MF output source + output sink on on/off valve

[spittet]

Hi,

I try to control 4 on/off valve (so, 8 coils) by using only 6 MFoutputs of a MC050-110 controller.
I have 2 outputs that are configured as Sink only and only one of these 2 is ON at a time.
I have 4 outputs that are configured as Source only and only one of these 4 is ON at a time.
So, at a specific time, only a pair of Sink and Source outputs are in the ON state, so only a single coil should be activated at a time.
My coils are wired as follow:
ValveA-Coil1 = Source1 & Sink1
ValveA-Coil2 = Source2 & Sink1
ValveB-Coil1 = Source3 & Sink1
ValveB-Coil2 = Source4 & Sink1
(Valves C and D are the same as above but with Sink1 replaced by Sink2.)

The problem is that the LED of the DIN connector is illuminated on 2 coils at the same time. By example, if I set Source1 and Sink1 at ON, ValveA-Coil1 is activated as it should, but the LED of ValveC-Coil1 is also illuminated, but function not active (coil not energized).

It seems like even if Sink2 is OFF, some current seems to flow and the LED illuminate. I logged FeedbackCurrent of Sink2 output in the above example and the value is really small at aroud -20 to 20 (-2 to 2 mA).

Is there a workaround for this? My goal was to minimize the number of outputs used.

Have a good day.

[spittet]

To add to the above post, I’m doing it this way because we plan to add 2 other valves (so 4 other coils) in the future. Doing it this way, I would only need a third Sink output (Sink3) to do another combination of 4 possibilities.

Also, now that I try the hydraulic functions a little more, I can see the function of ValveC-Coil1 moving really slowly when activating ValveA-Coil1. So, it seems that the leakage current that was enough to light the LED is also enough to move the spool of the valve when not wanted. This problematic.

My next move will be to add a diode pointing to the valve on each SourceX wire for each coil, configure all the outputs as Push/Pull and inverting the SinkX outputs, but this is kinda defeating the advantages I was looking at. I would really like to avoid the diodes in the harness as this is another point of failure.

[G30RG3]

Theoretically, If current consumption isn't an issue, you could set both coils in series to the main voltage, and use a single pin to sink or feed the middle voltage.
That would give both coils half current in "unpowered" state, and only one coil the full current in active mode. leds would be going out on "shorted" coil instead of being on when powered...

[oiltronic]

Yep to diodes.  With this shared-common approach three other coils get put in series when one is turned on.  (E.g. turning on A1 results in power through C1, C2, A2.)  End result will vary depending on whether flyback diodes are across each coil.  You would get a good picture of what's going on by measuring the voltage across every coil when trying to turn on a single one.