Overhead-camming a Type 1 engine

[Jadewombat]

Still pushing on with this science project. I finished the last spike center cap with a different approach of the cutter and the finish is a lot smoother. I have a pretty good feel for this lathe and I should be able to get things fairly close for another machinist to finish it. I also have access to a mill (two actually) now at a friend's shop.

Gonna push on here and get the junkyard parts from a Golf I need to mock this whole thing up with and get some ballpark measurements of the aluminum plate I'll need to buy.

[Piledriver]

We're scrapping out an old Zevatech robotic pick-and place unit at work, as support is no longer available.
Has a ~3'x4'x2' workspace... huge, heavy optical table slab, very high accuracy...

When I look at it , it screams to me it wants to become a huge 3D printer.
The mechanicals are perfect and the stepper drivers/hardware is std Galil stuff, its just the software was always buggy as hell.
EMC2 or the std OSS 3D printer software could probably run it fine.

If I had a big shop I'd have to buy it at auction.

[Jadewombat]

Yeah, the guy I've been metal apprenticing for the last few months has said 3D printing is the future of manufacturing (probably mostly true for a lot of parts). I think that'd be cool if we could shift that type of manufacturing back to the good old USA from overseas.

I have little to no experience with additive manufacturing and/or laser mapping the original. Anyone else have experience with it? Would like to mess with it somewhere down the road.

Gonna cut my teeth on Sketch-up next before I start any of this.

[petew]

Just came across this post. I've been dreaming off and on about this for years. There are 2 finished ACVW motors I know of that went OHC. One guy was using VW golf sohc heads and another using Subaru EJ20 dohc heads. Both were rallycross engines. I've also heard of another one with 911 heads as you suggested.

http://www.rallybugs.com/ppjracing.shtml

If I had the time/skills/money to buy machines as you suggest, I wouldn't actually be going OHC cam though. I'd be converting the motor to a direct injection 2 stroke, ala Ralph Sarich. It would require forced induction to make the best of it. Having said that, your limit will always be the crank/bearing setup.

[petew]

P.s. here's Peter Harrold's car in action...

https://www.youtube.com/watch?v=Oj_sMVvXSbc

[Jadewombat]

Thanks for the links. I definitely want to follow the same layout of toothed belts that Subaru uses with two tensioners either side of the crank pulley. I don't want to jump ahead on any of this. I'm just taking it real slow but the goal is definitely not some monster that revs to 10,000.

I want to start with very mild changes and keep everything apples to apples changing only the pushrod configuration on a 1600 to SOHC to see the differences in the rev. range and rev. characteristics, then go from there.

[petew]

Have a look at this... I reckon it could be directly bolted on to stock head instead of the rocker cover.

https://www.youtube.com/watch?v=S3cFfM3r510

[Piledriver]

Slick, someone finally figured it out.
The pneumatic valve springs would be a huge win by themselves... have actually been around for ages.

...You are not going to see much gain save from losing the foot long pushrods and heavy lifters.
The cam profile is what it is--- the motor is not going to see a benefit except at highest RPMs.
OHC provides no magic by itself, but it can allow you to do things at higher RPM.

On most production engines its done for packaging/part count reduction and assembly simplification, cost reduction reasons. not due to some inherent magic. There is no magic.

If you build a stock 1600 DP with stock profile overhead cams its going to run like a stock 1600 DP, with a little less lift, and maybe capable of slightly higher revs. It may be a little quieter, but thats about the gain to be expected.

[falcor]

Well the thing about the freevalve is that you have no camshaft at all. The computer dictates the opening of the valves, how much,how long and when. Basically you can have a virtual camshaft with square lobes.....

[Piledriver]

Well the thing about the freevalve is that you have no camshaft at all. The computer dictates the opening of the valves, how much,how long and when. Basically you can have a virtual camshaft with square lobes.....

I get that, but I kinda doubt Freevalve is selling modules/drivers retail for you to design your own heads.
They are pushing tech/patents to GM/Ford/BMW/VW etc. Its the Holy grail for valve control if it works.
But:
Pneumatic springs have been used for >70 years and are a huge win in some racing applications, and A Guy With A Lathe could probably even make working units. A DI air pump off a Mercury DI motor (Orbital air assisted DI) would probably make enough air pressure/volume and is probably fit for purpose.
If you aren't redlining well past 10K RPM probably a waste of time.
Commonly used on motors that idle at 10K.

[Jadewombat]

Wow, absolutely phenomenal. Thanks for the link. The Dodge (ahem, cough, barf) Dart has a simplified version of this. Not camless, but some fuel saving blah, blah, something or other.

Would be awesome to do this in a perfect world to a Type 1. It would make the engine a little wider (so will an overhead cam configuration) but overall lighter with no cam, pushrods/tubes, rockers, or throttle body. Don't think I can get my hands on a set of those lifters or an ECU to run them, but Man it'd take the 1930s design to almost the year 2020 in a heartbeat and only weight 200lbs.!

[petew]

Hmm, not exactly. We still really don't have enough bearings in the crank.

[Jadewombat]

Time to revisit this topic. I got a junked golf head and I started mocking up in wood what I think will work. The cam is a type 1 and the lifters are from a golf/jetta. The lifters sit on top of the valve/spring perfectly, not much wasted space. Type 4 lifters will not work for this because they're long and skinny and would make this plate very thick.

I spent quite a bit of time trying to understand the type 1 cam configuration out at the cylinder heads and what appeared to be a major problem why the lobes wouldn't work with this configuration...until I remembered the type 1 cam turns "backwards." So what you're looking at in the photo will be the drivers' side head.

The thickness of the plate is determined by how much those lifters stick out. On the golf/jetta head it's ~8mm. I'm using ballpark numbers in all of my measurements at this point on the mockup to understand the sequence of machining that will need to take place, the oil passages need to be drilled, cam saddles, how to seal the plate to the cylinder head and fabricate a valve cover, drive shaft and cam gear to mate to the timing belt, etc.

[Jadewombat]

Drilling the oil passages will be a bit tricky. Essentially one long bore will be parallel to the lifter bores, the three cam oil passages will be perpendicular, and the oil passages for the lifter bores will bisect the main bore at an angle. Looking at the golf head, the oil passages are cut like jets facing up at the camshaft to fill the groove in the lifters and pump them up.

What's also interesting (to me anyway) on the golf head is there isn't full surface coverage of the camshaft with one of the cam saddles. It's a U-shape to accommodate the lifter bore and lifter. I'll have to do the same U-shape for the T1 cam saddle near the #1/#4 exhaust valve. The upper portion of the cam saddle is full coverage of course.

I compared the amount of distance between the lifter bores next to each other. A bit surprisingly, on the T1 head there's 2.5mm more spacing than on the golf head. This is surprising because watercooled engines typically have less spacing between the cylinders.

[Piledriver]

Many 8v golf heads had bigger valves, and they used the ~whole bore.