- Personally I most distrust the airflow measurements. Getting this wrong by 50% would explain these results. We do not actually know where the air speed was tested. Lack of data here means we cannot check this.
- I think the middle of tube air speed, air temperature, can easily be determined. Airflow is more difficult because flow velocity profile will be non-uniform, reducing to zero at edges of a pipe and higher in the middle. Mizumo talks I think of checking airflow at different points around the reactor. that does not help, and also does not answer the velocity profile question. This could make the +50% apparent results, if ignored.
- The other uncertainties seem well bounded by me and although we would need careful and detailed results I feel these are less problematic
For about $20 I can buy a mass flow sensor from any modern car with a pulse or voltage output with a total uncertainty of around 1.5% (hot wire/hot film-type). Also, RTD measures a difference in mOhms in order to resolve temperatures, normally I use NTC thermistors coupled to an Agilent 34970A and we can get total uncertainty of temperature measurements to about 13mK when using triple point and gallium cell calibrated thermistors.
Display MoreTHH how about the details of your mathematical derivation
of instability which I have asked you for?
Shall I let you off the hook?
Higher COP means more unstable
Robert, just like an amplifier circuit, high gains give tendency for more instability than low gain circuits so I agree with what THH wrote.