AUSTIN SEVEN ENGINE REBUILD

Article kindly reproduced from the Herefordshire A7 Club.

Part 3 - Crankshaft & con rods

Part 1 – The crankcase, camshaft, oil pump, rear main housing.

Part 2 - The cylinder block and connecting rods

Part 4 - Main Bearings

Part 5 - Carburettors, manifolds, clutch and fan spindle

Part 6 - Flywheel

Part 7 - Reassembly

Part 8 - Attaching the block and cylinder head

Crankshaft
If you drive your A7 like a nervous granny, or the engine is to be used simply as a spare, then I believe it is perfectly feasible to retain an original Austin crankshaft, so-long as it has been carefully crack tested and found to be sound. A common area to find cracks on two-bearing crankshafts is the rear web just behind number four big-end journal but they can also crack on the journals and elsewhere. However, if you are inclined to belt along and rev the engine enthusiastically in the gears and/or seek an especially high level of reliability from your bottom end, then a modern replacement crankshaft is probably a good idea. This is simply because ‘two-bearing’ crankshafts tend to flex when revved hard (i.e. go round like a skipping rope) and this can (and probably ‘will’) eventually lead to a broken crank’ through ‘fatigue’ failure. Many original Austin cranks have been whizzing around for eighty or ninety years now, so it is mind boggling to think how many times they may have already flexed. You might be lucky enough to have a sound Austin crank that doesn’t need regrinding together with four beautifully matching conrods and I can understand the temptation to re-utilise these items. However, if the big-end journals (sometimes referred-to as crank pins) need regrinding and consequently the conrods need re-metalling, then this significant cost might be better spent as a contribution towards a modern crank.
The only modern crankshafts of which I have personal experience, are the 15/16” splash-feed ones made by Phoenix Engineering and I have found them to be perfectly satisfactory in service. My only note of caution is that my first brand-new Phoenix crank’ was supplied with one big-end journal 10 thou’ undersize. Happily, the item was immediately replaced without any quibble and subsequent ones have been spot-on, but nowadays I always check. Interestingly, I recently checked the big end journals on my original Phoenix crank’ after it had propelled a variety of A7s fairly enthusiastically for more than 20,000 miles and there was no measurable wear whatsoever.
So how do we measure big end journals? Well, when new or reground, the business part of a journal will start life as a precision cylinder, i.e. sides exactly parallel and constant diameter wherever measured. However, during its working life it will inevitably wear and because the heavy firing stroke load from the piston/conrod is applied to a particular side of the journal every other rotation - this will eventually cause some ovality. Similarly, any longitudinal flexing of the crankshaft can cause journals to wear in a conical manner. Therefore, to obtain a complete picture, we must measure the diameter of each big-end journal at six different positions. Firstly, a reading at each end (inboard of the fillet radius) and one in the middle, with the measurements taken at the same angle as the firing stroke load; then a second set of readings at right angles to the first three. The difference between the pairs of readings will reveal any ovality of the journal at each position and it is the maximum ovality that is of interest. Next, the difference between corresponding readings at either end of the journal will show the extent of any taper. Finally, a comparison of the end and middle readings will determine the extent of any ‘barrelling’.
An ideal big-end diametric clearance for a splash feed two bearing A7 engine is probably one or two thou’. This is where a conrod lubricated with very thin oil, will fairly easily fall under its own weight with the big-end bolts fully tightened. This suggests to me that an ovality of up to three thou’, and a similar amount of end to end taper or barrel shape can all be regarded as perfectly allowable tolerances without having to re-grind the crankshaft. Remember, whilst it may not be frighteningly expensive to have a crankshaft reground - the cost of the necessary conrod big-end white-metaling is considerable and several firms nowadays are quoting lead-times of three or four months.
Earlier, I mentioned ‘crack testing’ and I use the Johnson and Allen two part aerosol ‘magnetic ink’ method which is straightforward and seems to work well. It is used in the nuclear industry and by the military so, it should be OK for Austin Sevens! After thoroughly cleaning, polishing (with emery cloth) and degreasing the relevant areas, the Neopaint NPT16 ‘contrast aid’ white is applied and allowed to dry, which takes only a minute or two in a warm cosy workshop. Next, thoroughly agitate the black Neocol B black magnetic ink aerosol to ensure full dispersion of the magnetic particles in suspension and magnetise the item to be tested. I do this by holding a powerful magnet against the back of the area of interest with a sheet of paper in-between, to prevent spraying the magnet and don’t forget to keep powerful magnets well away from your pacemaker! Finally, the magnetic ink is sprayed onto the component surface and a careful visual inspection in good light will reveal the presence of any crack, as a discernible black line. This method is applicable for crack testing many other ferrous components including A7 conrods and will be referred to again in the next section. Obviously, if a crack is detected, then the crank’ should no-longer be considered suitable for use.

Conrods
If money is no object or you are building a racing engine, then I imagine it might be nice to have brand new conrods and several different manufacturers now offer suitable rods for our engines. However, I took expert advice some years ago that recommended sticking with Austin rods for road use (including enthusiastic use) so long as they have been carefully selected and equalised. My experience suggests this advice was sound because I have not yet broken a conrod and many of the failed ones I have seen were damaged for other reasons such as crankshaft failure or piston breakage. Interestingly, I have seen terribly damaged A7 engines where the conrods have bent but not broken which might suggest they are stronger than they look.
So how do I select conrods? I firstly check the fit of a new gudgeon pin in the little-end, it must be a firm push fit with absolutely no slackness, then check that the little-end pinch bolt thread is sound. In my view, these checks are of the utmost importance because I have seen several engines where poorly fitting gudgeon pins have caused considerable fretting to the securing bolt. If this is left unattended, the engine is almost certainly doomed to failure. I always use new HT bolts on assembly with internal shake-proof washers and a touch of thread-lock - all tightened very firmly (this will be covered in a future article).
I also file smooth and polish any potential stress raising marks on the flanks paying particular attention to the top of the web just below the little-end where many rods can be found to have cracks. The rods are then crack tested in this area using the Johnson and Allen two part aerosol ‘magnetic ink’ method described earlier.
Readers with an excellent memory, will recall the January 2014 Newsletter article containing a photo showing how A7 con-rods can be simply checked for fore and aft bending or twist, by passing a length of 0.500” diameter ground Silver Steel through all four little ends with the rods firmly attached to the crankshaft. For new members and those with an imperfect memory, we have a similar photo here. The rod should be a firm sliding/twisting fit without any binding or loose play.
Eddie assures me that slightly bent rods can be straightened satisfactorily. Very slight adjustments by cold twisting or bending are considered permissible. Usefully, the Dorset A7 Club website Technical Pages show how rods can be straightened using a vice as a press.
A later article will discuss engine assembly in detail, but it is perhaps useful here to mention that I favour being able to pass the conrods down the bores. This usually necessitates filing away the ‘bumps’ either side of the conrod big-ends if the bores are smaller than +60. This filing can often usefully be incorporated in the process of equalising the weights of the four rods. At this stage, it is also useful to ensure the big-end rods and caps are permanently marked to ensure correct positioning on re-assembly.
Fitting conrods to crank’ journals is straightforward in an existing engine where the journals are within limits and the white metal in the big-ends is sound. After thorough cleaning and careful inspection of the white-metal for cracks or other damage, the rods should be clamped in position on the crankshaft after lubricating the journals with light oil. I keep a set of old Nyloc nuts for this purpose and save a set of new ones for final assembly. Interestingly, it appears there are two different depths of 5/16” Nyloc nuts on the market and I prefer the slightly deeper ones because a socket fits more securely without binding on the cap. So, after torqueing the nuts to the required 18 lb ft and rotating the rod a few times to disperse the oil, check that the rod will happily fall from the horizontal under its own weight but without undue radial slackness. Slight (around 1/16”) fore & aft (rocking) play at the little-end is OK in my view. If the rod seems just a little too slack, it can be dismantled and the big-end mating surfaces rubbed on fine wet & dry paper supported on a truly flat surface, then cleaned and reassembled. Finally, give the big-end cap a reasonable thump via a stout brass drift and you might find this results in a better journal fit. If the big-end still seems slack the process can be repeated.
On the other hand, if the journal is a bit too tight, it will be necessary to indulge in the ‘dark-art’ of bearing scraping. I say this, because having discussed the subject over the years with a number of experienced practitioners, I have come to the conclusion that there are several different approaches. Anyway, the method I use (which happily seems to work) is as follows:

1. Prepare a ‘jig shaft’, ground to the required journal diameter plus the required bearing clearance. In our case say plus one to one and a half thou’ on diameter. Note: if you use the crank journal directly instead of a jig, you will end-up with insufficient clearance in the finished big-end bearing
2. Coat the jig shaft very thinly with engineer’s blue
3. Clean the conrod white-metal and bring it firmly into contact with the jig shaft and rotate it gently right around
4. When separated, the high spots on the conrod white-metal will be marked grey/blue
5. Use a sharp scraper to carefully remove these high spots, scraping alternately at plus and minus to the centreline of the bearing journal
6. Repeat from Step No 2 until the blue marking covers more than 75% of the white-metal
7. Repeat the whole marking and scraping exercise for the big-end cap

If the above process has been carefully carried-out and the conrod is cleaned, lubricated and assembled on the crankshaft, it should now happily fall from the horizontal under its own weight.
Alternatively, if we are building an engine with a new conrod/crankshaft combination, then the conrods will have to be white metaled and machined to suit the crankshaft journals. Sadly, the days are gone when every town had its own white-metal business and the relatively small number of remaining providers seem rather expensive. More irritating, is that some outfits now quote lead times of up to four months, which can be very inconvenient. Incidentally, I’m sure readers will recall that the September 2017 issue of this Newsletter contained a useful list of white-metal specialists in the South of England (thank you Ray).
Now, I have known people who have had white-metaling done without specifying exactly what they want. This might be OK if the firm is very well acquainted with Austin Sevens but I always make sure to specify the following:

o One to one and a half thou’ diametric clearance at mid journal
o An extra thou’ or so ‘bellmouth’ at each end of the journal
o Eight to ten thou’ longitudinal clearance along the crank journal (Woodrow suggests as much as 60 thou’ which seems rather a lot to me)

My approach gives a slightly looser engine than some firms might provide if left to their own devices but is based on advice I received many years ago from a very well respected authority on A7s.