Hello,
I have a problem with one of my PCBs.
The project:
Solar panel cleaning robot.
The robot is powered by two BLDC motors.
One can draw up to 5 amps and the other up to 15 in normal operating conditions.
The system is powered by a LiFePO4 battery with about 17V nominal value.
There is also an MPPT charger on the board to charge the battery from a solar panel when the robot is docking.

The robot used to work very well and several thousands from it were produced and deployed.
We recently started working in India with much longer routs for the robot to drive.

We started experiencing burnt PCBs.
95% of the time it manifested itself in the circuitry of the charger, however it always happens during the operation.

After investigation we have found out that due to the longer routes, the battery is depleted and the protection circuitry (BMS) of the battery shuts it off.
This results in an immediate loss of power to the circuit and a brutal stop of the motors.
Moreover, since that the battery in this situation is effectively not connected to the circuit anymore, there is no pass for the current to return to as in normal operation.

I would like to add some dissipative circuitry to protect the PCB but have little experience with such circuits.
I have read about RC and RCD snubbers but not sure where to connect it and how to design the.
Do I need to connect it across each MOSFET in the bridges or one for the entire circuit (such as on the battery + and - inputs) is enough?

Attached below are the relevant parts of the schematics.

Any help is appreciated.

 

I suspect that all you need is a pair of diodes connected like D11 and D12 in the schematic below connected to each winding.

The diodes must be capable of handling surge currents the same as the maximum current to each motor.
In some circuits high speed diodes may be required.

EDIT: You need to assure that there is sufficient capacitance to ground from the power supply to assure that when the current is dumped when the drive current is shut off the power supply voltage will not change much. I know this sounds subjejctive, but there aren't many details available at the moment. V- appears to be your ground.

Note: It looks like you Zener diode and TL431 are backwards.

 

There is much more to this problem than we are being told, and so while some will venture a guess, not much help can be provided.
In addition, a battery undervoltage shutdown is probably much different from a disconnected condition, at least it should be.
And if some portion of the circuit leads to " This results in an immediate loss of power to the circuit and a brutal stop of the motors", there is a serious design flaw present. That may be in a part of the system designed to fast-stop the motor so that fingers will not be injured reaching into a moving part

 

We started experiencing burnt PCBs.
95% of the time it manifested itself in the circuitry of the charger,

That's rather vague. Can you specify which particular components in the attached pdf have been found to fail?

 

I suspect that all you need is a pair of diodes connected like D11 and D12 in the schematic below connected to each winding.

The diodes must be capable of handling surge currents the same as the maximum current to each motor.
In some circuits high speed diodes may be required.
View attachment 295156
EDIT: You need to assure that there is sufficient capacitance to ground from the power supply to assure that when the current is dumped when the drive current is shut off the power supply voltage will not change much. I know this sounds subjejctive, but there aren't many details available at the moment. V- appears to be your ground.

Note: It looks like you Zener diode and TL431 are backwards.

Thanks, In my design I use MOSFETs so the diode is already there.

 

One of them is. Which part failed?

 

There is much more to this problem than we are being told, and so while some will venture a guess, not much help can be provided.
In addition, a battery undervoltage shutdown is probably much different from a disconnected condition, at least it should be.
And if some portion of the circuit leads to " This results in an immediate loss of power to the circuit and a brutal stop of the motors", there is a serious design flaw present. That may be in a part of the system designed to fast-stop the motor so that fingers will not be injured reaching into a moving part

Thanks Bill,
All the battery protection mechanisms (Undervoltage, thermal protection, over-current) lead to the same result.
It shows 2V, it is not enough to open the MOSFETs at the entrance so the circuit is effectively disconnected from the supply.
There is a bulk capacitor that holds a charge, but when the motors are running it will deplete very fast.
We have mechanisms to sense the current of the motors by the MCU so if something goes wrong, we stop the motor.
I am talking about situations that we do not sense, or protection if the software fails.
For example, if a motor will stall, we will have a massive rise in current, we handle that, but sometimes we miss it and the battery shuts down.

 

That's rather vague. Can you specify which particular components in the attached pdf have been found to fail?

Before we added the D21, the parts in the MPPT area would fail, mainly U4 and C66,C68.
After the addition of D21, we would see other parts fails as the MPPT was protected Such as U7 (or the STM32 which is not in the PDF).

 

It seems now that we are in a discussion of product design, more than repairs to an individual device.

So I am suggesting a review of exactly what the requirements are as far as protection from common conditions such as battery charge depletion, motor stalling, and obstacle contact. User protection must also be considered, given that a small potion of users are incredibly stupid. That is the harsh reality that products face today. It seems that current legal processes favor those with no willingness to avoid obvious hazards at all.
In addition, battery charging must be carefully regulated to avoid all incorrect conditions, including overheating and overcharging. Thus every aspect of product use must be considered once again. This includes the correct shut-down sequence.
OR, it may be that the TS is considering a more complete revision of the product control scheme only for their personal copy of the product.

 

Before we added the D21, the parts in the MPPT area would fail, mainly U4 and C66,C68.
After the addition of D21, we would see other parts fails as the MPPT was protected Such as U7

Perhaps you could add a Charge/Run selector switch to isolate the MPPT when not charging?
As for a snubber, would a simple resistor across each motor winding suffice?

 

Is this a discussion of product design, or is this an individual working to solve the problems in a one-off home repair situation???

 

Is this a discussion of product design, or is this an individual working to solve the problems in a one-off home repair situation???

You are right.
This is a discussion of product design.

 

Perhaps you could add a Charge/Run selector switch to isolate the MPPT when not charging?
As for a snubber, would a simple resistor across each motor winding suffice?

Thanks Alec,
Even if I will isolate the MPPT, something else will fail as the charge has nowhere to go when the battery is disconnected.
As for a resistor across each winding - you mean on each Half-bridge?

 

I mean at or close to the motor coil terminals, so that each inductor is bridged by a resistor to provide a permanent path for current circulation, albeit a lossy path for either current direction (unless a diode is included in the path).

 

From the descriptions of operation it seems that there are some functional changes required. This includes how the power is controlled when the battery charge level indicates that charging must be done.
Switching off the high power blade mower must not be a "brutal stop", but rather a controlled shutdown, as the motor certainly will act as a generator while it is spinning towards stop. Adding a connected shunt resistor will certainly be waste of battery charge and so is a poor choice. But an orderly motor switching off sequence is needed.
Next, dealing with battery charging, which may happen only at a docking charge station, although that is not presented at all But the issue of dealing with the charging power when not charging is presented. Why this is an issue is unclear. Is the charging system a part of the mower, or is it separate? We are given no hint as to which it is.

So as an engineer the first question would be: "what has changed?" Is the production in a different location with a different design? Has the purchasing department changed the source of some components? Or even selected a lower cost component? Is the battery management system been altered? Or replaced with a lower cost product? And if the battery protection system is unable to provide an advance low battery signal prior to shutdown, then it is unsuitable for the application.
It appears that there is a major flaw in the motor drive circuitry if it is unable to survive a loss of power without allowing for the power generated by the motor coasting to a stop.

In summary, the fist step will be to understand just what the present system actually does when the power is suddenly removed, and compare that to what should be the actions when the system power is lost. The second step will be to create a sequence of actions that will result in an acceptable sequence of actions when a low battery charge condition is sensed.
The initial evaluation is that no provision has been provided for the condition of battery system shutdown, and that the battery protection system is an inadequate design, and that the motor drive system is not able to correctly respond to a loss of supply power in a satisfactory manner.

 

I mean at or close to the motor coil terminals, so that each inductor is bridged by a resistor to provide a permanent path for current circulation, albeit a lossy path for either current direction (unless a diode is included in the path).

Wouldn't this require a change in the motor itself?
This is not something that can be done.

 

From the descriptions of operation it seems that there are some functional changes required. This includes how the power is controlled when the battery charge level indicates that charging must be done.
Switching off the high power blade mower must not be a "brutal stop", but rather a controlled shutdown, as the motor certainly will act as a generator while it is spinning towards stop. Adding a connected shunt resistor will certainly be waste of battery charge and so is a poor choice. But an orderly motor switching off sequence is needed.
Next, dealing with battery charging, which may happen only at a docking charge station, although that is not presented at all But the issue of dealing with the charging power when not charging is presented. Why this is an issue is unclear. Is the charging system a part of the mower, or is it separate? We are given no hint as to which it is.

So as an engineer the first question would be: "what has changed?" Is the production in a different location with a different design? Has the purchasing department changed the source of some components? Or even selected a lower cost component? Is the battery management system been altered? Or replaced with a lower cost product? And if the battery protection system is unable to provide an advance low battery signal prior to shutdown, then it is unsuitable for the application.
It appears that there is a major flaw in the motor drive circuitry if it is unable to survive a loss of power without allowing for the power generated by the motor coasting to a stop.

In summary, the fist step will be to understand just what the present system actually does when the power is suddenly removed, and compare that to what should be the actions when the system power is lost. The second step will be to create a sequence of actions that will result in an acceptable sequence of actions when a low battery charge condition is sensed.
The initial evaluation is that no provision has been provided for the condition of battery system shutdown, and that the battery protection system is an inadequate design, and that the motor drive system is not able to correctly respond to a loss of supply power in a satisfactory manner.

Hey Bill,
Thank you for the elaborate answer.
I will do my best to clarify the issue you have raised.
First, this is not a mower but a robot for cleaning solar panels but the idea may be the same.
There is a sequence for charging and stopping the battery charge - it is controlled by the MCU.
The charger is not working while we operate the robot but it is still part of the board.
There are also mechanisms to start and stop the motors, control the speed, identify over-current situations, stalling and to monitor the battery charge state.

The problem that I describe can happen so fast that the MCU cannot respond to that and the battery will shut down.
You asked what has changed, mainly the environment and the length of the operation.

When a shut down occurs, the motor will coast, but if it was stalled or even working in full load, it will not coast but rather stop.
In any case it will act as a generator.

I am looking for a way to respond to that, I was thinking of adding a resistor + diode on the (+) (-) of the board so the current will have a path to go through.

If you have other or better ideas I will be happy to hear them.

 

Wouldn't this require a change in the motor itself?

Not unless you have a DC-fed motor with internal commutation.
Do you have a link to the motor spec/datasheet?

Edit:
What provides the input to VIN_SUPPLY? It might be worth adding extra capacitance to that pin. At the moment the total for the various parts in the pdf which rely on that voltage seems to be about 30uF. I would have expected at least a magnitude greater.

 

OK, and thanks for the detailed response. I realized after commenting that it was a cleaner system, not a mower system. So the application is a bit different in that the path is well defined.
So it seems that the problem is that the battery protection system makes an instant decision rather than first delivering a notification of low charge. And the motor, acting as a generator, delivers power when it should not be.
If the process controller were notified of a low charge condition then it could conduct a safe transition through an orderly shutdown and transition to charging mode. I am guessing that this would prevent the damaging failures.
So the question is, how to know that the battery is in need of recharging soon? If the processor is already monitoring the battery then it could simply involve the addition of some code.

 

OK, and thanks for the detailed response. I realized after commenting that it was a cleaner system, not a mower system. So the application is a bit different in that the path is well defined.
So it seems that the problem is that the battery protection system makes an instant decision rather than first delivering a notification of low charge. And the motor, acting as a generator, delivers power when it should not be.
If the process controller were notified of a low charge condition then it could conduct a safe transition through an orderly shutdown and transition to charging mode. I am guessing that this would prevent the damaging failures.
So the question is, how to know that the battery is in need of recharging soon? If the processor is already monitoring the battery then it could simply involve the addition of some code.

Thanks Bill,
The battery can go to "protection mode" for many reasons, such as over current and temperature so adding code may not help.
I would like to solve this with hardware only.
That is why I was thinking on a snubbed.
What do you think?