Some four years ago, GAU 15, my 1932 20/25 Hooper-bodied special touring saloon, had a total engine rebuild in the hands of an ex-Coldwell Engineering mechanic with whom I am no longer associated. He reinstalled the radiator minus its spring-loaded, connecting pin between the Calorstat and the radiator shutter levers. In its place was, and still is, a somewhat unsatisfactory, large but loose split pin. Locating a replacement proper part has not been fruitful. In recent weeks I have had a seemingly intermittent problem with boiling. All sorts of explanations passed through my mind at various times in the very early hours of the morning, but my first port of call was to the water pump. On one of John Eastwood’s visits, we removed the pump and checked it out. It was in perfect condition. We moved on to the radiator shutter lever mechanism and the dreaded split pin. John was horrified to see such a ‘bodge’ having been placed there by a so-called expert. Its looseness meant that the calorstat thrusting rod had a very tardy contact with the first of the levers and in turn the shutters were opening insufficiently and too late. Adjustments were made, the shutters were now more open than fully closed when the engine was cold and the lever mechanism and all joints on the shutter blades were thoroughly lubricated. The boiling ceased for a week or two.

Depression Sets In.

The problem returned and my observations led me to the conclusion, again, that water was not flowing, but a telephone call to John sent me into depression when he aired the various possibilities:

1) The pump was in perfect condition and if it has now failed there is now a broken shaft or shattered bits of it somewhere in the engine. This could cause the engine to seize. Probably an unlikely cause of the problem.

2) Blockage in the cylinder head. A new aluminium head was fitted at the time of the engine rebuild and with its wider water passageways, again, probably unlikely.

3) A blown cylinder head gasket causing hot gases to influence the coolant temperature. A cylinder pressure test might be in order and any severe reduction in one or two of the cylinders could be a tell-tale sign and certainly a head off job would therefore be a matter of urgency.

4) The block is clogged with metal and other debris and the pump is unable to push water through the very restricted pathways. “Was the block cleaned up at the time of the engine rebuild ?”, John asked.
“I assume so, but I cannot be certain”, said I.
 “The engine will have to come out if it is fouled up”, was John’s response.
 My heart was beginning to sink fast.

5) “The radiator core is blocked with metal debris and sediment. When I had mine replaced on my Thrupp & Maberley 20, it cost me £3,000”, John informed me. The depression intensified.

Tracing the blockage.

I proceeded to the garage and conducted a pressure test on all cylinders. All of the readings were high and more or less identical. This was something of a relief but only partially comforting. John came over the following morning, Monday, Sept 29th, 2008. Between us we confirmed that boiling was commencing after just a few minutes from cold and that no water was cascading from the radiator feed pipe into the header tank. We discharged about 2 gallons of coolant from the system. This was a relatively slow affair. Next, rather than taking the entire water pump off again, we removed just the brass elbow linking it to the block and started the engine. The pump was working but not pushing much water at all, but water was flowing freely by gravity from the block. A little finger and then a thin screwdriver through the entrance hole into the block provided no evidence whatsoever of any blockages in there.

Elbow: Water Pump to Block

With John  closing off the exposed orifice of the pump outlet by means of a thumb, I filled the radiator again to see the effect of water passing in the reverse direction from header tank to cylinder head and on down into the block. Clear water flowed freely. This was not to suggest anything other than the fact that at least the front end of the block was unimpeded. We drew the tentative conclusion that the block could be ruled out.
Our attention was then focussed on the bottom water pipe between the radiator and the water pump. On removal, it was clear that the problem did not emanate from here, although it could have been within the flanged, cast aluminium outflow pipe on the driver’s side bottom corner of the radiator. Filling the radiator with water once more demonstrated that the coolant flowed freely back down through the block but not down through the radiator. John reminded me of the cost of a new core! With some difficulty, we removed the radiator outflow pipe and immediately eliminated this part of the system.

Radiator Outflow Elbow, Bottom Hose and Pipe to Pump. New Cork Gasket fitted at flange

Core Blimey !

With just the radiator left, the image of a £3,000 replacement core began to erupt in my cerebral hemispheres with consequent effects on my cardio-respiratory system. We deliberated on the alternatives remaining for us; to take off the radiator and back flush in the inverted position or to back flush in situ. We opted for the latter and with the outflow pipe back in place, complete with a newly fashioned cork gasket and sealant, the water pump bottom feed pipe was connected to it. Amazingly, the yellow hosepipe nozzle fitted snugly into the open end of the pipe and our final investigation was about to begin. Using first aid knowledge on stemming a haemorrhage, I bandaged the hosepipe nozzle into the pipe and secured the bandage with masking tape to prevent  premature explosive separation, turned on the water supply at full blast and stood back in anticipation and with a throbbing intrepidation. Within 30 seconds, water was gushing from the block and soon from the radiator overflow pipe and the header tank filler. Removing the hosepipe resulted in a strong rush of water from down below. Whatever the plug was, it had now been dislodged but not necessarily disintegrated.

Inflow Pipe and Hose: Radiator to Water Pump

The entire cooling system had now been effectively checked out and fairly quickly all parts were fitted back, the system filled with fresh water and the engine fired up. The water was flowing through the complete circuit once more and no boiling was evident after 20 minutes of engine activity. The cooling gradient down through the radiator core was now as it should be. A further rapid release of water from the radiator outlet pipe in an attempt to flush out loosened debris followed by another refill made us a happy pair.

Bad Advice from a So-Called Expert ?

John and I sat back for a while and wondered about the nature of the embolism. He was aware of the fact that the ex-Coldwell Engineering guy who rebuilt the engine at his former premises in Chesterfield had given me a large tin of his special, very dark brown, almost solid, pump bearing/gland grease to fill the rotatable greasing cap on the pump and that the advice given was to give the pump bearings a regular dose of the stuff. John’s feeling was that far too much of this stuff had made its way into the cooling system having melted easily by the normal hot water temperatures over a period of four years. We reckoned there was a strong possibility that on entering the radiator as a hot, fluid  oil it had changed back into the solid or semi-solid state by the time it reached the bottom as a result of the radiator doing its job and there it accumulated, uniting with fine debris and sediment to form a dense mass of material, some or all of which had migrated across to plug the outflow pipe.

Flushed with Success.

There was niggling thought; the mass of muck and grease still lurking perhaps at the bottom of the radiator. John recalled that a few months ago I had, according to the temperature gauge, a severe overheating problem in GDP 38, my early 20/25 open tourer. To cut a long story short, the problem here centred around cylinder number 6 where of course the temperature sensor is located. There was not much, if any, water flowing around the sensor. It was very dirty and muddy. Poking around in the sensor’s orifice, bottle-brushing and injecting with water appeared to partly solve the problem. Certainly the temperature gauge was offering me more pleasurable feedback. I had spoken to Dr Ken Brittan earlier about my problem and the possibility of blockages in the waterways at the back end of an old cast iron head and how he would tackle the situation. After I had bought some Halfords Radflush, he called me a couple of days later offering to send me some special pink powdery crystals he had purchased from a company in California. The chemicals therein are collectedly designed to degrease, de-scale and loosen up undesirable rust sediment and sludge in a cooling system. In the meantime, I introduced the Radflush and drove the car for several days before draining off. It actually worked and there has been no further apoplexy caused by the water channels around No. 6 cylinder.

Since I still had Ken’s generous present, I decided to use it in one short flush and one very long flush in car GAU 15, the first one over a period of 30-40 minutes whilst John and I had a sandwich and drink, the  second one over repeated trips out over several days. I am hoping that the chemicals have broken up and digested the offending conglomerate of hard grease and fine debris. What I do know is that the problem has not raised its ugly head for several weeks and hopefully for good.

Another saving of thousands of pounds sterling!!

The 1 pound (4.45 Kg) container of stuff Ken Brittan sent is called ‘Thoro-Flush’, by Irontite, located in El Monte, California.

What is a Cooling Gradient?

If the radiator and its fan are doing their respective jobs properly and the coolant is flowing around the system as it should, the water will cool rapidly as it trickles or is drawn down through the radiator’s matrix. This is so distinct that by putting a finger behind the radiator shutters and against the core itself at various positions down it and on both left and right sides, a reduction in temperature should be perceptible. To be more scientific however, and to be in a position to record the information, a thermometer should be used, but one whose bulb is capable of entering the air tunnels in the core. If you have a cooking probe thermometer used in the kitchen, this would perform equally well.

The operation is best done from the rear of the radiator for ease of access. One important word of warning should be heeded, I have to say. With the engine running, it is not easy to see the fan blades rotating or even the radius described by them. Fingers and certainly a glass thermometer are easily damaged by a close encounter with the blades in this respect and therefore the measurements are best done with the engine stopped for about half a minute whilst two or three of the temperature estimates are made. Three or four measurements down each of the two sides of the radiator and a sighting of what the temperature gauge in the cab is reporting for each series will provide you with a nice impression of what is going on. Repeat the measurements at various stages in the general heating up of the engine. This is all good stuff for producing a profile of data in graphical form if you are that way inclined and it will certainly tell you whether or not you have a radiator partly clogged with years of silt, limescaling etc.

Bear in mind that since the bottom outlet from the radiator is on the water pump side of the chassis rather than being in the middle, the rapid flow of water down through the core will be in direct lines from the header tank to the outlet pipe on account of the suction force created by the pump, leaving a region in the opposite bottom corner where the flow rate is significantly less. You will notice the difference in temperature between the two bottom corners. For the same reason, sediment will tend to gather in the ‘cul-de-sac’ corner.

In a similar vein, the water pump itself is so located on the engine as to induce a region of rapid flow, directly upwards and through the cylinder head towards the radiator feed pipe, and a region of slower flow, the back end of the engine. This represents a poor design fault in my opinion and it is for this reason that the No. 6 cylinder tends to have a much higher temperature than those further forward; the water flows less rapidly around it and since the thermosensor is located in the region of No.6, the temperature gauge reports the local state there rather than a more general picture.

 With radiator cap off and engine running, measure the temperature of the water in the header tank and compare this with that which the gauge indicates. They are very different indeed. The header tank water temperature represents the average of that coming from all six cylinders.