Hi to all... I have been struggling in an arduino application, is a control board... and there is weeks when everything is all right... and weeks were the arduino make weird stuff... Now...

This is the original question;
http://forum.allaboutcircuits.com/threads/how-to-avoid-return-v-to-turn-led-on.104370/


I ask in the forum and the suggested answer was; Electrical Noise...

Makes a lot of sense... because the arduino is in a box turning on and off contactors, and depending of a voltage line that a lot of inductance and motor loads...

So today I add A couple of snubbers to the contactors... And I intend to do a filter for the VCC to arduino.

Currently the arduino gets its VCC from a switched power source (Like the old cellphone ones)... so no great signal stability...

There has been a strong recomendation to add multiple capacitors , between the power source and the arduino, to get less noise of the main line.

So today I buy 5 capacitors; 3300uf, 2200uf, 22uf, 1uf, 0.47uf...
And I start to make a schematic on Proteus 8, I want to know if this filter is good enough for futures applications add It at the controls as standard part.

However.. I dont know how to simulate something like this... is a AC generator? a signal generator? ??
Or more concrete... how can I see if the selected values will provide an signal stability for the Arduino, I know Im not giving any parameters.. but I trust some one in the forum, wise and good will provide us with a parameter.

Here is the file;

https://onedrive.live.com/redir?resid=6727EC83F4A63D75!255

Thanks for the suggestions.


Thanks in advance.
-Alex.

 

Im doing this right?

This means the filter is good?

Im sorry I dont have a very strong electronics background.

 

You're saying you want to filter the supply to the Arduino? It has on-board linear regulators, I'm not sure a little supply noise will matter.

 

Yes precisely... the arduino has an On - board power switch how ever ... this particular arduino is in a control box, near contactors, relays 110, 220 V and has was discuss...
the opening and closing of the contactors may lead to a high voltage peak... induced on the mains... and for the lack of filtering on the arduino ... stop the program at a random place...

This is an observable effect and has happend several times.
I already add a couple of snubbers to the contactors poles, yet... I want to make the filter for the Arduino.

So... do you think the above graphs proves that the arduino will not experiment a variation higher than 20mV at a given time... or... I made a miskae on the simulation or in the results interpretation?

Thanks.

-Alex

 

This is how we do that.

 

Thanks.... So any other cap value is to big for the transcient? .

The above simulation is good?

Thanks.
Alex.

 

I don't simulate. Any size capacitor that works is good, but you need a high frequency compensator for the electrolytic capacitor. It is the ceramic capacitor.

 

A guy from the arduino forum told me that there are somo transient noise eliminators . a ferrite core with a copper bobin and in the other phase there is a similar bobin but in reverse direction .

Is a complementary solution along with the capacitor? Or is good to use just the capacitors?.

 

Those dual winding filters go on the power wires. They should be on the wires that go to the relays so the noise can't get very far from the relays.

I think your simulation is wrong because there is no impedance between the signal generator the the filters.

Noise problems can be difficult problems. You need to know what you're looking for and what you are measuring to be sure. The only things we can do here is show you some methods and you see if they work in your problem.

 

Yes I agree... I put a couple of snubbers in the two nearest contactors and suddenly the program don't stop more... Of course I need to filter the vvc for the arduino , there is plenty of electrical noise in the 220circuit.

So tomorrow I will draw a squematic of the intended new vcc filter... And maybe test it on the field

Thanks.
Alex.