You need your head examining!
Several times people have said this when I say I have three BSA cars.
Some think they have little street cred what ever that is. We BSA owners know better, and when things go well are amply repaid all the work and expense in achieving that perfect run. We must admit however that like all old cars things can go wrong. Head Gaskets on 4 cylinder engines for example. At this stage Vee Twin owners can retire with a wry smile.
Unfortunately BSA 4 Cylinder engines have a habit of blowing head gaskets, some-times repeatedly, in spite of the owners best endeavours. So lets try to establish possible reasons and remedies.
Firstly numbers 2 & 3 cylinders are close together, this is usually the area in which the gasket fails. Secondly these engines frequently overheat and boil, often followed by head gasket failure. Overheating can be attributed to poor design, the front wheel drive engines have the inlet and outlet of the cylinder block at the same end, this does not help the water to circulate. The position of the radiator over the differential limits its height and as a result to circulation of the coolant is restricted.
Radiator cores are in many cases years old, and partially blocked as are the water-ways in the block. Other causes can be ignition timing too far retarded, and carburation weak. In many cases these problems can be cured with careful set-up and an electric boost pump to improve circulation. Much has been written about these pumps and details are included in the Issue 2 Trike Manual, so I will not dwell on this aspect, rather look at the situation when in spite of attending to the points mentioned, gaskets still blow. BSA’s use a copper/asbestos (substitute) gasket which under the right conditions is perfectly capable of giving a good seal, however the surfaces of the block and head must be flat. If the engine has boiled it is likely that at least the head will have distorted, if not the block. Distortion usually takes the form of a sunken area developing on the narrow bar between cylinders 3 & 4. Determining if this area has distorted is not easy. You may be able to spot trouble with a good straight edge set along the head or block, any visible gap must be rectified. A better way with the head is to lightly spray the clean surface with a primer spray, then lap the head on a sheet of plate glass of lapping plate. The plate must be larger than the head. You will then be able to see if the paint is removed evenly or remains in the critical area.
The same technique can be used on the block by lapping with the head after removing all the studs. Great care must be taken with a partly built engine that lapping compound does not get down the sides of the pistons. If either surface is low in the region of the centre bar then it will have to be machined. In the case of the head, your local machine shop should be able to do it. With the block in the car there is no alternative but to remove it, strip the engine and again take it to the machine shop.
If at all possible have the head and block milled or flycut. Many shops surface grind the heads, this produces a lovely shiny surface, unfortunately the heat from the grinding operation frequently causes more distortion and nothing is gained. In my view the highly polished ground finish is not desirable, better to have fine machine marks that help to key the copper gasket. Photo 1 shows a head towards the end of a flycutting operation. At this stage
0.006″ have been removed and you can still see the low area which has not cleaned up. This engine required 0.010″ off the head and 0.006″ off the block to clean-up. The gap when the two were placed together would have been 0.016″, as this is localised no amount of tightening of the studs would pull the gasket up tight at this point, and early failure would have resulted.
Photo 2 shows another method of checking the block given a flat (machined) head. Position the head as shown, and check for any gaps with a 0.0015″ feeler gauge, the gauge should be trapped wherever it is positioned.
Having got both surfaces flat, then check the new gasket. Never re-use a gasket except in an emergency as it will have lost much of its resilience.
The area to be checked is again in the centre. Lay the gasket on the head as shown in Photo 3. Locate it with 3/8″ bolts, no nuts! as pulling the bolts down will damage the gasket, the object is simply to position the gasket. Look for any overlap on that centre bar. The ideal is when the edge of the gasket lines up with the step surrounding the valve pocket. If the gasket overlaps at this point as in the photo, the lip or folded edge of the gasket will not be properly trapped when the head is pulled down. In extreme cases gaskets can overlap so far as too touch the valves.
If an overlap exists try with a suitable piece of bar or tube, to tap it back as shown in Photo 3, either tap the gasket using the weight of the bar or with a hammer on the bar. The object is to gently knock the gasket back until it is flush with the edge of the valve pocket as in Photo 4. 
This obviously needs to be done at both sides of the bar. Some thickening of the gasket will take place however providing that the gasket is nominally flat and not creased or buckled it will squash flat when the head is pulled down.
Now if the studs have been removed from the block re-fit them, applying a bit of silicone compound the lower thread, this stops coolant leaking up the stud. Jointing compound is not normally recommended for copper/asbestos gaskets, a thin layer of grease can be used or I use a none setting, thin jointing compound applied with a brush in a very thin layer to the block face, then to the top of the gasket when its in place. Silicone gasket has insufficient strength to be of any use at the high pressures experienced on a head joint and so would be useless.
Tighten the head as described in the manual or owners book gradually tightening to 35 lbs/ft.
If you haven’t a torque spanner I would try hiring one, failing that a socket spanner with a handle about a foot long and a good pull should do the job. Unfortunately with a re-build all too often at this torque the studs will pull out of the thread in the block, console your self that its better to pull out now than loosening off on a run and blowing another gasket. Stripped threads can be repaired with Helicoils but I don’t trust them in cylinder heads so I would use a stepped 7/16″- 3/8″ stud in an oversized hole but that’s another story.
Lastly but I am sure a frequent cause of failure, check the torque of the nuts again when the engine has had its first short run, and don’t give it full throttle until you have.
My head has now had enough so I’ll leave it at that.
Mike Scott-Coomber
An Achilles heel cure?—–
It is probably generally accepted by the BSA FWD fraternity in the know (and even by myself) that the 4 cylinder ‘short’ engine, whilst being a sturdy, if plebian, unit, has one notable weakness which, far from giving it character, can be downright annoying. I refer of course to the serious lack of cooling water in the vicinity of the two centre exhaust valves numbers 4 & 5
These two valves constitute the hottest spot on the block and this results in all too frequent overheating problems, short valve life, burned seatings and at worst, cracking of the block. Such cracking usually manifests itself initialy as a hair line crack on the valve seating which hopefully can be rectified by fitting valve seat inserts. If this remedial action is not taken quickly, however, the exhaust flame will burn a V-section out of the seat in a very short time, Once the crack moves into the block it will spread down the valve port to the valve guide and across the block face to the cylinder bore. And this can happen on each of the two centre cylinders.
The trike block which I am now replacing was professionally welded about six or seven years ago, having exhibited all of the above symptoms, and re-sleeved to standard bore all at horrific cost – and to no avail. The welding concerned is of excellent reputation and consequently this experience? only reinforces my conviction that there is little point in restoring an original design fault – however well done. Consequently, I have been taking note of the detail block design around the offending area and how this altered over the design life of the engine.
Obviously my checks are only over a limited number of blocks (five in all) and cannot be claimed as representing an adequate cross section to prove a definite finding – but it looks hopeful.
On the earliest blocks the solid internal wall (which is the cause of the problem) is at its deepest, running along axis AA on the drawing from BB to DD, In consequence the central waterway on the right of the drawing is a slot as the ones above and below it. There is no waterway at F, nor is there a waterway trefoil section E, This block can be identified externally as having three core plugs along the dynamo side, the one examined having a casting date of July 1932.
We have only to move forward in time to September 1932 to find a slight improvement. In this case the solid internal wall has been reduced along axis AA from CC to DD Still no additional waterways at E or F, At least the water has been brought a little bit closer to the overheated valves This block has two core plugs along the dynamo side, On moving to 1934, I had a block similar to the 1932 one mentioned above but complete with waterway F, but without the trefoil section E, Now it is just possible that I drilled F myself at some time to suit a head I was using – but I can’t remember for sure, However, tne welded block of 1933 which I am now removing from the trike was definitely drilled out by myself to accommodate waterway F and a similar one in the nose of area A Since I was taking my lead from the waterways in tne current replacement head gaskets I assumed that by so doing I was bringing the block up to the latest spec. Not so! About four years ago, I acquired a rather sad looking engine rusted up solid and with the engine number oxidised away but identifiable by the mounting brackets as an early Scout one-say 1935 only my recent block problems provided the urge to de-rust and strip its and I was surprised to find that the block itself was was in sound condition with no trace of valve seat or port cracking. Now this block is a rather more ‘cheap and cheerful’ casing – no frills i.e. the inclined face on which valves seat on the earlier blocks has been eliminated. Remember that the valve stems are a few degrees off vertical and by using a horizontal seat it means that the valves are recessed on the inside edge and are a bit proud on the other edge.
Not the best recipe for good gas flow, but you can’t have everything. What is more to the point though, is that the improved waterways are as follows – waterway F plus the full trefoil E plus a 13/64th” diameter drilling passing down the axis AA between the two valve seats thus breaching that solid wall, So now we have the reason for that trefoil shape – to provide adequate lead in for the drill which enters at CC with a centre line 1/2″ below the block face and emerges at DD 7/8” below i.e. at an angle to encourage thermo flow So, is this the answer? If I get deluged with letters from members with a garage full of cracked Scout blocks I’ll know I’ve got it wrong, again!.
Anyway, it should be noted that anyone with a power drill, a set of rotary files and a steady hand should be able to modify the older blocks (and the head to suit), I can’t see it doing any harm even if 13/64″ is not a very big hole, but I lack the nerve to go much larger.
PS, I foresee much strong language if a pilot drill gets broken off in there — so take it easy ADDENDUM In spite of what I said about not being brave enough to drill out that 13/64″ waterway – I did in fact do just that and found that the true diameter, without cutting new metal, is 1/4″ diameter along axis AA The earlier false reading resulted from build up of time, calcium, chalk of- what have you over fifty years
The Late Andrew Macqueen
Keeping Those Exhaust Valves Cool, by Bill Carney August 1992
When I rebuilt my Series 4 Scout engine in 1986, I was told that I should have a l/4″ diameter hole running at 45 degrees between the two centre exhaust valves, .and to check that it was clear. This I did, not expecting any trouble. However, after completing the restoration of the car and running in the engine, the head gasket failed. On removal of the head, I found the head gasket had blown between the centre exhaust valves. The exhaust valves themselves had blue hot-spots on them, and on removal needed the seats recutting because of the severe overheating.
I thought I would see what could be done to stop the problem reoccurring. Someone had told me that if you were careful you could open up the hole between the valves to 5/16″ diameter. I did this. but thought that this would be futile if the small amount of extra flow had nowhere to go, so I had a really good look at improving the flow into the cylinder head. In the top of the engine block there is a tee-shaped hole. but in the cylinder head there is a small round hole, so I cut out the cylinder head to the same shape as the block. This revealed a small chamber with outlets leading in three different directions. One runs in line across the head and the other two lead each way over number 2 and number 3 combustion chambers.
These two waterways were almost completely blocked with years of crud. with various drills, rotary files and needle files I managed to clean out the waterways, the centre one to 5/16″ diameter, and the two opposed waterways to about 3/8″ x 1/8″.
At that time only fibre head gaskets were available, so one of these was easily modified. I reassembled and hoped for improvement. That was nearly five years ago, and over 6000 miles. I have had no reason to remove the head since, I suppose I should go in and look, but the engine has only improved with the relatively short mileage covered, and I am content to leave well alone.
Bill Carney
Postcript—-
More years ago than I would like to admit to, Jack Rowe used to swear by the following “fix” (not based on sound engineering practice!!), he would cut a wafer of copper from an old gasket, prise open the existing copper overlap on the new gasket where the problem is (between 2 and 3) and insert this new piece, finaly closing back the copper overlap. This effectively increased the gasket thickness at the crucial point., it did seem to work, but it would only be a “get-you-home dodge I think. psb