I have a device I'm developing that interfaces with an existing consumer device. I can't change the design of the existing device and it has a poor design with how a battery is connected to it. It doesn't come with a battery so they leave it to the customer to source. A very common and good choice is a 2 cell LiPO battery. Unfortunately, the input connector of the device is identical to the common 2 cell LiPO balance cable, but wired completely different. People see the same connector, plug it in and poof.

So I would like to make a tiny board that will plug into the device's socket and provide an identical socket for the user to plug their battery in to. The goal being that it would not allow current through (and will not be damaged itself) in the event it is wrongly connected.

Ideally the tiny board would have a green and red LED to indicate if the battery is in right.

On the original device, we're dealing with a 3-position socket with the middle being +ve and the two outer are -ve (joined together). The battery then is intended to use a 3 position connector where only the middle and one outer are used, but if a 2 cell balance connector is used there would be 3 wires (one at +7.2V, one at +3.6V and the other at 0V). So I guess that's 12 possible combinations mathematically but 5 would never happen (due to how balance plugs are designed). That leaves 7 combinations, 2 of which are correct. I need to protect against the 5 that aren't.

Below I have the possible/likely combinations. G is 0V, H is a positive voltage. The last one has L, which is in the event this is a balance plug (which should not be used but customers can be ignorant and/or careless) and is also a positive voltage, half of H. X is not connected.

G H X - Good connection
X H G - Good connection
H X G - Battery short
G X H - Battery short
H G X - Reverse Polarity
X G H - Reverse Polarity
G L H - I don't even know what to call this. Battery is shorted and device is powered?

I really don't know how to go about this. My experience is pretty much limited to circuits where it is a given that power flows the way it is supposed to. I get how to protect against reverse polarity. I'm fuzzier on short circuit protection. I really have no idea how to handle that last event.

Thanks.

 

Do you have a schematic or even photos of the device?

 

I assume that you can't use a couple of series diodes because of the voltage drop and you can't use parallel diodes because you don't want to include any fuses/overcurrent protection?

 

A reverse-voltage protection circuit using a MOSFET for low-voltage drop in normal operation is shown here.

Below is the simulation of a low-drop current limit circuit.

The plot shows load current and voltage drop from the input to the output of the circuit versus load resistance.

Its limit current is about 0.65V / R3.
The selected value of 0.65Ω ohms thus gives the shown current limit value of ≈1A.

The circuit voltage drop under normal operating conditions is determined by the voltage drop across R3 plus the selected MOSFET fully-on (Rds) resistance.

The P-MOSFET for both the reverse-voltage and the current-limit circuits can be any logic-level type that fully turns on at 3.3V Vgs (not the Vthres value) and is rated for at least double the maximum designed current limit value.

For both reverse voltage and short circuit protection you would put both these circuits in series (reverse protection first) between the battery and the load.

 

I got lost in your description, but could this http://www.linear.com/product/LTC4365 help?

A reminder that you can use a bridge to always get a positive voltage from a badly connected battery, but you lose 1.2V


## - + -
##1 2 3
0: G H X - Good connection
1: X H G - Good connection
2: H X G - Battery short
3: G X H - Battery short
4: H G X - Reverse Polarity
5: X G H - Reverse Polarity
6: G L H - I don't even know what to call this. Battery is shorted and device is powered?



I used some tricks to do that. But I assigned pin #'s and labels and the battery sign I could have used a table.

Put polyswitches in series with one or both of the incoming pins 1 and 3.

 

Potential problems are uncharged battery and if the system does battery charging and loss of battery life.

Another POTENTIAL idea is to look at pins 1 and 3 with another LTC part (LTC4365#2) and then look for aprox 1/2 the voltage. If it's present, inhibit the LTC4365#1.

I have no idea if this could work. Just thinking out loud.

 

The LT4365 doesn't have a current-limit function which the OP requested.

 

Sorry, I thought my question was ignored as I didn't get a notification there was any replies.

Thanks for the replies! I appreciate the help. Actually, I realized over the weekend that I was over-complicating things. K.I.S.S. Between voltage drops and the multitude of possible configurations (on top of wiring possibilities, there is also a voltage range from 6V to 12.6V)... I realized a low-tech solution was ideal. I've come up with a tiny board that simply changes the physical connector, so they won't be tempted to plug that 2 cell balance plug in to it. They'll have to solder on a new connector to their battery, but the device is r/c hobby related and they're used to that kind of thing. And if they don't want to, well.. then don't use the board and just be careful. By including it I think at least it draws attention to the issue and hopefully less of them see the magic smoke.

 

FWIW: http://www.linear.com/product/LT4363

 

Thanks. Something like that would be far too expensive anyway. I want to protect the 1%, but not at a significant expense to the other 99%.

 

I want to protect the 1%, but not at a significant expense to the other 99%.

if only you ran our government.