Reverse Protection

Key Points

1. The mosfet output stage of an eCom presents a near short circuit across the lane if a car is inadvertently placed or ends up backwards on a live lane, even if the eCom does not fail.
2. There is a high chance that the eCom will fail and produce a more severe short circuit across the lane.
3. A short circuit (or near short circuit) across the lane is quite likely to cause a controller failure, or the power supply to shutoff power.
4. If the car ends up backwards in another lane it's likely that another driver's controller fails.
5. A simple error or accident could easily result in the failure of both an eCom and a controller.

Additional details

Mosfets inherently have an anti-parallel diode as part of their structure, so if reverse polarity is applied the diode conducts. In an eCom there are three high-side mosfets and three low-side mosfets so under reverse polarity conditions there are three parallel diode branches that will conduct current. This effectively looks like a near short circuit across the lane.

The current is limited by factors such as track power supply, track wiring, controller trigger position, track and car braids etc. If the controller trigger is near its maximum end a high current can flow and the survival time of the controller's transistor is short, about 1ms if the current is 120A.

Under reverse polarity conditions at 120A the eCom mosfets survive for less than 1ms. The usual failure mode for the mosfets is short circuit, so the eCom goes from a near short circuit to a full short circuit.

A real life example of track prospective fault current: at our local UK club, the power supply can deliver 180A and even allowing for wiring and track braid a fault current of 120A would be expected.

The common ATOF automotive fuse used in controllers is intended to protect the controller wiring not the transistor, a 20A fuse would take more than 55ms to clear at 120A i.e. 55x longer than the survival time of the controller's transistor.

To have a chance of protecting the controller's transistor during a short circuit a 3A fuse would be needed, it would clear in about 0.7ms at 120A i.e. just less than the survival time of the controller's transistor. But, 3A is an impractical value for normal racing.

Whilst we might be willing to accept any potential damage to our own car being if it ends up backwards on our own lane, it is surely not acceptable that it could also cause problems for other drivers if your car were to flip into another lane backwards.

During a recent race we saw a car flip into another lane backwards 3 times.

A Scrutineering Proposal

A car flipping into another lane backwards means that the reverse polarity protection circuit has to rapidly switch between normal operation to reverse blocking, this is highly likely to be its most difficult mode of operation.

It should be part of the scrutineering process to prove reverse polarity protection is working, by:

• Simulating a car flipping into another lane backwards use a reversing switch to toggle between normal supply polarity and reverse polarity.
• Applying a substantial forward voltage (as close to 12V as is deemed safe), toggle the reversing switch to very quickly apply a substantial reverse voltage (perhaps 12V).

As an automotive 3A ATOF is slightly weaker than the transistor used in many controllers, a fuse put in the line can be used to prove compliance, if the fuse blows it means a failure.

Therefore if a 3A fuse were to blow during a scrutineering test we would know that a controller would be likely to fail if a car flipped into another lane backwards.

To be realistic a power supply with at least 10A capability should be used for applying the reverse polarity.