Assessing an unknown engine

[RobFrost]

My small reamers are hand type, 15mm or so taper on a 5/16" dia.

With head #6 I would be going to the guys in Nantwich to sort it out. The danger here is the guide centre line being some way off the truth of the valve seat centre line so making lapping the valve/seat really hard or impossible. Cylinder Head people will machine, re-guide, ream and reseat the valve to the seat in one precise go.

They have the jigs for the heads rob.

Plan is to get a good cylinder head on a jig, centre the mill using a 9mm bit in one of the guides, then put head #6 in its place and drill from there.

I guess headshop have a good jig to make sure both heads are in the exact same position. I do plan to use them for the twin plugging, if I go ahead with that.

Sent from my SM-S918B using Tapatalk

[RobFrost]

I centre punched the aluminium plug each end of the oil pipe.



And drilled first 2.5mm then out to 5mm, and cut an m6 thread into the plugs.



I put a little socket over with room for the plug to come out, screwed an m6 bolt in and tightened the nut to pull each plug out.





Sent from my SM-S918B using Tapatalk

[RobFrost]

They all came out nicely. Another enjoyable little job. Looks like I might need a new centre punch as this one is a bit off-centre.



The moment of truth -



I must admit I was expecting to find lots of aluminium in here... but it was all clean. Turns out I've wasted my time and created a job putting it back and re-plugging the ends.

But that's the price of peace of mind, I suppose. I'll spray them through with carb cleaner and re-fit. At least I know they're all clean now.



Sent from my SM-S918B using Tapatalk

[RobFrost]

The cams seem really snug and smooth inside the towers but it was time to measure them. I wasn't entirely confident in my micrometer as I read off the absolute value from the bore gauge in the cam towers, and it was telling me 46.98mm when they should have been 47.00mm. I don't know how accurately they make these cam boxes. No matter, I simply measured the cam journals and zeroed the bore gauge to the micrometer. Measuring the bores then told me the cam journals have 0.04mm - 0.05mm clearance in their bores. All were within a hundredth of each other.



Spec is .0025mm to 0.066mm with a wear limit of 0.10mm so they're all within their original spec, let alone the wear limit.

Sent from my SM-S918B using Tapatalk

[RobFrost]

I cleaned up the original oil pump. It was apparent from the outset the gears were in good condition, aside from a slight notch in one tooth.

The magnesium salts which had collected on the inside of the scavenge chamber, cleaned off to reveal the case in relatively good condition too. The magnesium has a peculiar quality, that the build up of salts onto eh surface can appear quite large but the amount of magnesium corroded is very little.

A more important indicator of condition is tightness of the shafts and the ends of the case, and those are all good.

But to accommodate the increased redline and the extra piston oiling, I sourced a used 964 oil pump. All seems super smooth and tight.



A bit of one of the webs will need to be notched in the case, to accommodate the larger pump.

Sent from my SM-S918B using Tapatalk

[911hillclimber]

Just a note Rob:
When I built my 3.2 the Carrillo rods were bigger than the stock 3.2 rods, and one hit the pump slightly, so needed a bit of a file.

Just saying to check all these rotating parts asap in case a surprise occurs when mixing parts.

[RobFrost]

Just a note Rob:
When I built my 3.2 the Carrillo rods were bigger than the stock 3.2 rods, and one hit the pump slightly, so needed a bit of a file.

Just saying to check all these rotating parts asap in case a surprise occurs when mixing parts.

Thanks Graham, good advice I'll definitely heed. Although I don't have your deep pockets for the Carillo rods!

[911hillclimber]

They were only just shy of £2800….

[RobFrost]

Ouch. I've been collating this list as i go. (Excuse the handwriting, it's on the back of a door). Spent a grand ticking the remaining items off this list, just a few nuts and bolts really. Plus some valves and a couple of panels for the forthcoming right hand sill.

Still making up my mind re pistons and cylinders. I was going 10.5:1 but have been discouraged. Fools rush in and all that.



Sent from my SM-S918B using Tapatalk

[RobFrost]

With the football this weekend and a bit of tidying up and holiday prep, I didn't get a lot done on the engine. I spent a lot of time filling a couple of baskets online with the many little items needed to reassemble the engine.

A job coming up on the horizon is the welding of the case, so I took some time out to practice that, this time moving on from aluminium to magnesium itself.

With this in mind, I had previously bought a couple of magnesium anodes. The key, apparently, is thorough cleaning away of any oxides, which form readily, and whose melting point is greater than magnesium, meaning they don't flow away easily.



I also made some flux from 50% magnesium chloride, 25% potassium chloride, and 25% sodium chloride. The purpose of this is to paste on the back of the work, which may not be protected by the shielding Argon, to prevent oxidation there.



The first weld involved joining two shorter rods of filler material, and on first impressions, magnesium is not difficult to work with at all.



As every pyromaniac schoolkid knows, magnesium, if you can actually get it ignited and the heat isn't conducted away quickly enough, burns ferociously producing extremely bright light and is extinguished by neither water nor CO2. This isn't generally a problem for castings since magnesium is a good conductor and conducts the heat away fast as well as self-smothering by coating itself with oxide.

Nevertheless, I set plenty of post-flow on the welder, for the purposes of fire prevention! I have no suitable extinguisher on hand so my firefighting plan, if needed, was to pull the earth off and use the argon to smother any fire.

Sent from my SM-S918B using Tapatalk

[RobFrost]

After every weld, the work and rods are covered with fine magnesium oxide so as much time is spent with the steel brush cleaning it away as welding. However it cleans away effortlessly, much easier than with steel.

Magnesium welding rods are £385+VAT per kg so I only have a small batch of 12 rods - which will be plenty for this one casting. Nevertheless I practiced at first using some cheaper magnesium firestarting rods as filler - no doubt full of impurities.

Once I'd spent some time doing that, I switched to the proper 3.2mm welding rods and practiced building up the casting. It won't win any beauty contests but it is structurally sound and although there is some porosity in the anode itself, the filler rod material laid down is non-porous.





Sent from my SM-S918B using Tapatalk

[RobFrost]

The fan could do with some cosmetic surgery and that requires drilling out the aluminium rivets.



That was soon separated and each part went off for a treatment appropriate to its composition.



I can't remember if I mentioned but the backs of the test magnesium welds experienced some absorption of atmospheric gases due to being outside of the argon shield.



So after a little research I mixed up some flux composed of magnesium chloride, potassium chloride and sodium chloride.



The 964 oil pump requires a shorter drive shaft, which arrived.



And the final part of the oil bypass mod is a longer spring, a deeper nut, a reinforcing inner tube, and a new plug thingy on top, which instead of having holes, has a channel around it. Looks like I forgot to order the new style plug thingy (technical term).



It's difficult to understand how I spent £500 on a few oil seals, studs, crush washers, nuts, circlips etc. but that's what I did.



This domed sump cover was among the haul, and was nowhere near as domed as it appeared in the pictures, meaning it was insufficient to accommodate the 964 oil pump integral strainer. Looks like it'll be a custom made part for there.



The springs, shims and spring seats are all cleaned, polished and oiled ready for reassembly and the heads went to a local machine shop for some jobs I didn't fancy doing myself. I'm time poor so can't somebody else do it?



Sent from my SM-S918B using Tapatalk

[RobFrost]

Heres the dreaded break on the case, all cleaned up ready for action. Time to give myself a sun tan so I pumped up the jam to 100 amps.



I got the case good and hot all over, and kept it that way with a blowtorch.



The surrounding areas and back of the area being welded were coated with my home made concoction of flux to reduce fire risk and to suppress the absorption of atmospheric gases. You can see it here.



The welding took a surprisingly long time. After every weld I had a very long run-on of gas, to prevent fire, but even with that, the entire area and the filler rod are covered with oxide, so it's a couple of minutes with the steel brush to get everything glistening again before pressing on.

Anyway, I was expecting the end-result to look a lot more like the elephant man than it did.



Sent from my SM-S918B using Tapatalk

[RobFrost]

It's deliberately proud of where it needs to be, so it can be machined back.

There were two other candidates for welding on the case. This spot of corrosion around one of the brass oil pipes.



And a more serious quandary - this pipe where the oil cooler mounts.



The latter is probably not necessary to repair, as the sealing face is actually inside the hole.



But I did anyway. What a nightmare. I've ground it back a bit here by hand but it'll need machining properly.



It turns out when the magnesium has a bit of corrosion in it, it's almost explosive as you first hit it with the arc. But the oxygen does seem to dissipate, leaving the magnesium behind. The filler rod is much nicer to work with than the casting. The magnesium casting has probably become a bit of porous and as you weld it, it seems the oil comes out of it and burns off.

The back seems to have been well-protected by the flux, with none of the porosity and bubbling seen when testing without flux. The uneven surface here is how it started out.



All in all, this step of the job had plenty of scope to go badly, which it did not do, so I'm viewing that as a result.

Sent from my SM-S918B using Tapatalk

[RobFrost]

I economise a little with the story above... The cylinder heads went to a local machine shop who were to face them on a lathe and open the two valve guide bores I damaged to 14mm (or whatever's required to clean them up), plus they were to turn two blank manganese bronze guides on a lathe to match the others, save for the wider diameter.

They mostly do cnc work and this was a job for the old boy who comes in a few days a week. However, the poor old soul collapsed on the shop floor (nothing to do with seeing the state of my cylinder heads) and not back for three weeks.

Becoming impatient, I took the heads to another carefully selected precision engineering shop half an hour away who said to bring them in and they would do the work. On visiting I got a really good feeling because it was clear they could see the difficulties involved and how to get around them. However it was not until I left the heads with them and returned home that they decided they didn't want the job because it wasn't financially viable - even though I had not given them a budget, they told me their hourly rate and said I was happy to pay for their time for however long it took. So the heads are currently 15 miles away waiting for me to pick back up.

I have now ordered the parts needed to make a jig, to drill the problematic head myself. If I ruin it a used replacement is around £250. But it's not the cost that matters - the head I have in hand date matches the others and is the same height, essential for the cams to turn freely.

Sent from my SM-S918B using Tapatalk