4

I am building a small TRIAC valve for ON/OFF control of a 300W heater. The current design problem is as follows. When the heater is cold, it's resistance is low (15Ω @ 220VAC) and the initial current is large.

Is an NTC the right way to tackle this problem? If so, what guidelines should I observe in selecting one? If not, what would be a better idea?

Nick Bolton
  • 2,043
  • 8
  • 31
Vorac
  • 3,091
  • 5
  • 29
  • 52
  • Also, I am controlling the triac from a microcontroller, so some (not very) fancy digital techniques are possible/preferred. – Vorac Feb 19 '13 at 11:44

1 Answers1

4

Possibly; this page from Ametherm has a lot of description and guidance on how to select one. However, the cases they describe are mostly power supplies, where the inrush current is into a capacitor, and the NTC heats up and passes full current pretty quickly.

Given that you're building a triac control circuit, I suggest looking at "triac dimmer" circuits and using one of those instead. You can then control current either by "dead reckoning" (gradually increase power over a time period that gives the desired result) or by measuring current flow (eg Hall sensor or current transformer).

You also need to consider what happens if the heater fails short-circuit. The triac needs to survive until the fuse blows or circuit breaker trips.

It may be simplest to just overbuild the triac to handle the inrush current.

pjc50
  • 46,540
  • 4
  • 64
  • 126
  • Do you think providing zero crossing detection to the uC would be a suitable option? Would it be simpler and no more dangerous than hardware dimmer circuit? – Vorac Feb 19 '13 at 11:46
  • It's required for doing "dimming" through phase angle control: http://pcbheaven.com/wikipages/Dimmer_Theory/ ; I don't know if _just_ zero crossing detection would fix your inrush problem, I suspect not. – pjc50 Feb 19 '13 at 11:51