Help to understand part of circuit found

Thread Starter

ahto555

Joined Mar 12, 2009
11
I am working on project to keep preset temperature inside chemical reactor by changing cooling water amount pumped through reactor. Because quite high precision is needed I was thinking about PT100 (class A or B 1/10) sensor as temperature pick-up. The first idea was to PWM cooling water DC pump but after reading I decided to go proportional solenoid valve instead.
After searching similar projects in internet following circuit came up - http://gillesenergies.webs.com/evalve.htm. As I understand from SSR used the solenoid switching frequency is not very high and simple/cheap NC (normally closed) 12VDC solenoid valve is used which is very temptating because of price difference.
Unfortunately I can not understand how this circuit works (555 and IC1D). Tried to simulate with different R17, R18 and C3 values on Tina TI but could not get oscillating waveforms like on diagrams.
Can someone lead to right track?
 

GetDeviceInfo

Joined Jun 7, 2009
2,192
I could not see any refernce to the valve itself. I would be very hesitant to operate a solenoid valve on PWM unless it was specifically designed for such.
 

Thread Starter

ahto555

Joined Mar 12, 2009
11
I agree with you. I tried to PWM simple 12VDC (NC) brass solenoid valve few years ago with 555 based circuit. The frequency was too much (I think something close to 20 kHz). So result was following - at high PWM % it was open and at about 50% it just closed down.
As much as I know today it should be possible to open/close "normal" solenoid valves up to 10 times per second. There are more expensive valves available (sometimes called pulsoids or something) which can handle up to 100 Hz. Seems from diagrams provided circuit opens and closes at low frequency (below 10 Hz) depending on temperature reading from LM334H but I can not figure it out how.
 

wayneh

Joined Sep 9, 2010
17,496
As much as I know today it should be possible to open/close "normal" solenoid valves up to 10 times per second.
I'd rethink the strategy. Putting a DC pump under PWM control would be simple and effective. A solenoid is OK as a switch but not so much as a linear control.
 

GetDeviceInfo

Joined Jun 7, 2009
2,192
a motorized valve could be arranged as 'linear', and are relativley common in industry.

The problem your going to run in to with PWM a soleniod is that the fluid can only permis a certain volumetric motion before the onset of cavitation. Prior to this, the fluid has enormous strength in resisting such tendencies.

I would be all for the 'linear' approach in utilizing a motorized valve, or a proportioning valve. I'm not aware of such, but there may be a proportioning valve driven by PWM, similar to the frequency controlled servo motors.
 

wayneh

Joined Sep 9, 2010
17,496
a motorized valve could be arranged as 'linear', and are relativley common in industry.
Heavens yes, but I took the OP's need to be much smaller than industrial equipment. I guess we don't know that for sure. I was picturing a small benchtop reactor and a coolant flow of maybe 1 gpm. Like an aquarium pump! OP, what exactly are we talking about here?
 

Thread Starter

ahto555

Joined Mar 12, 2009
11
Yes, we are talking about coolant flow max few gpm. Proportional solenoid valves controlled by voltage are available but quite expensive. One I need is in range 150-300 USD.
I finally also got how this circuit works. It is quite simple. 555 is few Hz astable generator and sawtooth waveform is taken from charging/discharging capacitor. IC1D is comparator where output pulse with and PWM % depends on reference input voltage.
 
Top