Early S7 Forks Rebuild




In March 2016, I had a 'phone call from the owner of an early S7 that was showing signs of extreme wear in the forks – general rattling and clunking over the all-too-common potholes. The head-stock bearings were OK and the movement seemed to be between the sliders and the stanchions. It seemed likely that the problem was extreme wear in the phosphor bronze bushes. Sadly, those bushes are no longer available.

And what is worse, the design of the early forks is completely different from the later S7 Deluxe forks and, unlike them, does not permit the bushes to be removed easily. The early stanchions have two different outside diameters at the business end of the stanchions, with two correspondingly different diameters on the inside of the sliders. The bushes, unlike the later ones, have no retaining mechanism except for a very tight interference fit into the sliders. The stanchions are able to slide all of the way out, leaving both bushes behind. The lower ones are a long way down, near the bottom of the sliders. The owner of the bike had tried a makeshift slide-hammer tool but had not been able to make any impression on the bushes.

The application of heat would not help to loosen the bushes' grip because the coefficient of thermal expansion of bronze is 50% higher than that of steel. Heat would only make the bushes grip even tighter unless a temperature differential could be achieved between the two metals. Even ignoring the inconvenient fact that the two metals are in intimate contact, a 'quick and dirty' calculation shows just how impractical that would be.



The owner and I agreed that he would remove the forks from the bike and post the stanchions and sliders to me so that I could try to find a way to get the bushes out. Replacing them with new bushes should, with luck, be a more straightforward problem to solve.

The forks arrived a couple of days later and I stared at them for a while, hoping that a good shake would encourage them to fall out and wondering what kind of nutter would design such a puzzle. The bushes did not drop out and I believe that the answer to the question is known to history. I wondered if Mr Poppe had copied the design from the early BMWs with telescopic forks, the R12 or R17 for example, but they look quite different.

We discussed a number of possible approaches, including:

  1. making a special tool to pass through the bushes to grab them from behind, so they could be pulled out;

  2. grinding or cutting through the bushes, to relieve the interference, so they would drop out;

  3. Filling the bottom of the sliders with grease and hammering the stanchions into the sliders, to produce a hydraulic lock to force the bushes out;

  4. cutting off the bottoms of the sliders, knocking the bushes out and welding the sliders back together again.





After discussing the pros and cons of each these methods for a while, we decided that the fourth option, although the most radical, was the most likely to succeed.

The welding would need to be strong, reliable. oil tight and very neat. My welding is adequate for many things but this was definitely a job for a professional welder. I telephoned the man that I use for all specialist welding jobs. He is an expert welder who has a workshop in a re-purposed farm building just a few miles away. He assured me that it is a straightforward job and the weld would be stronger than the steel the sliders were made of and virtually invisible, once painted. I believe him.

However, the first job was to make two dollies of different sizes that would be used to drive out the bushes from below. Knocking them out with a brass drift might work but it may damage the bushes or even the precision-machined inner surfaces of the sliders, so that was rejected. So I set about making the dollies from mild steel bar.





The only difficulty with that approach is that the dolly for the larger diameter bush would need to be passed up through the lower and smaller diameter bush. That is unless the sliders were cut into three pieces, not two. I discussed this problem with my son, Peter, an engineering designer, who immediately came up with a neat design that would be big enough in one dimension to knock out the larger bush but small enough in another dimension to pass through the lower, smaller bush. Better still, the design was simple enough to be made with the machinery that I have in my workshop.

However, looking at the design made us wonder whether it could be used the other way around. Could the dolly be passed down through the upper bush and used to knock it out with a slide hammer? Since the upper bush clearly had to come out first, it was worth trying it anyway. There was nothing to lose and, if successful, the same method could be attempted on the lower bush.

So I drilled, tapped and chamfered the centres of both dollies and adapted the end of the rod of my slide-hammer to fit. Clearly the dolly could not be inserted with the slide hammer rod attached, so it had to be passed through the bush and manoeuvred into position with a magnetic probe (one of the advantages of using mild steel) before inverting the slider so that the dolly would be held in position by gravity.

Then the slide hammer rod had to be screwed into the dolly from below with only the force of gravity preventing it being pushed out of position while fumbling with the rod. At that point, careful cutting of a nice clean thread with a deep chamfer enabled the rod to find its mark easily. Careful preparation always pays off.





The first bush was very tight and took a lot of hammering but it eventually came out, as did the upper bush in the other slider. Flushed with success, I then applied the same technique to the lower bushes, which were even harder to get out because the effective length of the slide hammer rod was considerably shorter.

Once the bushes were out, I could measure them and work out what size the new bushes needed to be, allowing for the considerable wear on the inside of them. Fortunately, there was very little wear on the working surfaces of the stanchions. Once the outer diameters of the bushes were accurately measured with a micrometer, it became clear why they were so hard to get out. All of the bushes were abnormally thick and had been pressed in with nearly 0.004” of interference. That is a lot and it explains not only the amount of effort required to get them out but the damage to the rather stretched thread on my slide hammer as well. Fortunately it held out.





Curiously, the lower bushes were turned out of a solid piece of metal but the upper bushes were rolled from a flat bar. I assume that was to minimise material costs.

The next task was to work out what to replace the bushes with and where to get the material from. The original bushes were made from phosphor bronze, a material that is used much less frequently now that Oilite has gained so much popularity. Sourcing a large enough piece of phosphor bronze to make four large bushes was likely to be expensive. Then there is the work involved in machining them down to within 0.001” of the required size. Oilite bushes have much better availability and will probably perform better in an environment where lubrication is sparse.




For those who are not already aware, the early S7 forks have no hydraulic damping mechanism inside the fork legs and, therefore, no oil sloshing about. Lubrication is provided by mist and occasional seepage from an oil-soaked wick in the centre of the stanchions. Oilite holds oil in pores in the material itself and so offers a better solution.

Incidentally, Oilite is not made from sintered phosphor bronze, as some sources state. It is made from sintered bronze. Phosphor is added to bronze to assist with the pouring and moulding process and not for the properties that it confers on the finished metal. Oilite is formed by pressing tiny particles of bronze at very high pressure. This results in a porous material that will hold oil but machining it requires extra care. It needs to be machined using very sharp tools, otherwise the machined surface becomes “smeared”, sealing the pores and reducing its effectiveness. I avoid machining Oilite if I possibly can but sometimes it is necessary to do so and then I use a new carbide bit and perform the cut at high speed to minimise the damage.





As it turned out, I managed to source bushes of the correct nominal dimensions for the upper bushes. But, for the lower bushes I had to make do with the correct nominal outside diameter and a somewhat smaller inside diameter, which needed to be bored out to the correct size.

Nominal sizes are one thing but the ideal finished sizes are another. Normally, when fitting bushes, I decide on the amount of interference that is appropriate, press in the bush and then ream it to the correct internal diameter. That is necessary because the bushes will always shrink slightly when they are pressed in and, if not corrected, the bearing will be too tight and might seize.



As an aside, about a quarter of all of the Sunbeams that I have worked on have either been suffering from a seized half-time gear bearing or bear the signs that it has previously been seized. That is usually because a new bush has been pressed in and not properly reamed to size.

In this case, however, there was no practical way of reaming the bushes once they were pressed in. That means that the amount of shrinkage had to be estimated, the bushes machined to the exact sizes and pressed in. Then they had to be proven by inserting the stanchion and testing for a good sliding fit, with minimal play. If there was too much play then the bush would have to be removed again and added to my scrap bin. Another expensive new bush would then be needed for another try. If too little play then the bush has to be removed and the internal diameter adjusted ready for another attempt.







Potentially, that process might have involved a number of insertions and subsequent extractions to make the required adjustments. Therefore, I decided that the new bushes would be inserted with somewhat less interference than the original ones. Neither my slide hammer nor my arms would have withstood the strain otherwise. I decided that 0.002” of interference would be perfectly adequate to ensure a permanent fix, while facilitating some trial and error.

As it turned out, the first fitting of a lower bush resulted in slightly too tight a fit. I removed it and adjusted the internal diameter to give a little more clearance. Upon re-fitting, it was clear that the fit was perfect. Then I used what I had learned from that first insertion to calculate the shrinkage more accurately and applied it to the machining of the remaining three bushes. All of them fitted perfectly so that the stanchions slid freely but without any noticeable lateral play. I cleaned all of the parts in the solvent tank, dried them with the ubiquitous blue paper towel and lubricated them. Job done.