The DeltaE FAQ (Frequently Asked Question) List.

1) Q: I'm not quite sure what STKCIRC does. It seems to me that total area, open fraction, r0, and length completely define the geometry. The fact that you also require "half of sheet thickness" disturbs me. What do you do with it?

A: Yes, both STKCIRC and STKSLAB have a parameter that is extra. The half-thickness off the sheet says something about the local solid heat capacity (affecting epsilon_s), but we often needed another parameter to be able to compensate for oclusions in the stacks (the fish-line in a spiral stack, for example), so that comes directly from open fraction. In the honeycomb (STKCIRC) it doesn't apply so strongly, but it's there.

2) Q: I cannot tell from the manual how HXFRSt and HXMIDl differ.

A: Perceptive question...they're exactly the same. There's only a name change for readability.

3) Q: I often am slightly irritated at the different number of decimal places for hdot and wdot in .dat files.

A: Hmm...me too, now that I think of it. There was some crazy reason for this that came up in discussion, but I can't remember it. We'll probably change it, though.

4) Q: It would be nice to have a "hot key" which would return the user to the main menu from anyplace in the program. For example, if I type "p," then change my mind, I would like to be able to return immediately to the main menu.

A: The current version lets you cancel after you type p, and in general, whenever a managed input is expected (when there is a default), you can type "x" to get out. CTRL-C also works on DOS (but will kill the program in UNIX during input). If you are running an iteration, you can interrupt it and abort with CTRL-C. This works on DOS and SGI systems. For the Mac, Command-period will always interupt what's going on. Sometimes, these can leave you in a strange state, though.

5) Q: It would be great if you could trap FORTRAN errors so that the program will not precipitously kick the user back to DOS.

A: I have done everything I can so far to do so, and it just doesn't work very well (yes, it used to be worse). Perhaps someday we'll find a compiler that behaves the way it claims it does. But there currently is a DOS trap routine in the code, and the coprocessor interrupts are also supposedly corrected, and yet, you get this frustrating behaviour. We sympathize, because we constantly put up with it, too. The Sun and SGI have better math implementation and are more forgiving. The Mac, too, will go much longer before crapping out, but then when it does, it usually takes the machine with it! There, too, I hope a compiler upgrade will help. We are working on a more universal solution, perhaps for version 1.3.