BLP Gas-Phase Replication Effort - Run 8….27JAN98 ………………..(best viewed with 1024 pixel display width)
Legend
Total Input Power (electrical)
Output Power (heat)
Chamber Wall Temperature (C)
Gas Temperature (C)
Gas Pressure (torr)
Filament Temperature (° C)
Room Temperature (C)
Inlet water Temperature (C)
Run 8 was another diagnostic run…this time with H2 gas, no catalyst, and an effort to limit the filament temperature to around 2000° K. The run starts with about 3 hours of the cartridge heater running at 30 watts to bake out the chamber under vacuum. To save time, the calorimeter was not allowed to equilibrate at zero input power initially. At hour 3.7, 10 watts was applied to the filament and the cartridge heater was adjusted accordingly to keep the total input power at 30 watts. Note the Tin trace which jumps up to about 1800° C and stays fairly constant for the next 45 minutes or so while the filament operated in vacuum. Then, at hour 4.5, we filled the chamber with 2 torr of H2 gas. This cooled the filament slightly but not below 2000° K (1727° C) so we just left the filament voltage at the previous value that had yielded 10 watts. As you can see from the Tin trace (which is based upon the measured resistance of the filament), erosion started promptly and proceeded for the next 2 hours until the filament finally burned out. It should be noted that the Tin trace no longer represents the temperature of the filament after significant erosion has occurred. After the filament burned out we left the system running with the cartridge heater at about 22 watts until Pout = Pin and then turned off the heater.
We got almost two hours of operation out of the filament by limiting the filament power in this run. Since the calorimeter time constant is about 1.5 hours, this is ample time to detect any significant excess heat. In view of the apparent difficulty of solving the filament erosion problems, we are considering a series of runs under these conditions with K catalysts present in the chamber.
Trivia: At about hour 8.3, there are some spiky dips in the Pin trace. These are line voltage dips caused by some relatively heavy aluminum welding going on in the shop area which is served by the same transformer that supplies the experiment area. Yes, we could be using a SOLA…but we're not.
Baselines:
We let this experiment run all night after the input power was turned off. As you can see, the Pout trace returns nicely to zero after several hours. Over the last 2.4 hours of the data collection, the average Pout value was 0.27 watts. This value can be seen to the right of the Pout: digital display in the upper portion of the screen. The n: 567 just below it means that 567 readings are represented by that average. The sampling interval is 15 seconds so 567 readings covers 2.36 hours.
It is also of interest that the Press trace was quite stable in the overnight run. This means our chamber is virtually leak-free.
BTW, we are getting a signal conditioning amplifier for the mV pressure signal which should greatly improve the signal/noise ratio.