270 Offy

[kvom]

Custom allen wrench for the win!

 

[petertha]

Hi Terry, just going back to your gaskets. Was the Oracal selected maybe because you could computer cut those smaller fiddly & has self adhesive for positioning or..? Any idea if/how much that material might change in thickness once clamped or do you just allow for the full nominal 2.5 mil thickness?

 

[mayhugh1]

Petertha,
The Oracal gaskets weren't computer cut. Although they certainly could be, the gaskets I've been typically using it for have been too small and separating the backing sheet would have been a hassle. For my gaskets. I removed the backing sheet and then press the pre-cleaned part down against the sticky side of the vinyl. An Xacto knife with a brand new blade was then used to trim the excess away from the part. The vinyl doesn't seem to compress any measurable amount on the surface areas I've been typically using it on, - Terry

 

[mayhugh1]

With the number one piston at TDC and the flywheel rotated so its engraved TDC mark was directly under the rear housing's timing pointer, the flywheel was installed on the rear of the crankshaft. A simple shop-made tool was used to hold the flywheel while the locking tapers were wrenched tightly together with the crankshaft nut. This left-handed nut will provide an alternative rear start for the engine.

The front cover/water pump assembly was installed next with a .010" Teflon gasket between it and the crankcase. This required patience because the shaft on the upper 60 tooth drive gear had to engage its cover bearing simultaneously with the tang on the oil pump shaft engaging the slotted water pump shaft. Since the starter clutch winds up captured between the water pump and front cover, the starter shaft had to be installed on the crankshaft at the same time. Installing the starter support bracket was trivial and completed the front end assembly.

With the crankshaft finally supported by all five bearings, it could be safely drill spun for a more thorough test of the oil pumps. In the process, I discovered the oil passage leading to the top-end was blocked. The -002 o-ring between the crankcase halves used to seal the passage in the seam had been compressed completely closed. After looking back over my notes, I realized the crankcase had actually been machined for a 1x1.5 metric o-ring. I had only to spread the crankcase halves slightly apart in order to replace it, but I had to start over on the just-assembled front end.

I then discovered the gear tower and the head (with its installed camshafts) had to be partially preassembled before either one could be installed on the engine. The front cover's protruding 60 tooth drive gear along with the "Y" shaped tower would otherwise require the camshafts to be removed from the head. Preassembling the pair was easier to do than to describe, but the cam gears still had to be removed from the cams and placed in position inside the gear tower.

My method for securing the head to the block was to use sixteen 5-40 SHCS's threaded up through the top of the block. (The original plans specified long studs running up through the block from the crankcase.) In addition, I repositioned the head bolts slightly to provide a little more head gasket 'meat' around the cylinders. Unfortunately, these new locations with their already limited clearances to the stock intake/exhaust ports allowed a thread depth of only 3 to 4 threads. The ports were later re-designed, and their new shapes provide more clearance, but I forgot to go back and deepen the threads for more engagement. Still not wanting to remove the camshafts, I wrapped the entire head assembly in vinyl tape for chip-proofing while the holes were deepened for an 8 to 9 thread engagement.

Since the head was machined from aluminum and the SHCS's are stainless steel, there is a potential for galvanic corrosion between the two inside the block's wet environment. For some protection against this, the screws were liberally coated with nickel (anti-seize) grease before being installed.

The o-ring cord surrounding the oil passage running up the rear surface of the gear tower eliminated the need for a gasket or sealer between it and the block. Oracal was used on its bottom, however, as a gasket to seal a potential leak around the oil drain back passage surrounding the 60 tooth drive gear. - Terry

 

[srobovak]

It's getting there! Really looking forward to see/hear it running...

 

[kvom]

For deepening threaded holes, how much smaller than minor diameter do you choose for the drill?

 

[mayhugh1]

kvom,
The original tap drill size was .101", and for deepening the holes I used an .096". - Terry

 

[mayhugh1]

While standing in front of the engine the intake valves are on the right, and the exhaust valves are on the left. Both camshafts rotate counterclockwise while the crankshaft rotates clockwise. The camshaft gears accumulate significant backlash from the five meshed gear sets that connect them to the crankshaft. Since the 40 tooth cam gears rotate once for every two crankshaft revolutions, their resolution winds up being 18 crankshaft degrees. In order to reduce this to a more manageable 5 degrees, each cam gear is pinned to its camshaft using a particular hole in a special pattern of holes that was drilled through the cam gear.

In order to realize the resolution offered by what is effectively a vernier, the proper tooth on the cam gear must first be meshed with its driving gear. The correct hole in the vernier pattern will be one that lies directly over (or is very close to) a hole in the camshaft flange. This hole will be among those located in one of the ten possible hole combinations made available by the vernier. Five of these hole combinations have been sketched in a photo. Each is separated from its neighbor by a single cam gear tooth. The sequence of hole combinations repeats every ten gear teeth, and so the probability of randomly selecting the correct tooth is only 10%.

In order to locate the correct hole for the intake cam, for example, the crankshaft was first set to the angle (20 deg BTDC) at which the intake valves should open. (Since the perimeter of the flywheel was previously engraved with five degree tic marks, a degree wheel wasn't required.) The as yet unpinned camshaft was then rotated to the point where the intake valves actually were beginning to open. The (unmeshed) cam gear was then rotated to a candidate position.

However, in order to be rotated and since the camshaft was already installed, its bearing caps had to be loosened so the cam gear could be raised free of its drive gear. With the bearing caps re-installed and the camshaft back into position, a drill was temporarily inserted in the hole with the best looking match. The result was evaluated by rotating the crankshaft and measuring the intake valve opening and closing points so they could be compared with the camshaft specs. A satisfactory result wasn't found with that particular hole (nor any of the others at that particular tooth location), and so the camshaft was again raised so the cam gear could be rotated to another tooth. In practice, it's harder to do than describe. The backlash and valve spring forces tend to rotate the cam while it's raised and make the tooth re-meshing something of a trial and error process. Once the best hole in the best combination was found (four tooth tries later), the drill was replaced with a pin, and the cam gear firmly secured to the camshaft with a nut and wavy washer.

It took several days to fully understand the vernier, wrap my head around the various measurement problems, and to come up a procedure to deal with them. Some upfront preparation turned out to be very important. For example, when machining the camshafts, I included a permanently installed grub screw in each end so they could be easily rotated with a hex key. (The front screws were drilled through for oil passages.) The ability to rotate the camshafts in small controlled increments was invaluable.

Since the lobes on the overhead cams prevent easy access to the followers, indicating the valves is difficult. A shop-made fixture using a needle probe working with a dial indicator was able to access the outside edges of the followers, but the results were inconsistent. I eventually settled on using a strip of .0005" shim stock between the lobes and their followers in order to detect the valves' opening and closing points.

Another issue is that the tiny index pins are nearly impossible to install without special tools, and there's a risk of losing them in the gear tower. One of the photos shows a pair of tools that worked well for me. The insertion tool is a length of .063" magnet-tipped drill rod that slides inside a close-fitting aluminum tube. With the rod fully inserted, the magnetic end of the rod is flush with the end of the tube. Withdrawing the rod slightly, allows the steel pin to be loaded inside the tube and safely held on the end of the rod. After the rod pushes the pin into the cam gear, the tool is swiped sideways leaving the pin in place. For extraction, I used a simple tool made long ago to test Hall devices. It's just an 1/8" diameter magnet pressed into the end of a brass tube.

Even with these tools I quickly realized that it was much easier to deal with the pins while the gears were horizontal rather than vertical. So, I also made a simple wood stand to safely support the engine nose up for some of the steps.

My final valve timing results were:

intake opens: 5 deg BTDC (spec was 20 deg BTDC)
intake closes: 60 deg ABDC (spec was 52 deg ABDC)

exhaust opens: 55 deg BBDC (spec was 48 deg BBDC)
exhaust closes: ATDC (spec was 6 deg ATDC)

I retarded the cams a bit in order to reduce the overlap at TDC so it would be more in line with the other model engines I've built. The higher manifold vacuum that may result might help later with carb synching.

Sanity checks on the valve timing were done simply by holding a thumb over each spark plug well while the engine was spun with a drill. Although the plug wells effectively increase the combustion chamber volume (and decrease its pressure) by some 50%, the pressure pulses felt strong and consistent among all four cylinders.

I was also finally able to thoroughly exercise the oil system. The crankshaft appears to be receiving oil from the pressure pump, and the top end is getting a share of the oil from the scavenger pump which also seems to be able to evacuate the sump. - Terry

 

[Vixen]

Is it possible to publish a drawing showing the angular relationship of all the holes in the cam gear and camshaft?

Thanks

Mike

 

[mayhugh1]

Is it possible to publish a drawing showing the angular relationship of all the holes in the cam gear and camshaft?

Thanks

Mike

Are you saying you want to see the other 5?

 

[Vixen]

In the bottom centre drawing (position n ) there appear to be a set of eight holes and a second set of four holes.
Only one of the four holes lines up with the eight.
Are the eight holes equi spaced = 45* ?
What is the position angles for the four holes?

Thanks

Mike

 

[Vixen]

Are the four holes angular positions something like 0*, 105* 190* 275* (5 degree increments)

Mike

 

[mayhugh1]

See post #222 at:
https://www.homemodelenginemachinist.com/threads/270-offy.31486/page-12
The four holes in the camshaft flange are equi-spaced at 0, 90,180, and 270 degrees. The 8 holes in the cam gear are the ones with incrementally increasing angles.
Terry

 

[Vixen]

Thank you Terry.
I can see that in the drawing now you have explained it.

Mike

 

[propclock]

Thank you again for your concise explanation of your experiences . I especially appreciate
your willingness to share the problems encountered and not just showing the things that
go as planned. Thanks again!

 

[Ken I]

I remember the old Alfa Romeo Julietta DOHC engines had 19 holes in the cam flange and 20 holes in the chainwheels thus the angular error between the next hole lining up was 360° ÷(20 x 19) = 0.947° per "step".
The Jags had something similar with Vernier wheels.

 

[mayhugh1]

Thank you Terry.
I can see that in the drawing now you have explained it.

Mike

Mike,
I was so close to the problem that I didn't realize that it wasn't clear in my sketch which holes were drilled where. I set the cam gear to 'transparent' in my CAD tool, and what was going on was much more obvious on my screen than it ended up being on the jpg that was created from it. Thanks for your interest. - Terry

 

[Vixen]

Terry
My interest in the Offy vernier adjustment relates to my 1/3 scale Mercedes Benz W165 engine. It also has vernier adjustment for it's four camshafts.The only design information I have is from these two archive photos. The W165 appears to have 19 holes in both the cam flange and cam gear; Both sets of holes line up and there are four bolts holding the gear to the flange, nothing else. Mercedes seem to be using a completely different approach to the Offy Mike
.

 

[Ken I]

Vixen - I'm guessing 19 holes in both allows 18.947° per hole while 10 teeth of the 64 tooth gear is 56.25° leaving an an error of 0.592° to the next hole (3 of 19 divisions = 56.84°) to line up. That's a pretty fine adjustment (0.592° per step) - and 4 bolts are close enough to symmetry as to not worry about balance.
In the case of the Alfa engines a single small bolt went through the correctly aligned hole and a central nut on the center did the securing by friction.
I like your (Mercedes) set up better.

Regards, Ken

 

[Vixen]

Ken,
Sometimes, counting teeth and hole spacing, from old photos, is the only way to unravel the detail.
I work it out slightly differently to you. A 64 tooth gear has 5.625* between teeth. The 64T gear can be fitted in any one of 19 angular positions. So 5.625* divided by 19 = 0.2961*. Now that is a fine vernier adjustment.
I think we may have trespassed on Terry's turf for long enough.

Mike