Power supplies dying

[kwazyjeff]

So I’ve been running my OSPi with an expansion board (16 zones total) for several months now. Unfortunately, I’ve killed 4 transformers in that period of time (currently running on my 5th). Everything works fine, then suddenly things are getting dry. I now know to go check the transformer, and sure enough its dead (output is a couple mV instead of the standard 26VAC). Any thoughts on what might be wrong and/or what I can check? Now that we’re in the hottest part of the summer, its getting where I have to be able to rely on regular watering – especially considering our restrictions around here (can only water one specific day every other week).

Thanks for any/all help.

[Ray]

Sorry to hear that. Two reasons I can think of, that may cause the transformers to be damaged, are either surge / over-voltage, or there is a persistent shorting somewhere. The latter is probably more likely, given that you have 4 transformers damaged. One thing I suggest you to check is whether there may be a shorted or semi-shorted solenoid somewhere. You can use a multimeter to measure the resistance of each sprinkler solenoid (the simplest way, without taking down the wirings, is to measure the resistance from COM to each individual station wire). Normal 24V AC solenoids should have about 30 to 50 ohm resistance. If the resistance is too low (like below 10ohm), you should replace that solenoid.

[rederikus]

Silly question. Are you running more than one zone (solenoid) at the same time?

Most transformers are designed to output about 1 Amp and that really means not more than two solenoids (main + 1 zone or, 2 zones) and the RaspberryPI-OSPi.

If you have an AC ammeter, try measuring the total current taken from the transformer at its output. That reading may give you a clue. Also try the same measurement when you activate each solenoid in turn. They should each draw about the same current when switched on one at a time.

[kwazyjeff]

Thanks for the replies guys. I’m using Richard Zimmerman’s program and it seems to be doing a fine job of avoiding multiple simultaneous programs.

I did check all solenoids and all were 25-27 Ohms or higher, so was still scratching my head. Then yesterday morning it was unresponsive again so I figured the P/S died again. However, it was just seemingly locked up – power cycle and it came back to life. Then it hit me that Wednesday is my watering day (were in a pretty severe drought here in North Texas – can only water 1 day ever other week), so I got to poking through the logs and I see that starting zone 3 is the last entry, even though there are many zones after that in the program. So I re-measured the solenoids last night and now zone 4 is showing 13 Ohms. By my math, that means its drawing somewhere in the neighborhood of 2A, which I’m sure its not happy with. It must have ended zone 3 and started zone 4, and the power drain was enough to lock up the Pi before it could update its log file.

Now I just need to get to the bottom of the low resistance. Hopefully its just a bad solenoid like Ray suggested and not a wiring short that only shows up when the soil gets wet. Still scratching my head on why the resistance varied so much from one week to the next.

[rederikus]

I am glad that you have found a fault. It is very rare indeed for a wound solenoid to lower its DC resistance. It could still be a wiring error if you somehow have two solenoids connected in parallel. Almost certainly the extra load is dropping the voltage to the point where it falls below the level where the voltage regulator feeding the Pi cannot provide a stable 5 volts DC and, it falls over.

Remember that DC resistance (R) is different from AC impedance (Z). http://hyperphysics.phy-astr.gsu.edu/hbase/electric/imped.html. That’s why I initially suggested that you measure the actual current drawn with an AC Ammeter. Nonetheless a 50% difference between one solenoid and the others would make me very suspicious indeed. I would completely remove the suspect solenoid and check it out of circuit. If it still exhibits the same problem then, it’s off to Home depot or Lowes.

[Author]

Posts