Burning up 24VAC Irrigation power supplies

[davidjustin]

I build the diy relay board circuit and soldered up a compact little board, seems to work great. 

I am powering the Raspberry Pi with a regular orbit brand 24VAC power supply and bought a little power buck converter from amazon to help power the relays separately from the Pi. I split the 24VAC from the power supply to the relays and the little buck converter board I linked to. The 24VAC runs directly to the relays for the valves, and the other side is converted to 5VDC for the relays.

Somehow overnight my power supply burns out.

I am not sure if my soldering caused the problem. I noticed once in a while ALL the relays are activated, maybe this condition burns out the power supply if left unchecked for too long? I can't find a reliable way to force the condition, but it never happens when the 24VAC is not connected...

I think my next test is with three different power supplies. Once for the pi, one for the relays, and one for the valves.

Any help or ideas are appreciated!

Thanks,

David

[dan]

The 3 power supply configuration would be better.

Did you include the 10K ohm pull up resistor on the shift register circuit? That was added recently to the circuit diagram and it helps prevent all the relays from switching on during power up.

Also see this post:
http://nosack.com/sipforum/showthread.php?tid=70&page=3

Look for number 25 for info about connecting the external power to the relay board.

[davidjustin]

The 3 power supply configuration would be better.

Did you include the 10K ohm pull up resistor on the shift register circuit? That was added recently to the circuit diagram and it helps prevent all the relays from switching on during power up.

I did add the 10k ohm resistor, and that helped the power to all relays on start up issue.

I am assuming the ALL relays on the issue is what is burning up the transformers in the power supply since they are all drawing too much power at the solenoids. I'm also thinking about adding a fuse to prevent further burned power supplies!

I am going to re-do my little circuit with simpler wiring to see if that makes any difference...

Thanks for the help!

[dan]

The fuse sounds like a really good idea.

[davidjustin]

So I soldered up a new power distribution board with a fuse block and soldered a new shift register circuit. When testing on the benchtop everything works perfectly. I have three power supplies, two 5vdc to the PI and the relay boards. Then I have the 24vac supply and run through the normally open side of the 16 relays.

For some reason when a program ends, or when I hit the stop all stations button, 4-5 relays will still randomly come on. This only happens when the 24vdc circuit is present.

The bummer is that I added a .75 amp fuse, I burned up a couple during testing. I matched the size from a normal sprinkler controller at 0.75 amps. Apparently, the fuse block I bought was cheap enough it melted and still blew the transformer... *facepalm*.

I need to do some more benchtop testing to see why the 24v circuit causes the whole thing to freak out sometimes... If you have any ideas I am all ears!

Thanks!

[dan]

I'm really sorry to hear about the problems you are having.

What type of relay board are you using, low-level trigger with opto isolaters? And the connection from the Pi > shift registers > relay board is just control wires and +5V to vcc on  the relay board with the ground not connected?

The 24V through the relays should be completely isolated from the 5V circuitry.

Also, if the relays are acrive low, there needs to be a voltage to the relays to pull them high to keep them off. If there is not enough power to the pi/shift registers that could cause the relays to switch on. I prefer to use active high (high level trigger) relays which only switch on when a voltage is applied.

[davidjustin]

Thank you so much for the reply! Don't feel bad about the troubles, this is all a learning experience for me, so I take it all in stride. I am very pleased you are willing to help!

Here are the relays I am using: According to this response they are opto isolated (https://www.amazon.com/forum/-/Tx3HQROER...B00KTELP3I)

I have three power supplies:

1. Powering PI, then the six wires from the PI runs to the small shift register circuit. There are two sets of nine wires coming from two shift registers. The 8 control wires and the +5v wire (no ground). The 9 wires are plugged into the grouped header pins on each relay.
   
2. Powering relays. I am using an old Apple USB charger and cable. I have the red +5v wure from the usb cable going to my power distribution board. This is split and running to the small headers where the jumper was removed from the relay boards. +5v is going to the JV-VCC and ground is running to ground on each relay.
   
3. 24vac is on a separate circuit on the power dist board. Before it splits between relays, this is where I placed the fuse block. One side acting as common and is running to the sprinkler valves. The other side is split going to the relays. Each relay has is chained together on the normally open side of the relay, and the middle (common) terminal is where the sprinkler vale wire is connected.  As far as I can tell the 24v is completely isolated. It's possible there is a short in the wiring going to the sprinkler valves, but I am not sure why this would effect the operation of the relays. 
   

I wonder if the USB charger is not powerful enough for the relays? 

Thanks again for the help!!!

[dan]

Sounds like you have everything connected properly.

Since the relays are turning on it seems that they are getting enough power. It is more likely that the Pi may not be getting enough power to keep the voltage high and keep the relays off. It may be an intermittent voltage drop from the Pi if it is getting marginal power.

Is there any possibility that your power distribution board is providing a path between the power sources such as a common ground?