Kimmel Collection: Williams Victress Boiler, Car and Engine

Williams Victress Boiler and Engine

Boiler

Williams Victress Steam Generator. This was located in the trunk of the Victress bodied Williams car that had the 105 c.i.d. in-line four engine under the hood. The engine was advertised as producing 250 hp. The burner was reported to burn about 10 gallons per hour of kerosene. The burner and controls were missing from the boiler when it came to my shop. The boiler is a standard Doble “F” style boiler, with a water wall of helical coils and 13 pancake coils. The pancake coils are 5/8” od and the 2 superheater coils and the waterwall coils are ¾” od. Both made from 306 stainless steel. There were many pinhole leaks in the tubing probably caused by chlorine corrosion of the 306 from using city water. Also the inside
		The Williams Victress
		of the tubes were coated with carbon from the lubricating oil getting carbonized to the tubes. There was no oil-water separator. There were two commercially available Burling Instruments, Chatham, New Jersey quartz rod microswitch temperature controllers. The Burling company furnished me with a copy of the original invoice to the Williams Engine Company, Inc. dated 4/4/67 Model VD-2S that used a 13” quartz rod, normal open switch set at 800 F and normal closed switch set at 900 F.

Without having any other information at this time, it appears that by using two temperature sensors the Williams brothers were able to solve the basic control problems of a monotube boiler that had vexed Abner Doble all of his life. Doble ended up using a normalizer, which is actually a de-superheater, to give feedback information to the water pump controller so that the boiler was not flooded.

The Williams appear to have used one sensor, placed just before the water wall section to turn the water pump on and off. They then placed the other one after the water wall and used it to control the de-superheater water feed. This would prevent flooding the boiler and only use the de-superheater as needed.

The steam generator is of conventional design otherwise. There is a tangentially fired stainless steel combustion chamber inside the helical coiled water wall that then exhausts into another chamber where the water wall is exposed and then down through the two superheater coils and the other pancake coils. There is good workmanship winding the coils, but other than that nothing out of the ordinary.

Williams 105 c.i.d. Victress Engine

Williams 105 c.i.d. Victress car steam engine. This engine is on loan from the sons of Calvin Williams. The engine and car were made in about 1960 and driven extensively and shown to Congress during the clean air car hearings. The engine is of the final Williams’ design; four cylinder in-line with poppet valves, sliding cam with 4 forward and 1 reverse cam grinds. These were all high recompression uniflow engines with a patented pressure relief valveinto the steam chest. Extensive tests show an 8 pound water rate. When the engine is in the lower cutoff mode without the auxiliary exhaust valves open it was titled the “Williams Cycle”.

The final design was single acting cross head uniflow. This is seldom done by other steam engine designers. The cross head feature made the engine taller and more complex, however it had the advantage of keeping water out of the crankcase.

The basic Williams’ engine designs used high recompression uniflow engines,meaning that one third of the exhaust steam remained in the cylinder after the rest of it had exhausted through the exhaust ports. The clearance volume was designed so that thissteam was recompressed to boiler pressure. The original patent claimed that the recompression to boiler pressure raised the temperature several hundred degrees higher than incoming steam and thus there was an added efficiency. People have been arguing about this ever since. It appears to follow Professor Stumpf’s design work.

The Williams’ construction methods are well thought out and worth copying. The problem with a poppet valve steam engine is that the valve faces are on different planes about ¾” apart. This is because the intake valve is above the port going into the cylinder and the auxiliary exhaust valve is below the port. The problem was solved by bolting a short cylindrical piece onto the head. This plug fits down into the cylinder making the effective head an inch or so below the top of the block casting. This way both the head casting and the block casting can be made with flat mating surfaces. The plug has two steam passages milled into it with a ball mill, thus leaving a very small clearance volume. The pressure relief valve, which was the big secret ingredient of theWilliams engine and one they did not disclose is merely a small spring loaded poppet valve that opens when recompression pressure is higher than steam chest pressure. Richard French told me that when the car was accelerated from a stop sign, it started in full cutoff and counter-flow with the auxiliary exhaust valve cam engaged. When the car got across the intersection it was shifted into straight uniflow by disengaging the exhaust valve and engaging a shorter cutoff intake cam. At that time it would produce much more power and spin the wheels. He said that when the throttle was opened at 60 mph the wheels would spin. The car has over 5,000 miles on it.