LV Dimmer design, comments...

Hi,

This is a design for a 12v serially controlled dimmer. It will work on the basic principle of PWM and use N-Channel MOSFETs to PWM the outputs (halogen bulbs).

As you can see it works around a PICAXE18X processor (U1), which will communicate serially with the data connector (X1). The 4 pins are +5v, GND, TX and RX. This board sends the 5v via the data connector to a remote chip which then turns it into standard PC serial.

6 outputs are shown, the resistor by each mosfet is there to protect the IC. The data plug is optically isolated.

Power either goes into the unit via X3 or X2. X2 is an onboard 2.1mm power jack which would be used for testing and programming the unit - it bypasses X3 if it is connected. X3 is a DIN plug which will connect up +12v and GND. The centre pin is for a 3mm LED built into the DIN housing, or if I connect one or more of these modules together I have designed a power board which uses it.

Each channel has a status LED. There is also a 3pin programming plug for the PICAXE (SL7). 7805 regulator provides 5v power to the board. I have just remembered to add 22nF caps either side of the regulator

Any comments on this is appriciated. The MOSFETs I am looking at are STP22NF03L.

Thanks,
comp

computer said:

Any comments on this is appriciated.

Click to expand...

My only comment is - does it work?.

The PICAXE runs a BASIC interpreter, with the BASIC program code stored in the few bytes of internal EEPROM (presumably stored tokenised?) - as such it's only got a very limited amount of program space, and runs very slowly (compared with assembler, or compiled languages).

So have you written code that works?, and is it fast enough?.

Heh. I'm not quite sure exactly, I've only built a two channel version with transistors for another LED project. It worked quite well, it could handle PWMing them out ok. I need to work on it a little more, though. I'll post final designs if it works. Does my hardware setup look correct though?