I want to replace the controller of a sprinkler control box with my own design, while keeping the backplane and module(s). The controller has a 14 pin ribbon cable connection to the backplane and 4 of those pins are feeding power to the controller from the backplane. The remaining 10 pins aren't powered without the controller connected and show 24VAC when the controller is connected. Using an oscilloscope, I've monitored the pins when changing the sprinkler zones, but haven't been able to figure out the control signals (the pins are always 24vac when I measure them regardless of what zone is on). Since it is AC, I really don't know where to go from here (if it was DC I might have a better shot). I'm assuming the signal changes too fast for me to see on the oscilloscope. Is there something I can use to 'watch' the connections? (Maybe cut the cable and add a circuit/monitor??) Is there a better method?

I've also searched for schematics of the controller, but haven't come up with anything useful.

Controller: The back of the board has traces and a reset switch, but no other components. On the ribbon cable: Pin 1 is one phase of the 24VAC and pins 2,3,4 are the other phase. There's a 9V backup battery that isn't present and isn't required for operation. The 6 pin connector goes to a rotary switch that makes controller selections. Controller

Backplane: A 24VAC wall transformer feeds the backplane with the two orange and single green wires. The yellow wire is a jumper for an uninstalled rain sensor (rain sensors are a normally closed switch). I've (poorly) soldered the module to the backplane so it stays connected while it is out of the cabinet. backplane

Module: The module has 7 outputs. There is a single solenoid valve wired to the module (the black wire with white stripes going to the screw terminals). One of the outputs of the module is always on and the other 5 are switched. The seventh connection is for the unswitched phase of the AC. Three more modules can be added to the backplane. The controller 'knows' when a module is not installed, but doesn't sense whether or not a solenoid is installed. The smd (transistor maybe??) is labeled 5D with a sideways 'p'. The triad is labeled PJ 600E BT134 A1449 D6. module


simulate this circuit – Schematic created using CircuitLab

Update: I tried to make my first schematic with the module soldered to the backplane and the solenoid still attached. I've updated the schematic with my second attempt (I probably have the triad oriented incorrectly). I'm using continuity on my multimeter to figure out connections. Now that I might have the module schematic close to correct, what makes the outputs turn on (trying to avoid the randomly applying power and burn it up method)?

Just to clarify what I mean by common output, is it is the second leg of the AC. All other 6 outputs are the opposite leg of the AC. I'm not sure if I am using common output correctly.

  • 1
    \$\begingroup\$ irrigation sprinkler, so other than some grass no lives should be at risk... \$\endgroup\$ – shaun5 Aug 30 '17 at 19:33
  • 1
    \$\begingroup\$ @jonk I've added pictures and some additional explanation. \$\endgroup\$ – shaun5 Aug 31 '17 at 0:08
  • 1
    \$\begingroup\$ How do you set a bounty of 100 when your rep is 3? Can you clarify in your question (don't spread info that should be in the question through the comments) (1) How much of the system are you replacing? (2) Why can you not build your own controller and use relays to switch the 24 V AC solenoids? (3) If you want to use the 7-terminal interface boards then you need to reverse engineer them and provide a schematic of one channel. Start by identifying the 3-legged fellows. They are probably triacs. There's a schematic button on the editor toolbar. \$\endgroup\$ – Transistor Sep 1 '17 at 22:59
  • 1
    \$\begingroup\$ @Transistor maybe his rep was 103... \$\endgroup\$ – Harper Sep 2 '17 at 1:15
  • 1
    \$\begingroup\$ @MatsK I've added the labels I could read on the module. Do you think reverse engineering the module is the best first step in reverse engineering the control signals? \$\endgroup\$ – shaun5 Sep 5 '17 at 18:18

I think it will be a lost cause trying to retain use of the modules. Get some DIN rail terminal blocks and don't look back. Edit: I know you really want to, but it just seems like a slow expensive process. You're sure to let some magic smoke out as you design/test, and what? You call Rain Bird and ask for part 635575 and pay full-boat parts price with shipping? That could be $65.

The fun and craft, if any, would be in the reverse engineering. You say you see 24V on all the lines. Can you measure the current on an active being-operated solenoid, and then figure the resistor value to make a dummy solenoid? (or just another solenoid lol). Seems like you'd need 2 or more solenoids to even troubleshoot this, just so you can be sure you are commanding them to be on or off. If the controller isn't doing what you think, you'll never figure it out.

I'd also watch for phantom voltage, it may show 24V on all the lines, but will a 22K resistor across the test points (1.1ma@24V) make that voltage go away?

I'm deleting the part about how to do it because you've already done solenoid control projects.

  • \$\begingroup\$ The goal of this project isn't to make something that replaces the entire controller box (I've made one using a Raspberry Pi and relays). Retaining the modules with my own backplane is probably easier, but I really want to only replace the controller. \$\endgroup\$ – shaun5 Sep 3 '17 at 0:48
  • \$\begingroup\$ Yeah, and I get you want to, just seems like a mistake. You're bound to make some magic smoke, are you willing to pay Rain Bird top dollar for a proprietary replacement board? ... ... well ... OK then, it looks like each pin is simply a driver signal, but you seem to have trouble detecting the AC vs DC components. Perhaps your next question is "how do I analyze a control on/off signal which has both AC and DC components?" \$\endgroup\$ – Harper Sep 3 '17 at 1:51
  • \$\begingroup\$ Would the 22k resistors go in series with each of the 10 connections, or just in parallel to the alternate AC leg when measuring the voltage (parallel makes a lot more sense to me, but obviously I'm no EE)? \$\endgroup\$ – shaun5 Sep 7 '17 at 1:49
  • \$\begingroup\$ @shaun5 between the terminals of the DVM, point being to discharge any phantom voltage during testing. So whichever two points you happen to be testing for AC or DC voltage -there. \$\endgroup\$ – Harper Sep 7 '17 at 4:18

Reverse Engineering 101

  • Look at return wire for ground switched solenoids from relays
  • measure DCR of solenoids for surge current rating (V/DCR)
  • use a dummy load (Triac outputs will not conduct without a return load resistance or sufficient holding current flow.
  • map out continuity with ohmmeter to make a wiring table.
    • label everything instead of a schematic e.g. P1-1 to J4-7


The only questions in this bounty are :

Is there something I can use to 'watch' the connections? (Maybe cut the cable and add a circuit/monitor??) Is there a better method?

The board with 5 transistor switches and 7 screw terminals implies you have 5 outputs and 2 inputs for 24Vac power and return.

My assumption is you know how to use Ohm's Law and trace circuits and do design at this level. Otherwise, much to learn.

With a load, it is trivial to measure when power is switched on and locating the common screw voltage to which the power is connected to.

There is no need to cut cable and monitor current unless you wish to diagnose connection faults in the solenoid, then use an ommeter with power off.

  • \$\begingroup\$ For the seven screw terminals, there are 5 switched outputs, one constant output, and the second AC phase. The 14 pin cable connects to the back plane and nit directly to the module. \$\endgroup\$ – shaun5 Sep 5 '17 at 3:25

This is just a helper and not a complete answer.

The 3-pin SMD package might be just a diode, explaining why you can't find where the third terminal goes. I found the image below on an eBay ad for "MMBD4148 1N4148 Diode 200mA 100V".

enter image description here

Figure 1. SMD diode. Three pin package used for automated assembly.

See BAS16 / MMBD4148 / MMBD914 datasheet for a similar diode package and pinout.


simulate this circuit – Schematic created using CircuitLab

Figure 2. Less unlikely schematic?

I've rearranged your schematic into a less unlikely arrangement. The varistors would be across the solenoid coils to absorb any inductive kick and protect the triacs. In this arrangement a negative pulse would be required on P5 to trigger the triac. (This seems unusual. A positive pulse would be more likely.)

Things to check:

  • Use the diode test function on your multimeter to see if the three-terminal device is just a diode.
  • See if my wiring diagram is better than yours!

Alternative solutions:

As is evidenced in the comments, you should consider replacing the control system with an off-the-shelf solution such as a micro-PLC (programmable logic controller).

enter image description here

Figure 3. A micro PLC with keypad and LCD.

These are available for the price of one of your output modules but include the controller! They typically feature eight inputs and four outputs on the base module with expansion modules available to plug in or network connection. The one in the photo is from Crouzet but most of the big manufacturers - Allen-Bradley, Siemens, Mitsubishi, etc., have a version. You would require the relay or triac output version for your application.

  • \$\begingroup\$ I've revised the schematic. I probably still have the triad oriented incorrectly. The control boxes have up to 22 channels. The PLC's don't allow me to recycle the old controllers and don't allow me to replace the controller with a web based controller. \$\endgroup\$ – shaun5 Sep 6 '17 at 1:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.