A component that let <12v pass but block >11.9v, zener diode?

Hello, i am searching for a component for my circuit.
The circuit is made of 4 dpst relay that allow me to switch from parallel to series the 3 li-ion batt (3x1cell) when a send 12V to the coils.

This works fine, however i would like to add a security beause i encounter a problem with my power supply:
When i switch off the power supply that control the coils, the voltage goes slowly from 12v to 0v. The relays triggers around 11v so they trigger one after another and this is very bad for my application (shorting the battery for a second).
However, if i unplug the cable quickly, things works fine.

So i am searching for a component that acts like a wire when there is 12v or more and and open switch for voltage <11.9v for instance

I could use one more relay to do this but i am sure there is a better solution and this is a educational project:)

I am not sure this is what a zener diode do, is it?

alimentation

Means nourishment according to the dictionary. Please let the ignorant (me) know what it means in this question.

Does it mean switch?

Please give us a drawing of how you are wiring the relays and batteries.

Weedpharma

I am sorry for my bad english i corrected my post. I meant power supply :)

Here is a diagram, all the coils are wired in parallel and the home station can be unplugged so robot works on battery |500x298

Alimentation == supply Tension == voltage

Groove: Alimentation == supply Tension == voltage

corrected!

Firstly, a couple of errors in the circuit. The relays should be normally closed in the series configuration as you want them to switch to parallel only when the charger is powered.

Next, the parallel contacts of RY22 and RY24 should go directly to the charger, and not through the contact of RY21.

Given this, it is sufficient to ensure RY23 and RY24 switch to series mode as the control voltage falls before any of the other relays, and only switch back to parallel mode after the others as the voltage passes the higher threshold. If you can set these separate thresholds, then it will be safe.

In general if you have a physical arrangement of switches or relays that has one or more configurations that shorts out the supply, you have a bad circuit and you need to fix it first. Don't try to work around, shorting the supply is never to be tolerated.

MarkT: In general if you have a physical arrangement of switches or relays that has one or more configurations that shorts out the supply, you have a bad circuit and you need to fix it first. Don't try to work around, shorting the supply is never to be tolerated.

.It is true that if one of the relay fail, it will lead to short circuit.

To protect from this i bought some li-ion 1cell protection board (over voltage/over current/low voltage shutoff) that i will put between the batteries and the relay. this is the only solution i have found. But for now those boards are on the boat... I am not sure this is the right way to do.

Firstly, a couple of errors in the circuit. The relays should be normally closed in the series configuration as you want them to switch to parallel only when the charger is powered.

I am not sure i get what you mean. In this exemple you have to consider that the robot is at the home station so the coils are powered. This mean the cells are in parallel In the diagram we can see all the cell - connected together up to the ground and all the cell + together to the charging board.

Look at using two 3 pole double throw, break before make relays. One relay for the positive sides of the battery cells and one relay for the negative. Even so, fuse each battery. Since the poles are tied together, they all act at the same time. You only need two relays. The way relays work in parallel is that as the magnetic field decays, one relay armature will release first. This causes a sudden lowering of inductance for that relay. It will become a lower impedance path for the current of the other relays, holding them longer. This repeats until the last relay drops. The circuit you have is always going to fail. Dwight

glub0x: I am not sure I get what you mean. In this example you have to consider that the robot is at the home station so the coils are powered. This mean the cells are in parallel

The way you have drawn it shows the cells connected in parallel with the coils unpowered.

The normally closed (unpowered) contacts in a diagram are those furthest from the relay coil.

Paul__B: The way you have drawn it shows the cells connected in parallel with the coils unpowered.

The normally closed (unpowered) contacts in a diagram are those furthest from the relay coil.

Oh yes this is true. i didn't draw the wires to the coils because this create quite a lot of wire and makes the drawing confuse All the coils are wired in parallel.

So i search and of course the zener diode doesn't do that ( at least i learned something ) I tried with another relay and it works fine tested for one hour yesterday with the batteries ( couldn't wait the protection board) BUT I have drawn my circuit here and i realise it shouldn't work... |500x339 When RY17 triggers, it is short circuited so it should go back to normal and then triggers ect.... [edit] Except for this "little" problem, when home goes off and tension slowly goes down, RY17 is the first one to trigger, this lead to all the other relay going off at the same time hence avoiding RY24 going off 1second before RY25.

Also RY17 is a dpst because i have 10 of them and no Spst ...

You still have (all) the relays drawn so that the batteries are connected in parallel when the relays are not energised.

And you have not taken note of my suggestion that the "parallel" contacts of RY22 and RY24 go direct to the charging circuit.

Paul__B: You still have (all) the relays drawn so that the batteries are connected in parallel when the relays are not energised.

I understood what you mean now! On this online program i use to make the drawing, i can't switch the position of the relays, i am sorry. This is the powered state and i cannot draw it properly :(

Paul__B: And you have not taken note of my suggestion that the "parallel" contacts of RY22 and RY24 go direct to the charging circuit.

this i just understood too, i must be very clever at 4 am, i will draw that tomorrow

thanks for help

|500x281 Here is the new diagram with Paul__B advice implemented :)

(this is without the relay to control the coils )

You need to consider what can happen if one relay operates before another due to any possibility.

If the charger/robot changes quickly and the others have not operated, there is the small possibility that the 3 batteries in series will be applied to the output of the charger. Many chargers do not like power being applied to their output, especially at a lot higher voltage.

Weedpharma

Use three DPDT relays . . . . .

(coils not shown)

Considering what happens if one cell goes faulty or simply starts discharging quicker than others, it's better to have three chargers.

weedpharma: You need to consider what can happen if one relay operates before another due to any possibility.

If the charger/robot changes quickly and the others have not operated, there is the small possibility that the 3 batteries in series will be applied to the output of the charger. Many chargers do not like power being applied to their output, especially at a lot higher voltage.

Weedpharma

The other issue with one deeply discharged LiPo battery connected in parallel to two higher charged LiPo batteries is ... FIRE !!!

weedpharma:
You need to consider what can happen if one relay operates before another due to any possibility.

If the charger/robot changes quickly and the others have not operated, there is the small possibility that the 3 batteries in series will be applied to the output of the charger. Many chargers do not like power being applied to their output, especially at a lot higher voltage.

Weedpharma

I guess a diode would protect from that, i will add it.

mrsummitville: The other issue with one deeply discharged LiPo battery connected in parallel to two higher charged LiPo batteries is ... FIRE !!!

This should not happen because 1 each battery will have a protection board such as this one http://www.ebay.fr/itm/PCB-Charger-3-7V-3A-Li-ion-Lithium-18650-Battery-Input-Ouput-Protection-Board-/252162638493?hash=item3ab610829d:g:ofkAAOSwF1dUR9LC

And i will charge all the batteries in parallel so there should not have much voltage difference between the cells

Finally i use LI-ion battery salvaged from a computer. It can go to fire but should less impressive than lipo. Yet my robot is made of wood so this is a real concern to me :(

@Archibald 3 chargers is an option BUT over time the 3 cells of the battery will have different voltage because they are never charged together. this will lead to one battery cell failling. I am not sure this is a real problem or not.

glub0x: @Archibald 3 chargers is an option BUT over time the 3 cells of the battery will have different voltage because they are never charged together. this will lead to one battery cell failling. I am not sure this is a real problem or not.

Over time, years hopefully, it is likely that one cell will start to fail before the other two. Then when your connect your cells in parallel, a large current could flow from the two good cells into the failing cell. Even with fairly new cells, when they need recharging, you cannot guarantee that the voltages will be equal before you connect them in parallel. I do not know how important this issue is. You could add three diodes (to the right of the relay contacts in my circuit diagram) but then you would need to adjust the charger if possible to allow for the voltage drop across the diodes.