Meanwhile I had my customer's loco running, and co-operated with Rob when it came to shunting a couple of wagons off the back of the works train, which had Charlie at the other end.
On Tuesday, New Years Eve, we paid a visit to Scunthorpe. Our plan was to concentrate on D2128, and we took my old stick welder with us for Andrew to secure the upper relay lever on the handbrake linkage. With that achieved, I squeezed down under the side of the engine and removed the cover on the side of the transmission where the pressure regulator is located, and after catching the oil that emerged, inserted a shim (aka an M12 washer) over the springs as we had done with the original transmission we had fitted. That done, the loco was taken outside.
But all good plans, etc. Tom had been parked up the side of the shed, where locos go that are in the doghouse, following a report that the flycrank on one side was still being polished by the knuckle pin. Sentinel went through a number of versions of knuckle pins, the problem being that the “clean paper” design of the Sentinel rod-coupled running gear came perilously close to the maximum width under L1 loading gauge, so much so that for the knuckle pin to protrude much beyond the flycrank cover would take it outside. [To complete the story, under Thomas Hill the knuckle pin was re-designed again and taking the bull by the head-handles, the pin was brought out into an M42 nyloc nut and a grease nipple on the end. Yes, it was outside L1 and yes, our competitors made a big thing of it – without admitting that it was just a nut on the side of the side rod - but the only place we ever had a problem was a certain cement works where a concrete wall was so close that the end of the knuckle pin struck it. The customer sent along a man with a jack hammer and a groove was taken out of the wall until the nut and nipple cleared...]
Anyway, after early experiments with circlips, the most common knuckle pin screwed into a plate on the outside of the side rod which was then locked with setscrews. To prevent the knuckle pin itself rotating, a “feather” (in reality about a half-inch of silver steel) dropped into a hole in the bottom of the pin and engaged a notch in the side rod. When we went round to examine Tom, it was obvious that the knuckle pin had backed itself out so after moving the loco to get the pin in a good position, we carefully knocked it out (and to our surprise, the side rod didn't drop) and sure enough, the feather was missing (and the hole and notch filled with dirt). We cleaned everything up, made a new one out of a bit of steel rod in the workshop and reassembled it all, adding Loctite to the outside thread for god measure. We'll monitor it for the moment but hopefully that should cure it.
So, back to D2128. The Cummins was fired up and after building up the air supply (and no, the reset unloader still didn't function, but the sanding valve didn't seem to leak as much) Andrew engaged first gear and almost immediately there was a clunk as it engaged. I say almost immediately – it is still not as fast as it ought to be (which is instantaneous) but less than half the time it took before. Second came in with the same reaction time and we shunted it up and down and around as the December daylight fell. Later Andrew watched the servo pressure gauge as I operated the gears – the base pressure is up 15psi or more and well into the approved range, but more excitingly, as the gear engaged it rose up into the range the book says (before the rise was so low and hit-and-miss that we were nervous of the clutch pack slipping). We drove the loco back in the shed – I wasn't going back under with a warm engine, but another shim should bring the pressure up towards the top of the manual's quoted range and may get the immediate response we expect.
Andrew had observed that the engine temperature rose rapidly given the ambient and what little running we had done, but I have always expected that we would get re-circulation of warm air within the engine casing heading back round through the radiator: it is time to get some sort of ducting made up to ensure it draws fresh air in.
As for the clunk when engaging, well, the RF11 final drive on an 03 is mounted to the frame with fitted bolts, and on these and 04s the fitted bolts fret with age and the gearbox starts to pivot over the jackshaft centreline. (You can often tell because apart from the clunk, the wingnuts on the oil filler cap scrape a clean patch on the underside of the cab floor). We will worry about this at a later date, for now it's get back to those outstanding tasks we had lost the enthusiasm for. But I fear we may have one serious issue with Scunthorpe when it comes to running the loco – to whit, keeping it within the speed limit. Take it up to second and open the throttle and whee! Does it go!
A final job for the day was to extract the Driver's brake valve which is in need of TLC and in any event had been changed in the past and left without the securing bolt to clamp it to the bracket, in other words it was held in place by the pipework. With the performance in getting that apart the valve clearly didn't require the bolt, but our replacement valve should be done properly.
New Years Day saw us back at Rowsley. We need to get back to Libby which has been languishing somewhat since the casing parts came back from shotblasting. But a couple of parts had become detached during this handling exercise so it was time to get them back in place and hopefully ensure that the casing was back to its original rather subtle shape. This inside work was fortunate as the weather outside was wet and windy. I had some conduit to change on the Matterson units. Another piece of casing top had been professionally re-fabricated some years ago but was slightly wrong – Andrew decided to cut and re-weld it to get it right, and in due course correct a bodge I made in another piece to get it to fit.
The fan dive on Libby was extracted, and the shaft end built up. Looking back, the shaft and hub are notionally to the same dimension, but the hub was machined in Sheffield and the shaft back in my days at RMS Locotec when we had a machinist in Dewsbury. Clearly they must have had independent West and South Yorkshire precision rulers. Anyway, Andrew built the end up with weld and between us made a good fit of it, then painted the various parts ready for reassembly.
Andrew was back at work on Thursday and Friday, but I had loads to do, much of it directed at the Shed Project. From collecting material for the Official “You are now entering a construction site” notice board, to reading through Health and Safety and Induction manuals, ready for the contractor to start on Monday 13th. But on Friday I was collecting the I beam sections needed to make up the extension pieces, and Saturday at Rowsley was largely devoted to a production line manufacturing process. We have 10 to do and 3 are now finished, and two more started. I set them out, Andrew tacked then continued fully welding while I got on with the next one and twiddled my thumbs. A minor “oops” was spotted – the base plates were derived from the drawings I had on CAD that we produced from measuring our columns, so have 22mm holes for M20 bolts. But of course I ordered this lot before we won the holding-down bolts on e-bay, which are M24. So another little job will be to drill the 22 dia holes out to 26 with the mag drill: there's only 40 of them.
We also refitted the driver's brake linkage for the train brake valve on the Drewry, and Andrew changed a leaky regulator under the cab floor. It turned out to be blocked with rust – this loco was piped before I learned the need for improved filtration when working with industrial valves.
Today we were at our grandson's first birthday party, more of a mother and toddler group meeting with decorations, if you see what I mean. Hence arriving home late and writing this up later. Stick around for the next few weeks as the weather attempts to thwart progress at Darley – already the two sections we dug out to recover the ballast have become impromptu lakes (and Rob reports that that is where the water table is at the moment,) but I am assured that pouring concrete into a water filled trench is not a problem. Although the formwork floating away might be.