peace of mind and get most of my money back in the end.

Outside Micrometers and Bore Gauge

In order to blueprint the engine I'll need a set of outside micrometers and a bore gauge to measure a range of parameters from roughly 1.5 to 3.5 inches. Each outside micrometer operates over a mere 1" range so I have to buy several that will match the following use cases:

The bore gauge naturally needs to be able to measure down to approximately 1.7 inches and have sufficient resolution to measure down to one ten thousandth (0.0001) of an inch. Needless to say, not all gauges do that.

While most bore gauges are analog, I am hoping to buy a digital bore gauge. These have the obvious benefit of direct reading from an LCD but also have a "minimum" function that automatically picks up the minimum measurement, which occurs when the tool is perfectly aligned to the bore. This eliminates a lot of measurement inconsistency that stems from the need to read the gauge while moving the instrument.

If you're wondering -- even after I have used these fancy gauges to measure all the parts prior to assembly, I am planning to use plastiguage as a sanity check / backup to absolutely ensure that the bearing clearances are correct.

Compression Ratio (CR) Calculation

After searching for the requisite data on bimmerforums, I've come up with some numbers to use in calculating the CR for the original engine configuration (84.0mm bore) and compare that with the 84.5mm bore so I can figure out how much the CR will increase as a consequence of boring out the block. To calculate these numbers the following volumes are needed:

  1. Combustion chamber volume. There is only one accurate way to get this and it involves some apparatus and a liquid to fill the combustion chamber to the face of the head. This was obtained experimentally by others. For the M52 head this is 36cc.

  2. Head gasket volume. Someone measured the factory gaskets and found the cylinder openings are 85.0mm or 1mm over the stock bore size. The stock composite gasket thickness is 0.071" or 1.8mm. A slightly thicker gasket (0.3mm or 0.012") is available to address cases where an equivalent amount of material is removed from the head and/or block. The volume is obtained by calculating the volume of the resulting cylinder which is 1.8mm high:

      πr2h
      π * OpeningRadius2 * Thickness
      3.1459 *42.52 * 1.8
      (3.1459 * 1806.25 * 1.8)/1000 = 10.23 cc
    
  3. Clearance volume. This is the gap between the top of the pistons at TDC and the deck of the block. If the top of the piston face rests below the deck, as is the case with the M52, then the clearance volume is added to the total volume. In the M52 the pistons are 1.88mm below the deck at TDC.

      πr2h
      π * BoreRadius2 * Clearance
      3.1459 * 42.252 * 1.88
      (3.1459 * 1785.1 * 1.88)/1000 = 10.56 cc
    
  4. Piston face volume: This is the volume of any reliefs or domes present on the piston face. In the case of the M52 the stock pistons are completely flat with four small reliefs milled into the face to accommodate the valves (the M52 is an interference engine).

    This value must be found experimentally as BMW does not provide this information. I had to dig for a while to find this number but apparently the stock piston reliefs are worth a mere 0.3cc.

    Reliefs add to the total volume in the cylinder while domes subtract from it. For brevity I am not including this in my calculation, but I must advise that if you are building something custom you'll definitely want to take the shape and hence volume of the piston face into account.

  5. Swept volume. This is the cylindrical volume obtained by using the stroke of the engine as the height and the bore. The M52 is a "square" engine, meaning its bore and stroke are the same out of the factory (84.0mm bore, 84.0mm stroke).

    We need to calculate two swept volumes -- one for the stock (84mm) bores and another for the reconditioned (84.5mm) bores.

    Stock:

      πr2h
      π * BoreRadius2 * Stroke
      3.1459 * 422 * 84.0
      (3.1459 * 1764 * 84)/1000 = 466.15 cc
    

    Reconditioned:

      πr2h
      π * BoreRadius2 * Stroke
      3.1459 * 42.252 * 84.0
      (3.1459 * 1785.1 * 84)/1000 = 471.72 cc
    

The static CR is the ratio between the volume in the cylinder at bottom dead center (BDC) as compared to top dead center (TDC):

  CRstatic = Vbdc/Vtdc
  Vbdc = SweptVolume + CylinderHeadVolume + GasketVolume + Clearance [(+/-) PistonFaceVolume]
  Vtdc = CylinderHeadVolume + GasketVolume + Clearance [(+/-) PistonFaceVolume]

Hence, for the stock CR (84mm bore) we have:

  (466.15 + 36 + 10.23 + 10.56) / (36 + 10.23 + 10.56)
  522.94 / 56.79 = 9.2:1

And for the reconditioned CR (84.5mm bore) we arrrive at:

  (471.72 + 36 + 10.23 + 10.56) / (36 + 10.23 + 10.56)
  528.51 / 56.79 = 9.3:1

You may notice that the calculated CR (9.2) is significantly below what BMW specifies (10.5). While it is generally accepted that BMW over-reports CR on its engines I have no explanation for this much of a difference. However, for my purposes, the absolute value is of little concern. I have answered the question I set out to answer, which is "what is the effect on compression ratio when boring out the cylinders from 84 to 84.5mm?" The answer is 0.1 or, in a word, inconsequential.

The point to take home is even if start with BMW's specified CR, 10.6 is well below the 11.0 often assumed to be the maximum for safe operation on AKI 93 octane pump gas. Will I have to run 93 octane? No. I typically run 91 now because I have found running 89 results in worse gas mileage, no doubt because knock is occuring and the timing is being retarded to compensate. So the good news is I won't need to change my fueling practices once the engine is overhauled. Going into this I assumed that would be the case but it's nice to see that the data supports the assumption.

The only other potential impacts on CR will involve the reduction in combustion chamber volume as a result of milling the head and the reduction in clearance when the block is milled. But I won't be able to calculate the new volumes until I get the head and block back and decide whether to use a stock thickness gasket or a thicker unit.

Consult with Autohead Performance

I recently spoke in detail with Peter at Autohead Performance about my head rebuild.

The first question I asked is whether he would recommend replacement of the springs. His response was that in his experience springs will "last a lot longer" than 265K. His standard procedure is to test the springs to ensure they are still producing the required force and reinstall them if that is the case. He added, however, that he will install new springs regardless if I am willing to provide them. I am inclined to replace the springs because they have a nasty tendency to work until they fail catastrophically, putting the entire engine at risk. While I must consider the possibility of infant mortality in new springs, I think the risk of failure is higher overall in the medium to long term (25K+ miles) with the existing springs. Given the cost ($360) I think it would be a fool's errand to leave the existing units in service.

The next question was related to the need to ream or replace the valve guides. He said that of the hundreds of heads he's reworked he's never replaced any guides. Considering that I have heard of many engines with half the time as mine needing at least the exhaust guides reamed or replaced, Peter's comment surprised me. I emphasized the mileage on the engine and he said he would naturally check the guides and advise me if any work needed to be done. He also suggested that very close valve guide clearance is actually not preferred because the gap naturally closes up as the head heats up. This was not exactly news, but it would not negate the need to replace out of spec guides in any case.

Mostly as a curiosity I also asked about the check valve that keeps oil up in the head after shutdown to minimize the time the cams run "dry" on the next start. Specifically, I wanted to know if this was something that needed to be replaced or simply cleaned. His answer: it's cleaned. It's nothing more than a simple spring and a ball bearing, and suffers no real wear.

Finally, I spoke to him about replacing the lifters. He said that if I wanted to replace those items he'd gladly install them, but generally speaking they are just cleaned and reinstalled in the same location in the head. He then asked me if there is any valve tap and I had to admit yes, for a short time after initial start on some cold mornings I have been noticing a bit of tapping -- what I think is from a single cylinder. He suggested that if I didn't want to replace all the lifters that I should get a mechanic's stethoscope, start the car, and attempt to localize the tick to one or two cylinders. That information would help to find the defective / clogged lifter and hence reduce the number of lifters we need to replace.

After our discussion I found that Pelican Parts is a source for INA (OEM) lifters. The price is roughly $15, which is $3 more than the piston alone and less than half the price of BMW OE units. That's $360 for all new lifters. I think that's worth the money to keep the valve train quiet and the cam lobes safe from spalling so I'm planning to replace the lifters.

Mileage: 265500