Although it's basically finished, there is still some tweaking to be done. Trying to test the unit to come up with the specifications detailed in part 3 has revealed some interesting points which I am in the process of researching.

One point is transient response. While attempting to measure transient response, I noticed that if I didn't include a minimum load, I had a strange effect as the load was removed. While experimenting, I found that the only way to completely cure the effect was to replace the existing current sink with a pull-down transistor, configured almost like a Class AB audio output stage. The quiescent current through this transistor was reasonable critical - about 30mA gave the best results. To facilitate this required a major rebuild of the board, so I've more-or-less decided against it. Next time, I'll design it in. Such a device is often called a 'down-programmer', and is typically found in programmable supplies to provide the fastest response time to a command that results in a lowering of the output voltage. Otherwise, the output capacitor might hold-up the output for longer than is desirable.

Outstanding issues

Here's a list of jobs to do - mostly for my benefit:

  • Front panel labelling
  • Sort out LED displays
  • Heatsink the bridge rectifiers
  • Attempt to investigate and improve the temperature coefficient
  • Add over-voltage trip (protects in event of regulator failure)
  • Add reverse-protection diodes across the output terminals

I'm getting back to the project that diverted me back towards this one. So, don't expect all of that to get done for a while...

  • Go to page:
  • 1
  • 2
  • 3
  • 4