Noise on Ice detection Circuit.

Discussion in 'Analog & Mixed-Signal Design' started by Theevan Venugopal, Oct 18, 2016.

1. Theevan Venugopal Thread Starter New Member

Dec 30, 2015
16
3
Hi Friends

Am trying a circuit for ice level sensing by means of three probes(1.PWM signal, 2.Minimum Ice level 3.Maximum Ice level). The logic is by means of the returning PWM signals from min and max probes, a comparator checks the voltage level and gives the output as high if it is water and low if ice formed. The circuit is fine when works with plastic containers but noise occurs when we do the same experiment with metal tank. So what happens is the following

1. Due to Noise, always a high level is detected by the comparator.
2. So the compressor is not cutting off even after ice reaches the maximum probe level.
3.Then we tried to give the earth(ground) to the system which nullifies even the PWM signal.

Please help us to find a solution in the same ckt or please recommend any other ideas that can be implemented practically asap. Here by attaching the circuit.

File size:
86.5 KB
Views:
28
2. AlbertHall Well-Known Member

Jun 4, 2014
2,261
447
Could you use temperature to detect the ice? Dehumidifiers use this method.

Or something which moves when in water, but would be stuck in ice.
A vane, an electromagnet, and a microswitch or proximity detector to detect the movement.
A motor in a box twiddling a magnet, with another pivoted magnet outside the box which will follow the magnet inside the box, and a reed switch or hall switch to detect whether the outside magnet is rotating or not.

3. Sensacell Well-Known Member

Jun 19, 2012
1,183
276
I don't think this approach will ever yield a reliable result.

I have never heard of a conductance probe system used to detect ice, probably because it is fraught with problems.

Theevan Venugopal likes this.
4. Theevan Venugopal Thread Starter New Member

Dec 30, 2015
16
3
No man..nothing like that.

5. AlbertHall Well-Known Member

Jun 4, 2014
2,261
447
Why not? Explaining that may help provide other alternative solutions.

6. Theevan Venugopal Thread Starter New Member

Dec 30, 2015
16
3
Could you please suggest any other methods?

7. wayneh Expert

Sep 9, 2010
12,379
3,232
Have you tried un-grounding the metal tank, to make it like the plastic containers?

By what mechanism are you detecting ice, by it's loss of conduction compared to water? The absence of ice means water (not air)?

I agree with @AlbertHall and @Sensacell: You need to better explain your system.

ronv likes this.

Dec 30, 2015
16
3

9. Theevan Venugopal Thread Starter New Member

Dec 30, 2015
16
3
Hi Wayneh

Yeah i have tried ungrounding the circuit. But it is not working good. When i went through certain articles available in net they all provide solutions like TI's capacitive sensor and all. But we are looking for a low cost system which is very easy to implement.

10. wayneh Expert

Sep 9, 2010
12,379
3,232
To do what?
Don't make us guess.

11. Theevan Venugopal Thread Starter New Member

Dec 30, 2015
16
3
Once again i am quoting my requirement @wayneh

My motto is to keep the ICE level from Minimum probe to Maximum Probe inside a water dispenser unit. The Compressor should turn ON when the Ice level reaches minimum probe and Compressor should be OFF when it reaches Maximum Probe. But what happens is since we are getting a noise in the system(always),the compressor is not getting OFF.

12. AnalogKid Distinguished Member

Aug 1, 2013
4,686
1,297
Consider for a moment that when 2 different people ask you the same question three times, then re-quoting your requirements is not a sufficient answer. For example:

When you say "ice level", are you talking about ice cubes in a dispenser, or a solid block of ice that somehow grows and shrinks, or what?

ak

Theevan Venugopal likes this.
13. Theevan Venugopal Thread Starter New Member

Dec 30, 2015
16
3
Yeah @AnalogKid Solid Block of Ice that grows and shrinks....Thank you AK...... and sorry all ...i didn't get what you mean before.

Apr 30, 2011
1,426
363
I'm aware of two methods used in commercial ice makers to detect batch completion.

One uses an electrical contact, possibly similar to your method.

The other uses a small tank with two float switches where the tank is filled to an upper limit, then this fixed volume of water is circulated over the freeze plate and back to the tank until enough of the water has tranferred to the freeze plate to actuate the lower limit. I like the simplicity of this method.

I still don't know what your system is intended to do so the foregoing information may or may not be relevant to you.

15. ian field Distinguished Member

Oct 27, 2012
4,447
791
A diode bridge with one of the 4 diodes serving as the sensor, simply feed the output of that into a window comparator.

It has been the basis of a number of automotive icealert indicators.

16. wayneh Expert

Sep 9, 2010
12,379
3,232
So you need to discriminate between air and ice? Couldn't you use a break beam sensor, the kind with the sensor right next to the beam? No ice, no light on the sensor. Covered by ice, the beam would be reflected onto the sensor.

A float valve mechanism comes to mind also. Lots of options.

Theevan Venugopal likes this.
17. Theevan Venugopal Thread Starter New Member

Dec 30, 2015
16
3
No i want to discriminate water and ice @wayneh

18. Theevan Venugopal Thread Starter New Member

Dec 30, 2015
16
3
the compressor and condenser are shorted so both turns ON and turns OFF at same time. The condenser coil is immersed in water and ice grows and shrinks on this coil. The minimum and maximum level probes are placed at a distance to these coils.

19. Alec_t AAC Fanatic!

Sep 17, 2013
5,968
1,135
Try increasing the values of all four capacitors on the comparator inputs. I suggest at least 10uF.
There is something wrong with your setup if grounding the tank nullifies the PWM signal. Can you post a pic of the probe arrangement?

20. Theevan Venugopal Thread Starter New Member

Dec 30, 2015
16
3
Thank you all for your support. Without you i can't have done this... see you soon