View Full Version : Controller Theory - PID and Compressors?
TheAMAZINGNorad
04-16-2007, 10:12 PM
I am working on finishing up my chiller, but I am stumped on how to control it. I would like to fix the temperature within a certain range, but how do I do it? I know enough from my classes to use a PID controller. However, you can't cycle a compressor on and off that quickly. I see various digital controllers used, both aftermarket and homemade (G-Froster). How do these devices control the temperature of the liquid/processor without dealing with issues of time between cycles? I was under the impression that you need to allow pressure to equalize before you can restart the compressor and that can take several minutes. What is the answer?
Thanks,
Chris
Xeon th MG Pony
04-16-2007, 11:33 PM
Delay on break timer. IE just use a delay loop befor the next cycle.
wdrzal
04-17-2007, 03:44 AM
don't you want coldest temps possible??? leave compressor run continually
a proper size txv or cpev will work if size & type are correct,can be complex if mixing gases(feel lucky and but so will a simple cap tube that will balance itself out.
marru
04-17-2007, 04:27 AM
I am working on finishing up my chiller, but I am stumped on how to control it. I would like to fix the temperature within a certain range, but how do I do it? I know enough from my classes to use a PID controller. However, you can't cycle a compressor on and off that quickly. I see various digital controllers used, both aftermarket and homemade (G-Froster). How do these devices control the temperature of the liquid/processor without dealing with issues of time between cycles? I was under the impression that you need to allow pressure to equalize before you can restart the compressor and that can take several minutes. What is the answer?
Thanks,
Chris
Those so called controllers are nothing but monitoring devices, the only thing they control is the compressor and the system. They are not designed to control temperature. That is a totally different story that you need a lot more reading to do.
Cheers
Ssilencer
04-18-2007, 03:13 PM
Those so called controllers are nothing but monitoring devices, the only thing they control is the compressor and the system. They are not designed to control temperature. That is a totally different story that you need a lot more reading to do.
Cheers
So, what is a so called controller?, if it is controlling and monitoring the compressor and the system are controllers, not thermostats or electronic metering devices SO CALLED electronic valves.
@TheAMAZINGNorad
XEon is right, just use a delay loop befor the next cycle
@Walt
Usually people that use chillers likes to keep the water at a desired temperature, why? for example to avoid condensation; so the use of a thermostat is mandatory
marru
04-19-2007, 03:06 AM
So, what is a so called controller?, if it is controlling and monitoring the compressor and the system are controllers, not thermostats or electronic metering devices SO CALLED electronic valves.
I haven't seen any electronic metering devices controlled by the controllers used around here thats all.
Yea, I was gonna suggest a thermostat..the only thing I`d say is be wise enough to set start and stop temperatures that will take minutes to reach so the compressor will have enough downtime to not throw a hissy fit
E.G.... compressor on at 22, off at 17 (avoiding condensation) tho I guess it depends on the size of the compressor and charge how long it`ll take to pull down the water temp
:)
n00b 0f l337
04-19-2007, 04:48 AM
I'd also imagine that if you needed a quicker equilization, a solenoid bypass, not a hot gas bypass (dont want to add more heat to evap!), that is closed with power, and open without could be installed. Run that with the thermostat and the unit and your equilization should be quick enough for even a 3-5 minute restart time.
wdrzal
04-20-2007, 04:20 PM
you can also use a potter b relay with a time delay built in or if you know enough about electronics you can build a delay timmer by using a 555 ic timer chip,really cheap chip.
Cronos
04-20-2007, 10:12 PM
Can anybody explain why we need pressure equalization?
Why just don't use solenoid valves to seal the low and high parts with power off, so that the pressure state will be preserved until the next start?
Xeon th MG Pony
04-20-2007, 11:02 PM
Harder for compressor start up when loaded, with a HST compressor it isn't an issue, all so known as Hard starting, Eq starting is easier on the compressor with less inrush but not 100% requirement.
Most systems I build I use a pump down cycle in order to eliminate the accumulator so a HST set up is needed. For him using highly restrictive cap tube bypassing the TXV would more then suffice that or a bleed port TXV for use on LST systems. and for control a delay loop with a fairly wide differential will be quite adequate.
There are many many options that one can take, so he must choos how his design will be made, HST or LST, Bleed port TXV, Cap tube, Standerd TXV, Pump down/ no pump down, Coolant return temp and differential, control system exchanger type and so on & on.
For temp control any simple therm with a 10C diff aut to work well, and a schmit trigger or a anti short cycle timer will be peachy (A nice wide diff usualy helps prevent that)
vBulletin® v3.7.0, Copyright ©2000-2008, Jelsoft Enterprises Ltd.