Issue 29 (April 2015)
MG TC Stub Axles
Well documented problems with the TC front axles include cracking at the root radius with the upright. In addition to visible signs of cracking, the axles on my TC, undergoing a total restoration, were very badly worn.
I have read several articles about fitting new axles to existing stub axles and acquired a pair of new axles via John James (which came across ‘the pond’ from Bob Grunau – Ed).
I entrusted a local machinist, recommended by a work colleague, to remove the old axles (the ‘pins’ – Ed) and produce a suitable bore plus spotface on the rear of the remaining stub uprights.
For assembly of the interference fit components I planned to heat the stub uprights prior to fitting the new axles.
The new bores in the stub uprights were machined and honed so that on assembly with the axles a resultant diametric interference towards the upper end of an H7 / s6 interference fit of + .0019in would be produced.
This was chosen because using a domestic oven and heating the steel stub uprights to 200°C would, according to my calculations, produce an increase in diameter of .002in.
The left hand axle outer diameter was 1.1274 in and the right hand axle diameter was 1.1272 in so the corresponding bore diameters in the stub uprights were honed to fall into the range of diameter 1.1255 / 1.1263in with a proviso to achieve diameters as close to 1.1255in as possible, to ensure the tightest possible fit on assembly (+ .0019in maximum).
A small lead in chamfer, to aid installation, was also machined onto the axle diameter that would assemble into the stub upright.
In the articles I’d read, some very beautiful jigs have been produced to enable the axle to be held squarely to the stub upright during the assembly process. However, with no means of replicating / machining any such jig in my modest restoration emporium I looked for alternative solutions.
I had already acquired an old Jones and Shipman ‘Made in England’ Arbor Press for £30 via an e-Bay auction, which would assist in assembling the axles to stub uprights. All I needed was the jig to hold everything square.
As a design engineer by profession I am familiar with the statement ‘think of the problem in reverse’.
Having stripped the old worn bearings from the original front hubs these were found to be a nice close running fit onto my new axles due to spinning that had been taking place during TC0894’s past life.
‘Thinking of the problem in reverse’ produced an assembly of old wheel bearings, groups of washers (held together by masking tape), axle nut and the stub upright which, when the nut was lightly tightened up, ensured that everything was pulled square as shown in the photograph.
The two wheel bearings have identical outer diameters and the TC front hub has a deep lead-in inner diameter for assembly of the outer wheel bearing before the close tolerance diameter where that bearing fits is reached. As this diameter is slightly larger than the bearing outer diameters it thereby provides a running fit, which would act as the guide for keeping everything square. Additionally the overall diameter of the outer threaded hub section is just right for supporting the stub upright as shown in the photograph.
The distance from outer hub face to internal, outer bearing flange feature, is also sufficient to allow for the required movement of the axle during assembly into the stub upright.
I used one of my new front hubs for the jig.
As the assembly was to be effected by heating the stub uprights, everything had to be ready for a very quick assembly after removal from the oven to avoid any significant heat loss that would probably result in incorrect partial axle assembly.
Therefore, to avoid wasting any assembly time I bolted a thick aluminium base plate to the base of the press onto which two wooden guides were screwed. This ensured that the axle / hub assembly could be positioned quickly / correctly / directly under the press arbor as shown in the photograph.
Each axle was initially held in a vice so that once the stub upright was removed from the oven, quick assembly of the bearings and washers could be effected whilst also allowing easy gentle tightening of the axle nut to square everything up prior to sliding the resulting assembly into the front hub.
A length of scaffold tube was also at hand in case the arbor press lever required extending to increase the available assembly force.
Both axles assembled perfectly and the scaffold tube extension was required to increase the available assembly force during each axle’s assembly. The left hand axle having the larger outer diameter resulted in all my weight hanging off the scaffold extension to achieve correct assembly into the left hand stub upright.
A finished stub axle assembly is shown in the photographs below.
S. Cameron
TC0894 (under total restoration)
Ed’s note: I’ve confirmed with Bob Grunau that he still has the axle ‘pins’ in stock, but freight charges from Canada to the UK are a consideration (the ‘pins’ are surprisingly heavy) as is the potential liability for Customs charges. Bob can be contacted at grunau.garage(at)sympatico.ca {please substitute @ for (at)}.
Tim Patchett can supply newly manufactured stub axles at £650 per pair. He may well have a pair left from the last batch he had manufactured. Tim can be contacted at happypeople222(at)gmail.com {Please substitute @ for (at)}.
I have seen examples of this repair which also included the welding of the stub axle base. Another source specified Loktite.
I would be interested to hear of any failures from competition.
On no account should welding be contemplated. Both the stub axles and new spindles are made from alloy steel containing carbon and heat treated for optimum strength and fatigue resistance. If welded together, not only would the specific heat treatment be lost but the sudden heating and cooling down would embrittle the alloy steel resulting in cracks developing either during cooling down or as a result of external stress.
I realize 200 degrees C/392 degrees F. is nowhere near a welding temperature. I am wondering if the process of heating as described in this article (not talking about his jig assembly or any of that) and bore out dimensions are considered the accepted process for this fitting of pins? Is there another process, not mentioned, involving bore out to a dimension, and using a hydraulic press to fit the pin? Thank you-
I recommend buying a few kilos of dry ice pellets(solid CO2, Chillisticks, £25 by post). Pour a litre or so of methylated spirits into a lidded polystyrene box and then top up with dry ice pellets. This creates a liquid that will cool the spindles more thoroughly than just the pellets. You will see bubbles from the CO2 ‘boiling’ (actually subliming, but whatever) until no more bubbles means the spindle is at the same temperature as the coolant (lower than -79 deg C) and is then ready. I used no jigs but happened to have a 20kg block of steel with the perfect diameter hole to locate the knuckle. The knuckle was rushed from the oven and placed on the block exactly as the pin was lifted from the dry ice and dropped into place. It should have been a loose fit but in fact I had to drive the pins home tight with a heavy drift and a sledge hammer. That’s why I wouldn’t attempt by heat alone; I badly needed the cold shrink as well. Practise, practise, practise every tool and movement, operation and safety concern that you will use. Triple check that you have, say, the nearside knuckle and nearside pin, both clearly marked, and put the other side parts safely out of reach so it is impossible for you to assemble the wrong parts. This makes for an enjoyable and exciting evening, and once finished crack a beer and play with putting roses and bananas in the ice bath for 2 minutes and then hitting them with a hammer. Bob Grunau suggested 700F (371 degC) and domestic freezer which he finds just drop in with a 0.003″ interference. I think my oven only went to 280 deg C ish but the kingpin bushes in the knuckle seem to have survived with no change to the kingpin fit. Never put dry ice in a sealed flask; it expands and explodes.