Norton Desaxe cylinder offset

jseng1 said:

With the wedge shaped base gasket the amount of deflection at the top of the 850 through bolt is .048". The bolt hole for the through bolt is .034" oversize so that gives .017" clearance on each side - reducing the bolt deflection at the top to only .031" I tested it and the through bolt seems to thread in no problem.

And no it doesn't change anything with the timing gears or timing cover.

Click to expand...

You can always hold the barrel vertical to start the bolts, but final torquing will always cause extra friction.

The stock bolt threads are not perfectly aligned with the bolts causing the bolts to wobble when you thread them in. Check it yourself and you'll see. This wobble is greater than the offset created by the wedge base gasket. Its nothing to worry about. I can make a vid when I get the gaskets.

The real concern is the extra wear and stress your engine is subjected to because the Cylinder is offset in the wrong direction as set up by the factory when they went from 650 to 750 without correcting the cylinder to crank center alignment.

Will there be a force on the wedge-shaped gasket, trying to move it towards its thick edge, when it’s squeezed between barrel and crankcase mouth?

Sorry if already asked somewhere up there.

The tapered base gasket will prove or disprove the theory. If successful the crankcase mouth could be milled at an angle and even thread inserts installed to align the threads to the gasket face

It will be a couple months before I can install one of these gaskets. I don't really want to disable my test bike until winter hits. When it happens I'll document it with video.

And yes there is the option of milling the cases at an angle and I already did that back in the 80s when I needed to lower everything to fit the first shorter lightweight Wiseco pistons. I noticed the deck was off from the crank centerline. I thought it was just bad machining. So I trued it up. I also installed timeserts - not because I was worried about the angle but because Ron Wood told me he had stripped the threads in his 750 short stroke so I did it as a precaution. This was for my all out monoshock 850 racer as below.

I suggest desaxe is a worthwhile train of thought, but with the Commando engine the concept is not normal, it already delivers massive torque - the problem might be more about using it effectively. When I first rode my Seeley 850, the normal gearbox was useless on a race circuit. The Manx gear cluster with high overall gearing fixed the problem, but the high first gear caused slow starts. I am not used to bikes which accelerate faster when you raise the gearing. When I first raced it, I suspected that it was revving without pulling. The only way to find out is to raise the overall gearing, and use close ratio gears. The steps between the gears are too large in the normal gearbox. Revs lost on up-changes are a problem when throttle response is poor. I have had a lot of experience with old garbage - the Commando 850 motor is different from many. I suspect it is a different or alternative concept. Norton motorcycles were developed from race experience. Most motorcycles go light crank and higher revs with lower gearing. Manx Nortons were usually limited to about 7000 RPM.

Jim-

I would think it would be a simple matter to 3-D print such a gasket. The mating faces could be printed using TPU for sealing purposes as it is resistant to oil and gas and can be stable to temperatures in excess of 320F (160C). If compressibility is a concern, the core can be printed using a different filament such as heat-treated Protopasta Carbon Fiber HTPLA which can be used up to 310F (155C). Alternatively, one could probably use only TPA filament but use something like 80-90% infill.

PM if you want to discuss this further.

Thanks,

-Robert

VooDoo Pistons said:

Jim-

I would think it would be a simple matter to 3-D print such a gasket. The mating faces could be printed using TPU for sealing purposes as it is resistant to oil and gas and can be stable to temperatures in excess of 320F (160C). If compressibility is a concern, the core can be printed using a different filament such as heat-treated Protopasta Carbon Fiber HTPLA which can be used up to 310F (155C). Alternatively, one could probably use only TPA filament but use something like 80-90% infill.

PM if you want to discuss this further.

Thanks,

-Robert

Click to expand...

Tthe thin side needs to be .010" thick max.

Jim, other than correcting Norton's wrong direction desaxe, what tangible benefit do you hope to get from this exercise?
Cheers

jseng1 said:

The thin side needs to be .010" thick max.

Click to expand...

Thanks, Jim.

I understand the 0.010" dimension at the rear of the cylinder base. Is the dimension at the front of the cylinder base 0.080" or 0.030"? It's hard to tell from your drawing.

Is it also safe to say that these dimensions are not the "relaxed" dimension of the gasket but rather the compressed or installed dimensions at full torque? I would think this to be the case. If so, what is the acceptable range of variance at the front of the cylinder? The problem can be solved to achieve the target values, but it will just take a little more on the percentage of infill to achieve the target values. The filament used for the mating faces, possibly TPU, will have the same thickness on both the top and the bottom, but the compressibility of the core material might need to have a varying percentage of infill as you move from the rear of the cylinder to the front.

Thanks,

-Robert

robs ss said:

Jim, other than correcting Norton's wrong direction desaxe, what tangible benefit do you hope to get from this exercise?
Cheers

Click to expand...

The reclined position of the cylinders makes for relaxed riding.

One more thing...

I have a set of cases in the mill right now that have a total variation across the deck of 0.013" which can be seen in the images here:


I am seeking the target dimensions from the CL of the crankshaft to the deck and the CL of the camshaft to the deck so I can ensure the barrels are correctly oriented relative to the crank and cam.

Does anybody have these dimensions on hand that they can share?

Thanks!

-Robert

VooDoo Pistons said:

You should correct the Desaxe while you're at it. Or just use my specs and drop the rear side .070" (or just 1/16"). You want the centerline of the cylinders to be at least 1 mm in front of the crank center (2mm may be preferredi). I would never use any base gasket that takes a compression (I don't use a base gasket). Paper base gaskets compress more over time and then they leak and you have to pull the head on an 850 to retorque. Even alum base gaskets have been getting a bad rep because some people anneal and soften them too much. The copper base gaskets and head gaskets I sell do not compress and annealing them is pointless unless you want to clean them.. The new wedge base gasket will be steel instread of copper simply because copper is more expensive. Andover's compression reducing base gasket is also made of steel. You can always use one of my various thickness copper base or head gaskets to restrore to original compression after milling your cases.

Click to expand...

Please show us a photo of your jig for mounting the cases in your mill.

Thanks, Jim.

I could correct the Desaxe while I'm at it, but first I need to know what the factory targeted in their blueprints.

I have referenced the side-to-side axis of the cases by measuring off of a precision-ground shaft that I have installed in the bearings in place of the crankshaft. This shaft extends on the outboard side of both bearings through the cases and rests on v blocks along the X axis. These V blocks have been shimmed and tweaked such that the shaft is exactly 90 degrees perpendicular to the quill, or Z axis. This arrangement can be viewed in the photos at this link: https://www.nortonownersclub.org/forum/camshaft-crankshaft-centerline-dimensions-relative-case-deck

However, to get a reference point for the target deck height, I still need to true the cases in the Y axis which is why I am seeking the camshaft and crankshaft centerline reference dimensions.

Once I establish this theoretical deck plane I can then explore alternative milling options, but until I have that theoretical plane, I'm just guessing.

-Robert

With any of these changes, ultimately we are stuck with working with the result. With a Commando engine, an improvement in torque cannot be detected without altering gearing to optimise it's use, on a known circuit. I suggest torque improvement is probably intangible with the heavy crank. Theoretically changing desaxe should make a difference. Compression ratio is a different story. The balance between comp. ratio, ignition advance and jetting can be restored, by adjustment.
I do not know the relationship between dyno results and performance on a circuit. It is not as simple as more power is good.

VooDoo Pistons said:

Thanks, Jim.

I could correct the Desaxe while I'm at it, but first I need to know what the factory targeted in their blueprints.

I have referenced the side-to-side axis of the cases by measuring off of a precision-ground shaft that I have installed in the bearings in place of the crankshaft. This shaft extends on the outboard side of both bearings through the cases and rests on v blocks along the X axis. These V blocks have been shimmed and tweaked such that the shaft is exactly 90 degrees perpendicular to the quill, or Z axis.

However, to get a reference point for the target deck height, I still need to true the cases in the Y axis which is why I am seeking the camshaft and crankshaft centerline reference dimensions.

Once I establish this theoretical deck plane I can then explore alternative milling options, but until I have that theoretical plane, I'm just guessing.

-Robert

Click to expand...

A better way to go might be to measure the desaxe on a Commando motor, and do some research on the performance of modern motors. Back in the olden days Japanese motorcycle companies used to publish their research papers in engineering institutes journals. Germany and Italy might also provide information. Noticing trends can provide predictability.

RobSS - the tangible benifit I'm hoping for is less wear on the piston skirts, less piston rattle, possible a slight improvement in smoothness and torque.

VooDoo - I checked mine by machining an alum tube dia so it fit the main bearings. Once installed I could measure down from the deck with a square and check the offset.

The measurement in the front in my diagram is .080" (not .030").

jseng1 said:

Please show us a photo of your jig for mounting the cases in your mill.

Click to expand...

Hi, Jim.

I have uploaded the pictures here:



Thanks,

-Robert

acotrel said:

A better way to go might be to measure the desaxe on a Commando motor, and do some research on the performance of modern motors. Back in the olden days Japanese motorcycle companies used to publish their research papers in engineering institutes journals.

Click to expand...

Hi, Jim.

Agreed. I could measure the Desaxe on the cases that are in the mill now, but what do I use as a reference point? From a reference "zero," I am -0.007" at the front corner of the cases on the the timing side and +0.006" at the center on the rear a total variation of 0.013." This precise data points can be seen in the images posted on the NOC forums.

Again, my first step is to know where I need to be before I start deviating.

Once I get a true reference plane, I could then print a variety of base gaskets on a dry build to see how the change in geometry affects other variables.

jseng1 said:

RobSS - the tangible benifit I'm hoping for is less wear on the piston skirts, less piston rattle, possible a slight improvement in smoothness and torque.

VooDoo - I checked mine by machining an alum tube dia so it fit the main bearings. Once installed I could measure down from the deck with a square and check the offset.

The measurement in the front in my diagram is .080" (not .030").

Click to expand...

Thanks, Jim.

Your technique is generally the same as mine except I used a sold piece of 304 precision-ground bar and then mounted the entire affair in the will and swept the measurements with a dial gauge fixtured to the quill. Everything was locked in place with the exception of the X axis so I could essentially "tram" the precison ground bar.

Good copy on the 0.080" dimension.

As soon as I can get the cam and crank centerline dimensions I'm looking for I can deck the mill and print up a "dummy" base gasket to evaluate its effects.

Stay tuned!

-Robert

I suggest a few base gaskets which cause a variation in slope might be a good idea, then work towards a good compromise. I never believed in the Commando motor until I worked with it. I suggest it is a different concept to the norm. I doubt that Norton made an error with their desaxe, however they were building a road bike, not really a sports bike.
If I was doing this, I would have a programme of rebuilding and testing variations in desaxe. With my bike, I had access to Winton Raceway which I knew extremely well, it made tuning improvements detectable. Changes in torque change gearing requirements. Bikes become faster and slower on various parts of a race circuit. So you need to know what the fast guys are riding. If a bike comes out of corners faster, it does not need as much top end to stay in front.

Whatever you do in changing desaxe, the result will not be a matter of getting it right first time. You will need to assess the data by feel. If you are used to a race circuit, you will know where the bike is better or worse. Load requirements in corners change. Do not use your best tyres for tuning.