16-32 bit D register conversion in FX1N

[CJH 0]

I have not worked with Mitsubishis before, but the FX1N seems to be an ideal solution for a motion control project I will be embarking on shortly. The biggest selling point for me is the apparent 32-bit integer maths that it is capable of. My concern, however, is that nowhere in the manual can I find how you take a 16 bit D register and make it usable in a 32 bit operation (eg DDIV) (and vice versa) Can anyone clear this up, or am I totally missing something? Thanks

[Sergei Troizky 26]

For 32-bit instructions, the least significant word addresses are directly specified as the instruction operands. The most significant word is the register with the next following address. Example: DADD D0 D2 D4 The 32-bit registers in this 32-bit addition are D0D1, D2D3, D4D5. P.S. It is a good practice always to use even register numbers. This simplifies instructions converting to 32 bits, should it become necessary.

[CJH 0]

Sergei Thanks for the reply, but what I really want to know is, say for example, D0 is a 16 bit value. Do I need to explicitly convert D0 to a 32-bit register in another location (eg D8D9) before I can use it in a double operation? I'm concerned about the location of the sign bit (MSB) in these conversions.

[Sergei Troizky 26]

In correctly written program the 32-bit value should be such from the first appearance. In your case you can perform MUL D0 K1 D0. Edited 22 Feb 2005 by Sergei Troizky