The MGA With An Attitude
Original Radiator Top Tank Baffle - CO-100B

At 09:51 PM 3/29/06, b57mormaz (Martin yd3-2226) wrote:
"To build twin cam rad I took apart several MGA rads. They all had flat plate, no perforation, in the upper tank. The plate does not go all the way across, 2-IN gap from end, and was about 1/4-IN above tubes. Just from simple curiosity, what is it there for? As the twin cam upper rad tank is way shallower (lower MGA tank) should I install it? Any benefit?

It would help to mention which end of the plate has the 2" gap. As I recall the plate in the top tank is at least 1/4 inch above the core, and maybe a little more. The cross sectional flow area of the space below the plate is approximately equal to the cross sectional area of the thermostat outlet, so there would be very little restriction to flow across the top of the core. The plate runs all the way front to back and left to right except that it stops about 2-inches short of the end where the inlet pipe is located (right side of car for push rod engine). When you pour water into the filler neck (left side of car). the water has to flow across the top tank to the opposite end of the radiator before it can drop into the core. For normal circulation with engine running the coolant can flow from the thermostat outlet through the top hose into the top tank and may drop directly down into the core area.

I wasn't there when the engineering was going on, but I can take an educated guess. I believe the plate is there to even out flow through the whole core. The water pump creates suction (partial vacuum) at the lower left corner of the core. There is a high flow rate at that point and progressively less flow across the bottom tank. Motion of the flow from right to left through the bottom tank is driven by a pressure differential with relatively lower pressure (higher vacuum) near the outlet pipe. When there is a constant pool of coolant in the top tank covering all of the down tubes, the higher bottom vacuum near the outlet pipe would draw relatively more fluid down the tubes on the left side.

I believe the baffle compensates for this inequity, directing more flow to go through the right side of the core by slightly restricting flow across the top tank just above the core. The baffle in effect creates a slight pressure differential across the width of the top of the core similar to the pressure differential across the bottom. Where the bottom suction pipe tends to pull more fluid through the left side tubes, the top inlet pipe tends to push more fluid through the right side tubes.

For the Twin Cam radiator where the top inlet pipe is on the same side of the core as the bottom suction pipe, I would think it is even more important to direct the inlet flow across the top tank to the opposite side before dropping into the core. If the flow was allowed to enter and exit on the same side of the radiator there could be a disproportionate flow of coolant through the left side tubes and relatively insufficient flow through the right side tubes.

With high coolant flow at high engine speed this may not be much of an issue, as there will be relatively high pressure at the top and relatively high suction at the bottom. Considering the flow restriction of the individual vertical tubes it is likely they would all carry similar flow with the high top to bottom pressure difference.

With low coolant flow at low engine speed this could be more of a problem. At low flow rate the restriction in the vertical tubes is nearly nil, so it would be easy for the bulk of the flow to drop directly down the left side of the core with much less flow further to the right side. Such a differential in the flow within parts of the core can only detract from the efficiency of heat transfer. Considering that most of the cooling problems seem to happen at low speed, the radiator core can certainly use all the help possible to retain efficiency of flow and heat transfer with low flow rates.

There may be another good purpose for this baffle plate. If the fluid level was substantially lower than "normal", such that part of the core away from the inlet was starved for flow with the top of the core exposed to air, then suction from the pump might well pull some air through that part of the core. Running partly air through the radiator and engine block and head in place of some of the liquid has to be a horrible detriment to the heat transfer function all around. Apparently this design will allow use of a much smaller top tank (for a lower bonnet line or taller core).

With the presence of this plate, and the only fluid entrance being at the inlet port, this inlet area (that 2" gap at the end of the plate) should be always submerged in fluid. This would prevent entry of any air into the space below the plate such that full atmospheric pressure (14.7 PSI) would be brought to bear to push the coolant into the cavity below the plate, and there would never be any air in the space immediately at the top of the core. The only time this would fail would be if the coolant level was so low that there was no fluid at all above the plate. For that to happen you either have a leak, or it was boiling to blow out steam.

Since I do not have any laboratory type test equipment for such things, I cannot quantify any of this. But someone "back in the day" must have thought the baffle was a good idea. Some thought about flow processes tends to make me agree, even if I can't prove it one way or the other. Removing the baffle without knowing intimately how it works ranks right up there with changing bore sizes for water passages in the engine without knowing the effect on local flow or temperature changes.

Thank you for your comments -- Send e-mail to <Barney Gaylord>
© 2008 Barney Gaylord -- Copyright and reprint information