PorKen

I did a quick run with ~2 degrees retard.

Boy, that sucked.

You might be able to run the cams retarded, if you have a manual, and can keep the rpms up, but with the auto, it's a dog, no power under 3K. (Might be good for supercharging.)

I did find that there was some parallax in the way I was sighting my hub indicator to the housing mark. Enough to give the advance overshoot I had.

I am going to set it to 2.5 advance and leave it. No dyno required.
...

jpitman2,

I am considering having a hub laser cut from steel plate. I don't believe that I could make one myself with the proper dimensions. Anyways, I'm happy with the modified factory hubs so far.

The one bolt hole that appears close is actually offset from the keyway. The other holes are not quite equally spaced (they are offset to act as stops in the slots), but could be equal without problem. I see the keyway as being the most difficult to cut manually.

You could line up hub marks with a gear opening, but then you're counting on the gear markings being correct WRT to the housing.

I strongly believe this the single best upgrade you can do for the 16V engine, as a starting point for all other upgrades. Even if you set it to stock (no advance/retard), it's makes the engine much, much smoother. But I wouldn't worry about finding parts, from my read of our 928 community, most will not bother with a mod like this.

SharkSkin

Ken, I really, really like this upgrade. However, I'm about 1500 miles into a new T-belt.

Whatever happens, next time I need to get in there, I'm going to want to spring for the newer gears and set them up like yours. I think I'm getting the idea how the pointer works, but I'm still a bit fuzzy on the exact details. It seems like a very straightforward problem to solve, so I'll re-read your posts and try to uncover whatever I've missed.

This is really awesome work BTW, and I think that if you could get it to either a kit or a set of dimension drawings for parts fab/modification you might be surprised about the uptake. There's a lot of people with 10 or 20 times the mileage on their T-belt compared to me who simply aren't in a big hurry to dig in there. Give it time.

jpitman2

Ken, I am a couple of years away from NEEDing to do the belt, but I am also very keen on this mod. When you say even just setting to stock is better, do you think is due to the error that always seems to be there in the 1-4 bank cam alignment? I had it with my belt, and many others have reported it IIRC. The WSM and others say to rotate cam Anticlock until belt matches wheel, but this is retarding isnt it? Wouldnt advancing be preferable here?
I had though of getting a new keyway cut in existing 16V wheels to give 2-3 degrees advance - I have seen wheels on other makes with 3 keyways, providing for std, +x and -x timing. What is the relationship of the keyway in the wheel to the timing marker on the edge? The wheels are only one part number for both left and right cams, so are the cams differently keyed? I need to get my hands on a cam and a wheel and see how things line up.
jp 83 Euro S AT, longing for more bottom end ......

PorKen

Ah... back to power on demand. (I set it to a conservative 2-2.5)

One thing that I have read and probably forgot to mention, that lowering the torque band rpm range optimizes the engines power for our stock, low-stall-speed torque converter. The engine can then make better use of the torque converter's ability to almost double the torque to the wheels (and increase the smile factor by fivefold ).
...

SharkSkin,

Basically the pointer or 'hub indicator', which I've started calling it (to differentiate it from the cam housing pointer), acts as the mark on the back of a 16V gear.

Since the 32V gear can move relative to the cam position, the gear mark is variable, but the hub indicator is tied to the cam key, and does not vary.
...

jpitman2,

Yes, the 1-4 cam is the worst, but the 5-8 is usually off a tiny bit too. Advancing or retarding a single cam to fit the belt doesn't matter, either way you are introducing an offset in the overall timing of the engine, which makes one bank more or less powerfull than the other, which translates into vibration.

The keyway's center, if broached (cut) properly, will line up with the 16V gear mark.

pvoth

after your sure the bugs are all worked out and you get everything right on and bullet-proof I will surly be interested in this.....
timing belt change.....this winter...will plan on //start getting all the other parts together
put me one the list
peace

SharkSkin

I get that, and the part about getting off of the cam lobes so you can move the gear without moving the cams. I guess what I was looking for was a precise way to determine how many degrees advanced/retarded each cam is. IIRC, that is done by counting gear teeth, dividing by 360, then estimating using a sort of "fraction of a tooth" unit, measured using the eyeball Mark II, correct?

PorKen

SharkSkin,

I am, in fact, using the Mark Deux eyeball; quesstimating by the position of the crank dampener marks. I am setting my cams to the mid point between 0 and 10, ~5, or ~2.5 cam. I'm pretty confident that I can get the same amount for both cams, but the exact position is an approximation.

The only way I've thought of to be more exact, in degrees, in relation to the crank position is to have a laser cut degree marked wheel. The wheel would either sandwhich between the stock hub and the gear, or be attached to a custom hub. Either way it needs to relate directly to the cam key. This is assuming, hopefully, that the keyway is cut exactly on every cam.

Perhaps the only way to be as accurate as the 32V setup, would be to drill and tap the cam housing for an inspection port over the proper cam lobe, whence a dial-indicator could be employed to verify the cam position.

SharkSkin

I KNEW I overlooked something with the cam tower R&R.

That sounds good enough for the average tinkerer. If it's off a degree, it's still better than the way these motors go together normally.

OK, so it looks like the hub/spider gets machined to 10mm, and the one piece that is most critical is the pointer. Will you be making a jig for these? Looks like you only use one, and move it to the cam you are checking. However, it seems that the overall process of adjusting the timing would go somewhat faster if you had two pointers on. After all, the setting on one cam should not have any effect on the setting of the other cam. I think I'd want two of them.

You mentioned moving the torque curve down at 2.5 degrees... can you point me to some of your dyno sheets? I'm just curious, and would like to speculate to myself as to how different settings would "feel" with a 5 speed.

Again, great stuff.

PorKen

Hub to 9mm. Jig, yes.

Here is the comparative dyno run at 0 versus ~3, with the single idler. With two idlers the torque curve was more flat, started earlier, and ended later.

SharkSkin

Er... yes, you said 9mm, but you also said the belt was running a bit forward of center on the gears, right? Wouldn't cutting to 10mm just about center it?

PorKen

I'll settle for 3/8"*, that'll make it easier to spec from plate steel.

Both banks have slightly different offsets, and while running, the belts move too. I suppose that's why the gears are wider than the belt.

At 9mm, on one side the belt is almost in the middle, the other somewhat forward.

*9.525mm