1. Calibrations
2. Calibration File Types Explanation

Getting started doing your own cals but confused about the different files and what they're for? Well here's a quick rundown of the different types of files you'll see and what they are and what they do.
The files:

The different types of files you'll see are xxx.bin, xxx.s19, xxx.tbl, and xxx.calx. What are all these files? Which one do I burn to the chip?
First, the .bin is the actual file that gets loaded into the memory chip. This file is the code and calibration data that the CPU uses to run the SMEC.
The .s19 file is basically a copy of the original binary file, but in a special format used by Motorola in it's bootloader. Usually, you do not need to use this file. D-Cal creates it by default when you save your files.
The .tbl (table) file is a database that contains the information used by D-Cal to allow the user to edit the binary file. This is data types, addresses, and scaling factors. Think of this like a file-allocation-table (FAT) for a hard drive.
A .calx file is basically the same as a table file. Geoff just changed the format so he could add more information to it (such as groups, injector scaling information, etc.).

How are these different files used?
D-Cal and CHeM are binary editors. That is, they take the binary file, interpret the data in it and present it to the user in a readable format so that it can be edited. But, D-Cal can't do this without knowing where the data is stored in the .bin file. That's where the table (xxx.tbl) file comes in. This file is essentially a database of the tables and constants stored on the .bin and used by the CPU. D-Cal reads the .tbl file to find out where the data is in the .bin file.
In addition, the scaling factors for each table/variable/constant are built into the data and/or code. The binary values that are stored have no sclale associated with it, they are simply a value from 0 to 255. So, the .tbl file also tells D-Cal what scale to apply to each table so that the data makes sense to the user. Without the table file, the editor doesn't know what scale to apply to the data. Changing these scaling factors only changes them in the .tbl file. It has no effect on the .bin at all - and therefore no effect on the cal.

Where do these files come from?
The binary file is the actual file used by the LM/SMEC/SBEC. It is the code and data used by the CPU to run the engine. It is created by Chrysler (and as such is considered copyrighted material). It is downloaded from the stock chip using an EPROM burner/reader. Or, with a properly modified harness, D-Cal can read the file through the SCI port (the DRB connector under the hood).
The S19 file is a version of the binary file that is created by D-Cal. It is only used by some Motorola-style bootloaders. Most will never use this file.
The table file data (addresses, data type, scaling factors, etc.) has to be created by the user. D-Cal has a utility to detect most table types and find thier addresses. But, it can't know what scales to apply to the tables or constants. Your best bet is to find a compatible table file or use a well-known cal that has a fairly complete table file. Since the table file is user-created, the absence or addition of a table or constant (when compared to another cal) will usually have nothing to do with the .bin file itself - it will likely be a difference in the table file itself.
The calx file is essentialy the same as the table file. It simply contains more information and it's in a different format specific to CHeM2 (zChem). In addition to the addresses and scaling factors, it groups the data into categories to make finding and tuning certain aspects of the cal easier. It also contains data about which tables need to be scaled for injectors. Again, the file is user-created. So, finding a calx file that is complete for the cal you are working with is very important.