Having a slight issue with interferance of the one wire sensors in my panel.
My panel has a 12v safety circuit for manual overide of control to the heating elements via contactors.
When these contactors are pulled in or out, there is a momentary jump in the readings to the one wire sensors to 85c. This can cause issues if it happens while a script is running as the script sees the 85 ( for example if heating to strike temp and the actual temp is 30c but the momentary spike causes the script to think strike temp has been reached and then progresses on to the next step)
While building the panel i did my best to keep all control and mains wiring on opposite sides of the panel but as all my inputs and outputs are on the top of the panel it was unavoidable to have them close together in this area.
Im certain this is a hardware or wiring issue and not an issue with brucontrol but thought i would ask here as it may help others avoid the same thing happening in their build.
Is there any easy way to avoid this from happening?
A couple thoughts i had were to
A) add some cable shielding to the one wire cables which are near mains voltage wiring inside the panel
B) add something into the script which would allow the software to avoid momentary anomalies in sensor readings.
Any thoughts or advice would be appreciated.
Electrical interference can be tricky to debug. First make sure your wiring adheres to standards. Many problems stem from improper grounding. All ground points of devices should be wired into a star pattern, meaning one common ground point. Having multiple ground points separated by wiring can creat ground differentials and allow for noise.
Another area for interference is improper or inadequate power supplies. You should always use a quality switching power supply, rated 2-3x your design’s actual max.
Anothet area for noise is relays/contactors not having flyback diodes. Assuming these are DC coils (as I think you mentioned) they should have a reverse-biased diode across the terminals. When the coil power drops, the stored energy in the coil indices a voltage opposite of the driving voltage. Without a diode, that voltage spikes back into the control system. Without specific part numbers and schematics, I can’t provide any more feedback whether this may be your problem or not.
Short length wiring, isolated wiring, twisted/shielded wiring, and noise filters such as inductors and/or capacitors can help. Perhaps a simple high value capacitor across 1-wire power and ground wires can help filter spikes. A ~5.5V zener diode across the 1-wire power may help shunt voltage spikes too.
Maybe some pics/schematics of your setup can help so we can address this in hardware before we chase the software fix (script).
In your example schematics you have the mains earth connected to control ground point, i was unsure as to wether this was a good idea during wiring so i left it out.
These are the contactors i used, not sure if they have those flyback diode or not
https://uk.rs-online.com/mobile/p/contactors/7448014/
I will try a capacitor across the 1wire see of that helps.
This is the power supply i use
https://uk.rs-online.com/mobile/p/din-rail-panel-mount-power-supplies/0428467/
It powers the arduino, a 3a pump, small panel alarm and 3x contactor coils
Looking at the datasheets, I don't see any mention of a diode. I am not sure if this is an issue for sure since you are seeing the problem on both power on and off transitions. But I think its worth a shot... use a general purpose / recitifer diode like a 1N4001. Wire it with the line on the + terminal of the coils and the other side to the other terminal of the coil.
I think the PS is good - not sure that is causing the problem. If you had a scope, you could look at the pulses on the +12V line and see what is up.
We added the earth ground to system ground as an optional line - it can help reduce noise and may add some safety to the system. Give it a shot.
I stand corrected, it is only when the contactors drop out that this occurs not when they pull in. Have ordered some diodes so will try that.
I will try the extra link between mains and control grounds. When wiring it up this just didnt seem right to me, thought if there was an earth fault this might damage something on the control side, but im only an amateur with electronics so ill trust your knowledge and add this in.
Thanks, will update if it solves the issues
Ok. The diode will likely fix it. Let us know!
The 1N4001 diode across each contactor coil and extra wire between control ground and main earth terminal has resolved my issues. All working well now.
Thanks.
Awesome!
