Dougs951S

I had an idea today. It's probably a bad idea, but I will speak it aloud here anyway. It is a fairly common thing for people with 16 valve engines to do away with the timing chain and go to a twin cam gear setup to drive the cams. This is more reliable but you still have to deal with the belt which is in my mind, the achilles heel of the 944 engine. What about driving the cam directly off the crank via a system of gears? It would be unfailingly reliable and more accurate than a belt, and would never need maintenance. The downside is that it would add rotating inertia. The real downside though is that it would require re-engineering of the accessory drive, and the large distance between the cam and crank would require either two very large gears, or 3 smaller ones. A modified water pump could possibly serve as the intermediary between the crank and the cam. It would definitely be an engineering project but I think it would be worth it to have a bulletproof cam drive. Any thoughts rennlist?

Scott H

I would remove the belt covers and make the engine non-interference before I even thought about trying to engineer something like that. Converting it to chain drive would be next on that list.

refresh951

+1 My thoughts exactly.

But I think it is cool that you are thinking about ways to improve the platform!

Dougs951S

Thanks!

Making the engine non interference would be the best solution, but that would require a rebuild of the engine. Converting to chain drive is also a very good idea and one I had considered as well. Most modern OHC engines are roller chain driven and they are extremely reliable, just slightly less so than a gear drive. One would just need to engineer a way to tension the chain. From an engineering perspective you just cant touch direct drive for simplicity and reliability, but when you consider a real car in the real world, a double roller chain is damn close.

Of course then you have to consider the fact that thats a damn long chain, with a long unsupported run. Chains stretch over time as well, that would alter the cam timing. Perhaps design something of a hybrid, converting to chain drive but using two separate, offset chains with one driving the water pump off the crank, and another driving the cam off the water pump.

divil

+1

I would be curious to know if it could be converted to chain drive without going to way more trouble than it's worth. I presume you would need a way to lubricate the chain (although the way my car leaks oil in that area it probably wouldn't be an issue lol ), and also modify the water pump, or replace it with an electric one?

Still, curious as I am, I would just stick with the original design. Maybe I'll change my tune if I ever have a belt failure, but so far I have had no issues. I am not sure how often a properly tensioned oem belt with fresh rollers/water pump etc. fails, but I bet it's not as often as people think.

Now, if we could figure out a way to drive the valves independently of the crank, so you can have any valve timing you want at any engine rpm...But that kind of mod is probably in another league

Reimu

By the time you did it you would have probably had an easier time just converting to non-interference. Not to mention you could pick up some displacement on the way..

txhokie4life

A chain will need a tensioner -- which is another source of failures.
Either a chain system or gear system will need to be lubricated.

I really don't see the belt as an issue is you keep on top of things -- and anyone
that is willing to replace it with a chain or a gear system is probably already more attentive than most.

the real achilles is the #2 rod bearing -- I've lost 3 race/track engines to that via one means or another --

BTW, there is a good size 951 racers presence in Austin, lots of tinkering and contemplating how to make these cars last.

cheers,

Mike

txhokie4life

combine a sensor on the crank with a stepper motor and control circuity and you have what you've described -- I'm just not sure how reliable it would be.

Mike

divil

Driving the camshaft with a motor might be a start, so that you could advance/retard it a little, but the real payoff is not having the camshaft at all. If you want ideal valve timing you would need to be able to control every aspect independently, i.e. duration, overlap etc. Servos of some kind are what you want, that you can control electronically with speed and precision.

I think you need a lot of force to open the valves so the power source that drives it need to be up to the job. Then of course you would want the timing to depend on load, not just crank rpm.

theedge

Anyone who wants chains needs to go change some chains on a VR6 or something.

Remove the transmission, take the covers off, intermediate gear, two chains, two tensioners, plastic guides all over the place, line everything up on both chains and all the shafts, line the cams up, lock the cams with a special (although fairly cheap) too, reseal the covers so it doesnt **** oil all over the place, an oil seal for the shaft....

Yeah, we suffer greatly with our belts and easy plastic covers....

Dougs951S

I AM attentive as hell about my car and totally OCD about PM. I spend WAY more on PM then I do on mods, and I am rewarded with a car that runs and drives like its brand new, zero issues or hiccups. That being said, I've worked on enough mechanical things; cars, boats, aircraft, motorcycles, ect to know that even if something is MOSTLY reliable, you basically have to look at it from an entropy standpoint. The longer something runs, the more it tends to catastrophically fail. Put another way, the farther a system is from the ideal 100% reliability, the more likely it is to suffer a random failure no matter how diligent you are. So even if the 944 belt rarely breaks, we all know that it happens and my feeling is that anything would beat the factory set up, even if it was more complicated. In my ideal 944 engine, there would be no AC, power steering, or balance shafts. The water pump would be electric and the alternator and cam would be direct driven through gears. This is stated simply from an ideal engineering standpoint of reliability.

I've suffered my fair share of totally random failures, the types of things people tend to scoff at as "never happening." Case in point, I had a 3.0 mitsubishi 6G7 engine that literally snapped the crank in half...when I pulled it apart I expected to find a broken rod/piston/ring or toasted rod/main bearing. nope...the crank just snapped for no reason. go figure.

Shawn I realize Sid is the person to be asking this, but since its relevant, do you happen to know how he achieved his non interference build? I know he was using a stock head and maybe a different deck height, so was his engine non interference to start with, or did he simply cut valve reliefs in the pistons? I ask because I'd be happy to take a small cut in CR in exchange for a non interference fit if I can do it with a stock head and stock rotating assembly.

refresh951

The non-interference design in my case is the result of the ability to choose from a large selection of off-the-shelf small block Chevy pistons. The piston choices on a traditional build are fairly limited. By choosing a piston with a given pin height and inverted dome volume, a combination can be achieved that results in the piston running a bit down in the hole at TDC. This kind of setup can result in a non-interference setup but should be verified by "claying" the valves. There are a lot of considerations here, such as squish area, but both Sid and I have had good results with this type on configuration running E85. Sid, of course, did it first. With the stroker motors we also have the flexibility to adjust stroke to obtain our desired CR.

On a traditional build I am not sure you could get there without custom pistons. I have not looked at this in any detail however because I will never build a traditional motor again.

BTW - my pistons cost $94/each

txhokie4life

Camless design

http://www.cvel.clemson.edu/auto/sys...ve-timing.html

Dougs951S

Sorry Shawn, I didn't mean to discredit you or your incredible build. Your work ethic, speed, and ambition are inspiration to me. I mentioned a stock rotating assembly simply because I am not in the position to be building a motor right now since I'm about to go back to school. When the time comes, I will absolutely be building a 2.85 hybrid with a stock head and an HX35, which will eventually get big valves like yours or possibly a 16 valve head. I simply didn't realize your engine was also non interference. Your initial 2.85 build was not, but that changed when you upped the stroke right?

Paulyy

Gears = more rotating mass.
Gears can also get chewed up from an foreign object
If the gear gets damaged, it'll get expensive to replace.
Bearings will need to be replaced on those
Gears can fatigue.
Gears need grease & lubrication


There is nothing wrong with the timing belt. The reason they break is because the owner didn't maintain it properly. The belt should be replaced every 45,000 miles. / 80,000 Kms. When you change the timing belt, the water pump should be changed also.
Thew water pump can damage the belt if that decided to go.
I've never heard of a story of a belt failure from a new belt that was installed correctly with a new pump

This belt issue is getting bashed to quick but the one to blame is the owner or who is maintaining the car.