Picasa Web Albums - Andy Bytheway

Search Images Maps Play YouTube News Gmail Drive More »

New Features | Help | Sign in

Restoration of a South African Windmill (8' SA Climax by Stewarts & Lloyds), May-June 2007

North of Cape Town there is a beautiful valley known as the “Riebeek Valley”. Windmills are a feature of the region but many are now in need of repair. The one seen here, a “Lloyds Aeromotor”, is working, but many of them are not. The one we are going to look at was a very bad case ...

First sight of the windmill we are working with (in the distance, over the gate) immediately suggests serious problems. The head unit is missing, the top of the tower is incomplete.

Happily, there is plenty of space to work in. The trees (the smaller ones) are almond trees and they need the irrigation that the windmill will provide. This windmill is an early example of a Stewarts and Lloyds “SA Climax” ... like the Aermotor, they were originally imported to South Africa, but subsequently were manufacturerd here in large numbers.

The fallen tail piece. Note the “S&L” in the white circle, and the stencilled “S.A. CLIMAX” - Stewarts and Lloyds made a lot of use of stencils, and some of the examples from the period of mass production have very poor paintwork, as a result of the stencil not being properly applied to the work. This one is not too bad, but (as will be seen) it was ultimately re-painted and re-sign-written.

A close-up view of the top of the tower suggests previous restoration work has been attempted. One upper leg section is missing and the two remaining sections are clearly not original. The deformation of the legs suggests that there has been a disaster, almost as if the head unit had fallen to the ground. Later investigation confirms that this was probably the case.

This early “S.A.Climax” is based on a three-legged tower that tells us specifically that it is pre-1957. No three-legged Climax towers have been made since then. Further evidence of previous restoration work is the mix of twisted-wire and threaded-rod tensioners between fixing points.

The parts of the windmill were found all around the 3,100 square metres of the plot. The wind wheel, and the vanes (or "sails") that belonged to it, were found in very poor condition, but the wind wheel was recovered, straigthened and put back to use; the vanes were of course irrecoverable and were replaced.

These are the other parts that were found in the grass, and in the garage. Anyone who knows windmills will see immediately that there are parts missing. Other pieces were found much later, in a old bucket at the back of a water drum. 'Never give up' was the mantra at this stage ...

The wheel hub is all there (later versions of the windmill used a split wheel hub) and the slot for the locking key is clearly visible in the housing for the main shaft. The locking keys were, of course, nowhere to be found (but happily they were later obtained from Southern Cross in Bloemfontein, as spare parts).

These are the brackets that hold the vanes off the wind wheel at the appropriate angle to optimise the effect of the wind. More than half of them were unusable, and it was decided to replace them all with new brackets, as new vanes also had to be purchased.

The gear and crank set was problematic. The two cranks had both been fractured and welded together - how on earth do you break two crank arms of such substance and strength? One original pinion gear (the small one at the front) sits with two new ones, and with a new side cover plate for the head unit.

Here is the most interesting piece - the head unit itself - that proved to be very much the worse for wear and none the better for an extended period lost in the grass.

The tower cap (right hand side) and the base plate (at the left) are interesting, because of the three-legged arrangement. Note how they are both made for 90-degree angle iron, and yet the tower is manufactured with 60-degree angle iron. One theory is that these castungs are from a three-legged windmill, but not the one in this property; another theory is that the angle iron has to be bent to fit the 90-degree housings. This mystery remains unresolved but the problem was solved by re-constructing the upper section of the tower with replacement 90-degree angle iron.

The lower end of the mast tube (the unit is up-side-down in this picture) is very badly corroded and worn. The slot at the right is the result of years of movement of a bent pump rod (that is the one that comes down out of the tower within this mast tube, and drives the water pump down in the bore hole), with no lubrication.

Heaven knows what has been living inside the head unit, but in the end it cleaned out OK and had no oil leaks.

Many of the parts have the part number cast in place. That made it easy to identify the necessary replacement spares.

Time to get to work! This was clearly a bad day, when everything had to come out of the tool boxes to find out what would work and what would not. The white box evidences the 1 ton chain hoist that was purchased in order to lift the gear back up the tower. The drill press has since done sterling service working on the replacement leg sections.

The first job was to put all the parts together to see whether the darned thing would even turn around. The main gear shaft was missing, but an aluminium TV mast tube substituted for the purposes of this assembly exercise (inside the head unit, at the right). The problem was that the head unit is extremely heavy, and it had to be bolted (using a convenient lug) to the bench. This was inconvenient, and so ...

... a small “top of the tower” assembly was thrown together with scrap angle iron to make it possible to work safely, and to turn the head unit as might be needed. Here the new "main shaft" is fitted, with new bearings.

The 'half moon' that is pulled up and down the mast tube in order to “reef” the windmill is a clever arrangement. Happily, the old half moon unit was found and once the chain had been replaced, and a new split pin inserted, it was ready for deployment.

This is the clever arrangement that gets the reefing chain out of the mast tube, and round the pulleys to the tail unit (not fitted at this stage, of course).

Getting the main gear shaft installed was not easy. The tolerances on the shaft are critical, but it had been machined to within a fraction of a “thou” (thousandth of an inch) short of what was needed. The new bearings are sealed (as can be seen), and so they could only be heated very marginally - insufficiently to get the new shaft through the bearing. The answer in this case was to ”flat off“ and polish the new main gear shaft until it was a perfect snug fit.

This is the (new) ball race that takes the whole of the weight of the head unit, wind wheel and tail unit. The mast tube is now the right way up, and the (new) end sleeve that locates it over the ball race is inserted. The flexible ring above the sleeve is a rain protection ring that stops rain water getting into the bearing. There is provision for greasing the bearing surfaces here (in the base plate) and above in the tower cap, but this load-bearing ball race is probably the most critical feature of an SA Climax windmill that must have regular attention. The old bearing, needless to say, had completely disintegrated.

A clearer view, showing the (new) stop plates that are to be bolted over the mast tube end-sleeve when assembly is complete. They lock it down and prevent the head unit from rising (a most unlikely eventuality, given the weight of it, unless the whole thing is turned upside down).

A good trick, in a restoration job like this, is to find an old windmill and beg, borrow, or steal it! It will be a source of spare bolts, angle iron, and inspiration. For this project, a neighbour kindly agreed to donate this old 14' SA Climax, which he wanted out of the way in any case. (These photos were taken by Vidi - thanks!)

Taking it down presented a serious mental challenge ...

.... solved by clearing the undergrowth ...

... climbing the tower to affix a rope ...

... and pulling the whole thing over with a tractor (unseen, at the left).

Over she goes ...

A pity that some of the gear was bent on the way down, but that was the decision of the owner. Not enough labour and lifting gear to do a piece-by-piece disassembly.

From the donated windmill, here are three new legs for the restoration of the top section of the tower, cut, drilled and ready to go. Note the short section of 60-degree angle iron from a leg of the original tower.

The new vanes are assembled to the inner wheel rim (thanks, Pule!). Despite the weight, it was decided to assemble the whole wheel on the ground and hoist it to the top of the tower as a unit. As can be seen, some final adjustments are needed to the setting of the mounting brackets (right hand side).

Jonathan, who proved to be the one with the best head for heights, sets about the assembly of scaffolding around the tower. We were at this stage still wondering exactly how the head unit will actually be hoisted into position, but the scaffolding seemed like a good idea. Note: fitting square scaffolding to a triangular tower presents certain problems, solved here with an impromptu timber-based solution (the "V" at the right). It worked well.

The finished scaffolding, and a test run of the hoisting plan, with the new leg sections assembled with the tower cap and suspended within the lower structure. Happily, it was established that the whole kit and caboodle could be lifted INSIDE the old tower, and that is what was done in the end.

Checking the chain hoist for clearance. An extended 8m load chain had to be fitted, to avoid lifting the final assembly in three or four separate ”lifts“.

Jonathan is happy, the “lifting plan” is complete.

Yes, the whole unit really did go up this space, in the end!

Back to the workshop - here the new spare parts are on display, freshly arrived from Southern Cross in Bloemfontein, who now own and service the "SA Climax" range.

Not everything fits first time! The angle grinder is put to work on a non-critical part that needed some adjustment.

All the original internal parts are now checked and pressure-cleaned, and they are assembled with the new parts. Split pins are replaced, and bent down to secure the assembly.

This detail shows one of the “oil rings” - the large wire circle hanging on the upper spindle. For a while the function of these rings was unclear (to your un-informed narrator), but they really are rather clever. As the unit turns, they fall on the lower “works” (which is running in an oil bath, of course) and pick up the oil and carry it to lubricate the upper bearing surfaces ...

... like this. The oil bath is not yet filled here, but you get the general idea!

Nearly there. Not far to the tower - just a few details now ...

... like the bonnet (ugly, but necessary!) ...

... and a coat of red primer, and a willing crew. Off to the tower ...

Lift the head unit up a little way ...

... quickly insert the new upper tower section ...

... and assemble the two ready for the final lift.

The final engagement of the mast tube sleeve in the lower base plate, and securing the stop plates to hold it all together. Much easier done at ground level than at a great height.