Cam followers - Andover to the rescue?

I guess I must be doing something wrong.

I am sure you know I have run many tests on big cams in my spintron. But I have yet to be able to complete a long term test with the PW3. I did a careful setup yesterday and was hoping to get a clean 24 hour run. It was running smooth at 5500 rpm last night when I turned in. And this morning at 7:00 it was still running but something did not sound right.

Just one more broken pushrod. Nothing else was damaged.

Wow Jim, what a strange place to break, and with no other signs of issue / damage.

Some strange stresses going on there...

Yeah, I hate to quit but I think the PW3 is going to go back where it came from and I will stick to something easier to get along with. Like maybe an N480...

Snotzo has confirmed my suspicion that the pushrod fracture was due to a harmonic.

I have watched the effect harmonics has on the valvetrain with my high speed camera. Sometime it makes me wonder how the parts stay together as well as they do.

I have some video of a pushrod that has a serious S curve to it at one certain speed, and the springs can get really out of shape.

I think the key is the engine does not dwell on (held precisely at) any one rpm range for too long - not a marine diesel.

When you run a Commando hard, the harmonics and other nasties are transient.

In industry there are acoustic hammers that measure the harmonic frequency(s) of an object that it strikes.
Could be helpful in an exercise such as this given that the exciting frequency is known (1/2 RPM). 5500rpm = 46Hz
Ideally you don't want items with lowest harmonic below maximum exciting frequency.

So maybe Steve Maney was onto something with his steel pushrods ...

Fast Eddie said:

So maybe Steve Maney was onto something with his steel pushrods ...

Click to expand...

Definitely Nigel - harmonic frequency is proportional to stiffness and, as a material (ie: excluding design differences) steel is roughly three times as stiff as aluminium.

The other thing to consider is that, as I've mentioned before, aluminium does not have an endurance limit so will die of fatigue.
Without examining it closely, that failed pushrod of Jim's looks like a brittle fatigue failure.

Dances with Shrapnel said:

I think the key is the engine does not dwell on (held precisely at) any one rpm range for too long - not a marine diesel.

When you run a Commando hard, the harmonics and other nasties are transient.

Click to expand...

You are correct. If I would have run the engine at another speed or varied the speed the pushrod failure would not have happened. I just happened to pick a speed where the spring and pushrod frequencies were similar and the aluminum fatigued first.

Fast Eddie said:

So maybe Steve Maney was onto something with his steel pushrods ...

Click to expand...

The steel pushrods will change the frequency of resonance. And would likely survive longer. But if resonance is maintained either the pushrod or the spring would fail sooner or later.
Of course in real operation you would not likely maintain one speed long enough to ever reach the fatigue limit of either material.

jseng1 said:

Welding hard material onto the steel lifter seems like one of the best ideas. This is what SRM decided to do instead of brazing on a stellite pad. The same welding process (not brazing on a pad) could be used with Norton lifters (if made of steel). I bought as many of these as SRM would sell me, sent them out to customers and haven't heard of a single complaint.

The Norton lifter is a complicated shape and expensive to machine - that's its main drawback (and its weight). The BSA lifter is nearly the same design as the Triumph lifter and the longevity is pretty much established in high mileage bikes (compared to Norton lifters). But the overall shape or design isn't the biggest problem. Whether its a Norton, Triumph or BSA lifter - the main issue is the pad material. So why aren't lifters with welded on cam contact surfaces being made at this time? Seems like a no brainer.

Hint:
SRM used stellite #6 hardweld - approx 50 HRC - softer than Hero 56 or 60 that is used on cam lobes by people like Web Cam (I think), So it should be compatible. BSA used 5 to 10 hardness difference between cam and lobe.

Click to expand...

No hint needed. Welding is an expensive process, and what do you think the hardness of these one piece tappets is, it is Stellite msterial.
It you buy SRM Stellite tappets, or any Stellite weld products ask for the serial numbers of the rods used as they are all traceable - how do you think I know that. The manufacturer who makes the Stellite rods was ask to look at hard facing tappets. There was nothing in their armoury that could could be applied cheaply that would achieve the hardness needed by any method, weld, laser or plasma application.
Yes, hardness separation is an ideal starting point when you have one cam material, but Nortons use 4 different cam materials with different hardness, so that also complicates the issue when trying to find something that fits all. If it was just one cam and tappet that would have been easy. Make no mistake this is a challenge to the finest brains in hard facing technology in North America and Europe, it is not Joe blogs casting and victorian machine shop involved in this project.

Just thinking off of the top of my head here, but...

if making one follower type to suit every cam type is so difficult... how about making specific ones?

meaning, one type for iron cams, one for steel cams and one for hard welded cams...?

How on earth did Norton make over 200000 followers half a century ago ?

Time Warp said:

How on earth did Norton make over 200000 followers half a century ago ?

Click to expand...

Cams were case hardened and that was your only choice. One cam, one follower.
Allegedly, the originals were one piece cast iron, then the version where the tips fell off, so they have struggled in the past to make 200000 tappets in the past.
To make individual ones would need different materials, some would cast some will not. Then it is back to the two piece type, some tips would braze, some not. Even the spigotted type needed a spot weld to the tip prior to brazing.
The ideal material would be cast iron, easy to cast and can be made glass hard, the problem is that iron based material introduces more problems, each tappet would need heat treatment to ensure consistency, furnace does not not achieve that even with gas quenching. Then there is the issue of grinding material that heat is treatable, if the cut is too much or coolant, not enough, wheel type etc can cause the interface to soften or change the structure in the layer. Best to use a material that has a natural hardness not affected by further processing.
How do I know this, testing the two piece tappets at each stage. 6 years ago a batch of tappets where the body is glass hard was found when the grinders refused to grind then.
I think a part two to the tappet story in the Source is needed.

In Europe the diesel engine development for performance car applications created the need for very high compressions, this coupled with the greater downward pressure from the combustion event due to increased power meant a high failure rate in the top big end shell even when using the highest fatigue strength bearing materials eg the leaded bronze with overlay plating also known as Trimetallic bearings. This was because the hydraulic wedge oil film was being breeched by the pressures from above so contact was being made between the big end journal and the top shell bearing causing the top shell bearing to fail.

This seems to be close to what is going on between the cam and cam follower, the forces from the cam profile are breaking down the oil layer and there is metal to metal contact for a significant period and the materials are failing.

The solution for the top bearing shell was a process known as sputter, whereby an even higher fatigue strength bearing material was deposited onto the steel shell.

The process

https://en.wikipedia.org/wiki/Sputtering

and the sputter bearing.

https://www.highpowermedia.com/Archive/what-are-sputtered-bearings

https://www.substech.com/dokuwiki/doku.php?id=sputter_bearing_overlays

There have been further developments, eg a plasma deposited layer of Aluminium Tin over a leaded bronze sub layer over steel.

Fast Eddie said:

Just thinking off of the top of my head here, but...

if making one follower type to suit every cam type is so difficult... how about making specific ones?

meaning, one type for iron cams, one for steel cams and one for hard welded cams...?

Click to expand...

The issue then is a business one, you need in your example 3 times the stockholding, of visibly identical parts, or visibly marked parts, if you include +5 thou variants that is 6 types, you need to create and manage your stock by variant part number. And so do your dealer network!

And you need to complete your R&D and testing on all variants.

It may not be an aerospace application but we are talking aerospace complexity and parts management, guess what cost that comes with!

Refering to pushrods for racebikes that revv higher. You will reach a point where the springs can't keep up, the valves will bounce, tangle and destroy the top end. You can increase spring pressure but then the cam flexes and pushes back after valve bounce - making the valve tangle problem even worse. The other option is to lighten the valve train and that includes the pushrods and this is important in a Norton because the ratio is 1.14

Stronger 2024 alloy alum pushrods are an option compared to the softer stock alum pushrods. But if you go to steel pushrods then you increase the weight and the valves will tangle at a lower RPM. Lots of positive things are said about steel pushrods but if you make a steel pushrod for a Norton to match the weight of a high strength 2024 alum .058" wall pushrod, the steel pushrod would have only about .020" or so wall thickness. The 2024 alum tube of equal weight is nearly 3 times thicker in wall thickness. That thicker wall increases the structural integrity and resistance to buckling.

I dare anyone to race a steel pushrod with .020" wall thickness. I've tried to make them and had problems with the ID loosening up and deforming where the pushrod tip is presses in. You don't want the ends falling out.

PS - the steel pushrods used in Nortons have .035" wall thickness.

Really, the Maney steel pushrods that were in my racebike for years and are now in my streetbike are lighter than the original aluminum pushrods. I haven't pulled them apart to see what the wall thickness is.

Madnorton said:

I think a part two to the tappet story in the Source is needed.

Click to expand...

I like the last one with reference to the Classic Attacks.

Any follower Andover Norton produces would be better suited to the camshaft they stock initially ?
A matched pair.

Back in 2011 Moto Guzzi on their 8V engines were using flat followers with a DLC coating on the base, it was known to fail with dire consequences in some cases and were forced to supply (begrudgingly) roller kits to rectify the problem and roller only from 2012 models of the big blocks.
So it seems to be nothing new, you would still think there would be coatings that would adhere and remain reliable.

Keep at it.