Relay-controlled valve

[mirh]

Long story short, my flow valve has a motor that works with polarity reversal.
And through the power of DPDT switches, we could just say it has two commands/states: "more" and "less".

I tried to half ass together something in Motor.ino, but clearly there's so much more to it.
Would it be too hard and disruptive to add support for this?

[SK21]

Are you using the motor option or the standard valve?

[mirh]

I tried to tinker with the standard valve code, after having retooled the PWM pin to be just a digital one outputting the opposite of the direction state... but it kinda failed.
But I think this would be worth a new type of motor tbh (is "proportional" even the right word? I have no control on the velocity, maybe "linear valve" would be better?) since nothing in the PID settings tab would apply.

[SK21]

I am not clear on how your motor works. A standard valve has more and less commands. The motor control varies the voltage from 0 to 12 v.

Do you have a model # or link to a page about the motor?

[mirh]

I have this brochure (last one) about the body, and this should be the valve-only.

A standard valve has more and less commands.

Yeah, but you also have PWM to regulate speed, while "effect direction" is the state of a single relay.
Here instead, you just supply +12V or -12V (which at the end of the day I made to correspond to two separate relays) and that's about it.

[SK21]

In the last brochure about the valve it looks like the motor only controls on/off, not flow.

[mirh]

It's on/off in a sense, yes.
But the "obturator" can end up sitting in more than just two states, if I can explain.
And this way (since polarity reversal makes for two separate "ON" eventually) you can control flow.

[bduncan175]

It sounds like his valve is similar to a standard raven flow control valve that is plumbed as a bypass system, but can also completely shut off. https://www.spraysmarter.com/raven-precision-1-poly-control-valve-flanged.html?gclid=CjwKCAjw-ZCKBhBkEiwAM4qfF1c-M0oKpccFcGTAFIvWbRJOzAW5RACnZJ9z2HZ7LF6Ss77eIKoxUhoCoMEQAvD_BwE

[mirh]

Sorry, I snafued the valve brochure. This should be the right one.
To reiterate again: with +12V (i.e. one relay) the ball rotates in one direction, with -12V (an other relay) the opposite one.

[SK21]

So you would like to control the pressure with 2 relays, one for increase and one for decrease? A motor controller would not be used. I could change the code to add another control type that would use relays instead of a motor controller.

Motor.ino:

[SK21]

Should be:

[SK21]

It could also work with a motor controller if the PWM is set to maximum. This would give 12V.

[mirh]

A motor controller would not be used.

I think the confusion is due to the valve mechanism technically having one, but the DPDT wiring I used making it logically be just two switches (I mean, I'm not sure how else I could have controlled polarity inversion otherwise).

Anyhow

// use 2 relays for flow control
if (FlowEnabled[i])
          {
              if (pwmSetting[i] > 0)
              {
                  // increase
                  digitalWrite(FlowDir[i], 1);
                  digitalWrite(FlowPWM[i], 0);
              }
              else if (pwmSetting[i] < 0)
              {
                  // decrease
                  digitalWrite(FlowDir[i], 0);
                  digitalWrite(FlowPWM[i], 1);
              }
              else
              {
                  // no change
                  digitalWrite(FlowDir[i], 0);
                  digitalWrite(FlowPWM[i], 0);
              }
          }
          break;

was exactly what I needed (at least when I replaced the fast close valve block for testing purposes).

I think maybe you could replace the PID tab in this case, and have some kind of "tolerance" value to be set? Without derivative control the relays aren't gonna have a good time otherwise when near the target rate.

[SK21]

In the PID tab Deadband is the "tolerance" setting. In this example if the error is within 4% no adjustment is made. The other settings will just be ignored.

This occurs in PID.ino. sDeadband is the 4 from the PID tab. The sub returns 0 if within the tolerance, so the relays should both be off in the your block above.

[mirh]

I see, sorry (not that my coarse testing with the "manually spun" flowmeter could have shown it working probably, lol)
I guess it would still be kinda more polished if the UI reflected the underlying enabled logic, but it's no biggie.

Btw while flow control responds and offset adequately when I turn off some section, when all of them are closed it seems like signals are still being sent to the valve.

[SK21]

I think the "if (FlowEnabled[i])" statement needs to be removed in your block above to prevent signals still being sent. Just use: