Back-up battery charging circuit from IFI

After a fun weekend with the back-up batteries, we were told about the back-up battery charger on the ifi website that charges the back-up battery from the robot’s main battery.

Has anybody made this circuit and if so, what are your thoughts, suggestions, etc. We want to make one in the fix-it-window before the Detroit Regional, but any lessons learned, advice, or whatever is appreciated.



We learned that the 8.2V zener is absolutely necessary. :slight_smile:

Luckily we tested the output with a multimeter before putting it on the 'bot… :rolleyes:

The backup battery should be charged initially at bot startup or your camera will have problems.

The backup battery should be charged initially at bot startup or your camera will have problems.

It will be. But we just want it so that the back-up is actually what it is called, a back-up battery for powering the RC when the main battery dies.

We just want the charging circuit so that the camera servos and other servos don’t drain our back-up batteries during testing and a match.

We used it, the only problem we had was a slight trickle through of power to the RC when the breaker was open and the backup battery connected, it wasn’t even really a problem, somewhat amusing is all. I don’t think we ever had to remove our backup battery and everything worked great for us, it gets a thumbs up from me.

Anybody else?

The circuit looks a little “interesting” to myself and some co-workers, so I’m looking for some experiences of it working.

We are probably going to try it anyway…

At my school we used the Vex backup batteries, which are more powerful, and last longer. Also they can be quick charged using a standard RC car charger. We had three we rotated, and it worked perfectly :D.

I’ve got one that I wired up, but in the haste to ship it never got onto the bot. It seems to work fine in bench tests.

My team had the same thing happen. We also discovered that if you push the robot for a few yards while the breaker is tripped and even when the Anderson connectors are unplugged, the same thing happens.

The reason that happens is because the motors are acting as generators, powering your breaker… I wonder,

If you push your robot too fast, could you burn anything out?

One recommendation I would make after seeing the schematic is to mount the LM317T (regulator) to a heat sink or piece of copper. This is the unit handling the power dissipation and it may get too hot.

And of course you could blow something by pushing your bot. Voltage is voltage, and current is current (i wont say resistance is resistance because negative resistance gets tricky) But, the motors would have to spin faster than they do with the power on.

We originally thought that might be the case, but after testing the completed circuit it turned out heat dissipation isn’t a problem at all. A plastic housing will work fine.

Note that the tab of the LM317T is electrically hot, meaning it has output voltage (and not ground!) on it. It needs to be electrically isolated from everything else.

We built the circuit, and saw the same issues as others: The backup battery ‘powering’ the bot, making for some anxiety but no real issues. We used it for testing, but took it off for competition.

Anyone wants a PC Board etching pattern, PM me.


My team wanted to reduce the amount of extra electrical components on the robot, so we used the Vex batteries which are also 7.2 Nickel Cad, which are allowed under rule R68 (IIRC), which allows any replacement backup battery. You get longer runs with them, and charging them is the same as charging RC batteries, so they can be quick charged in 15 minutes, so with two there should be almost no downtime.

They do add a bit more weight, but it is negligible compared to what you get in return, which is extra power for the camera and servo’s as well as anything else that gets it’s power from the RC through the backup battery.

Edit: Seems I had already told my story, sorry for the double posting :o

There are two problems with the schematic from IFI. The first one is that if the 8.7V zenner comes off or breaks, you risk supplying a full 12V to your RC and backup battery.
ANother problem is that of the main battery is dosconnected, the backup battery will end up “replacing” the main battery. It will supply power to the motors through the 43 ohm resistor. At 7.2 V, the power dissipation inside that resistor is about 1.20W. You will end up frying the resistor.
My recommandation is that you add a diode(any diode that can take over 1A will work)before the output.Disconnect the “+” ouptut, and connect the anode of the diode to it. Connect the cathode to the wire you just disconnected. Or just build a new circuit with the extra diode. This will prevent power from the backup leaking back if the main power drops too much, or the main battery is disconnected. However, this modification will not protect the backup if the 8.7V zenner fails.

We have determined that there are too many things wrong with the circuit as given from IFI to use on our robot at competition. We have a modified circuit that works as it should with all the proper protection, but since it is not the IFI circuit, we won’t be using it in Atlanta. It will just be used on our practice field here…

You are completely right. The LM317T is set to provide 8.26V. D1 has a forward voltage of at least 0.66V, That means it will only conduct when the voltage is 8.92V, but the zenner diode will limit that to 8.7V. The entire LM317T part of the circuit does nothing.

The way the LM317T is used in the schematic, it should pull down the voltage on the output, but it has to do that through the 1.24KOhm resistor. It wasn’t designed for that.

The only way to use this circuit is to exclude R1, R4,D1, and D2, and connect the “+” output to VOUT. This should provide approximately 8.26V to the backup battery.

Actually, better would be to put the diode at the 12 volt input. The diode exhibits an approximately 0.7 volt drop across the PN junction, which will reduce the circuit’s output voltage if put in the output. The input is sufficiently high that the diode drop will have no impact.

On the other hand, I agree with Don Wright and others, I just do not like the circuit, and although we used it during testing, it never went into the crate.


I’m not sure if this statement is correct. I’m going to attempt to describe what I believe is going on and the CD community can let me know if I’m off in left field. :slight_smile:

When the backup battery is drained (< 8.3 volts) the very low impedance of the battery will pull the output voltage down (and the 8.7v zener is not active). At this time the lowest impedance path would be through the 22 ohm resistor from the 8.3 volt regulated supply. This will charge the battery (and supply the RC).

Once the battery is charged to a voltage higher than 8.3 volts the voltage potential will no longer draw from the regulated supply and then current will flow from the 12 volt source with the current limited by the 43 ohm resistor.

This should create a staged charger for the backup battery as well as supply the RC.

Edit: I would like to see IFI add some protection to prevent any back feeding from the backup battery when the 12 volt supply is disconnected. (Or just diode OR the 12 volt and battery sources internally to the RC).

We used this circuit (with minor modifications) in our robot, and never once had a voltage-related issue with the camera, nor did we change the backup battery. It was a nice reprieve from previous years. The circuit does look a bit… odd… and pieces seem to be redundant, but there is a reason for that: FIRST likes redundancy a lot. If the programmable regulator ever blows, or if there is a current spike, or a voltage spike, or any one of thousands of possibilities, the circuit won’t cause a catastrophic failure. In short, the circuit works as advertised, and I’ll be advocating its return for next year.
Oh, and no heat problems surfaced here either. We did, however, place a rather large resistor in the line between the charger and the circuit breaker - the darn backup battery kept the Vics on with no 12v source. Having a resistor also kept the circuit from blowing itself out in case of a heat-related issue.
That’s all. It’s got my vote.