Analoge output controlling loading solenoid valves!

[naifmbo]

hi everybody,
can i use the pid-loop output to control my recip-compressor's loading solenoid valves, for example:
solenoid valve 1 will be energized if pid output > 10 %
solenoid valve 2 will be energized if pid output > 30 %
and so on..

and why?

thankyou alot.

 

[MintJulep]

PID loop output of what?

Probably not, at least not directly. You could probably cascade the analog output of the PID controller into an analog input, and use the value to trigger one or more digital outputs.

Why? I don't know why you would want to do that. The value of a PID controller is largely lost if you have only on/off devices to control (unless you have a lot of them)

 

[IRstuff]

A solenoid is highly non-linear and will probably give your servo loop fits.

One possibility would be to convert the analog into a pulse width modulated signal that can control the solenoid.

TTFN

 

[sterl]

The solennoid on most recip compressors is in fact a pilot device and causes cylinders to unload by either applying or bleeding away oil pressure, which in turn usually causes the suction valves to either seat/unseat automatically in response to compression space pressure or to remain unseated, thus unloading the machine.

If this is a large machine using such an arrangement, you could do a lot of damage to unloader mechanism, valve discs and plates applying pulse width modulation at anything like normal pulse rates and intervals.

If the motor and the system will tolerate it and you need to use P & ID (though I don't know why you would) consider a VFD. We have applied them through 600 Hp on recips...

 

[HVACctrl]

Won't the PI (I left off the D b/c it is sometimes not used in tis type of control) be effective, still, in eliminating the "offset" effect of thromstatic control? Consider that it is 1.8 deg. too cold, but heating doesn't kick in until 2 deg too cold. A thermostatic or offest control (ie, a receiver/controller) would not let the space get warmed up. PI control would count the duration of time away from setpoint and appropriately allow heating.

I don't know much about loader sol. valves, but I knnow the loop can be used to drive DOs in this fashion- whether it will harm the machine in this scenario, I can't tell you.

Ed

 

[ClydeMule]

Sure, it will work. However, as posted, you won't get full modulation because you are only controlling on/off devices, but you will remove the offset like with a standard differential control.

Check out

http://www.kele.com.

They have this nifty contoller that will hook up to the output of a process controller. You then use a potentiometer to control when the out put relays come on.

So if your controller has a 0-1VDC output you can set unloader #1 to 5 V and unloader #2 to 10VDC.

 

[pennpoint]

You can use a PID output but it should be an ON/OFF signal to the solenoid not analog.
Analog is coventionally 0 to 5V, 0 to 10V or 4 to 20Ma or...
Although there are solenoids that are proportional to a variable signal these devices are not usually adapted to the device or machine you want to control, eg, Carlyle Refrigeration compressor unloaders will not respond to a process signal of any kind except on/off at the rated coil voltage.
Thats not to say they will not respond to a time proportioned signal of X% time ON (open)and X% time OFF (closed).
IMHO

Good Luck

pennpoint


Best regards
pennpoint