Boiled (deep fried?) coolant pumps

[Lawrence Bardfeld <LB6116@PITCAIRN.com>]

I would appreciate some educated speculation (or better) as to what 
might account for the destruction of two pumps in two days.

As background, a few weeks ago I purchased a '93 S4 (75k).  Everything 
was stock, except for a K&N filter.  I took it immediately to a highly 
recommended (by several q-listers) mechanic who specializes in Audis, 
George Baxter.   George did lots of work getting it into top shape - 
new timing belt, tires, fluids, etc.  For purposes of troubleshooting 
my pump problems, the most relevant changes were: the engine was 
"chipped" (stage 1, ostensibly 280 hp), and Evans coolant replaced the 
former conventional water-based coolant.

The car, after the changes, always ran hot, but I was told this was 
normal, and would cease once the small amount of residual water that 
was left in the Evans (the system had been flushed) evaporated.  To 
expedite this, I was instructed to remove the cap after using the car, 
so the water, in the form of steam, could evaporate from the coolant. 
 Inevitably when I used the car, the needle on the temperature gauge 
would, within 15 minutes, pass the halfway mark, and then just prior 
to reaching the next mark, the ECS (?) check light would come on, then 
the coolant temperature warning light.

Tuesday, running a short errand in the morning, the coolant pump 
failed - more precisely, the horizontal tube coming out of the pump, 
in the direction of the rear of the car, broke, spewing coolant all 
over.  I didn't do a post mortem, but the pump was replaced. 
 Yesterday, the replacement pump failed in the same place. I did look 
more closely at it this time, and it looks pretty clearly like a heat 
problem - there wasn't any clean fracture of the plastic, rather the 
material was partially distorted as if it partially melted and failed 
as a result of intense heat.

Any suggestions?  I find it difficult to believe that the temperature 
of Evans coolant circulating normally through the system would be 
sufficient to melt the plastic, since the temperature gauge registered 
one mark away from the "hot" mark.  Is it possible that the pumps 
weren't working at all - perhaps an electrical problem - and the 
temperature of the coolant near the pump was much hotter than where 
the temperature gauge senses it (I don't know where that is).  I took 
a quick look in the Bentley I just purchased from TPC, and it didn't 
shed any light on the problem (but the fault is probably mine - a 
vehicular ignoramus with a Bentley is still an ignoramus).

Thanks for any help.

Larry Bardfeld
LB6116@pitcairn.com