Drill motor improvement to avoid component wear when shut off

Question

Story

Get my hands on two Metabo (made in Germany) cordless DC battery powered hand drills previously thrown away because of..... the batteries, how sad. The overall construction is simple however pretty stiff and clever designed with quality components.... a keeper! "The Post Apocalyptic Inventor" aka TPAI made a video about making obsolete tools useful again. The solution is simple, use an external battery or PSU to power the tool by installing a connector. I did it a little different, used a different and more regular connector and install the connector on a different location. The result is actually equal to the TPAI version.

---

Problem

At the first try, I notice that after starting the motor (huge inrush current anyway) it stops immediately (with some reverse turn shake feedback). Through the ventholes of the motor I noticed some huge sparks when the motor stops. This abrupt "kick out" behaviour is not very healthy for the switches, bearings, the gears and the brushes, possibly effecting the lifetime of the device. I have notice similar behaviour at other brands, only at high speeds and release the (speed controlled) trigger quickly. These ones has no electronic speed control, it is just on or off in selected direction.

When disconnect the power source (remove it), this problem does not occur and the motor slows down by friction and power loss.

After figuring out how it works (very clever designed in matter of simplicity of construction) I understand why.

The motor is shorted (in feedback/generator state) when releasing the power trigger, because of the clever configuration of the two 2-state power on slash direction switches in parallel. However this results in a huge feedback current and build up coil voltage inside the motor. That is the reason why the brushes spark intens and the motor stops immediately after high speed rotations. To avoid feedback to the batteries this is good but can have some negative impact on other components because of the sparks and internal feedback force.

---

Question

1. I have tried to design a new addon solution (mod) in the existing configuration. Cut the negative lead and added two diodes and a resistor. I am not a professional so I wonder if this is a plausible solution to slow down the decrease of speed and limit the feedback current/voltage inside the circuit. What do you think, it this okay?
2. If this solution is plausible, why manufacturers do not apply this even at quality tools? To avoid the best quality? Wear is a selling point?

Answer 1

^{simulate this circuit – Schematic created using CircuitLab}

Figure 1. A pair of diodes can snub the motor when running in either direction.

The diodes will conduct only during transition of SW1 or SW2.

---

^{simulate this circuit}

Figure 2. Simulation circuit.

Figure 3. Simulation results. Top: voltage at Node 1. Bottom: current traces.

1. SW4 (representing the NO contact) opens at 0.1 s. Voltage rapidly swings negative.
2. SW5 (representing the NC contact) closes at 0.2 s.
3. D3 keeps the current running through the motor.
4. The current through the NC contact decays from a low initial value to zero.

Answer 2

One possible solution would be to replace the diodes in your idea with NTC thermistor current surge devices. You can get them rated at 25 amps with an initial resistance of 2 ohms and resistance of 0.02 ohms at rated current. It would cause a delay before maximum current would be available. When the switch is turned off, the motor would see the 2 ohms of the cold thermistor, so it would experience a relatively slow braking at about 6 amps initially for a 12V battery.

Answer 3

A inexpensive and relative simple solution for dynamic braking would be to place a low value in the braking circuit maybe in the 2 ohm range. This will increase the reliability as there is no semiconductors involved. When you release the switch the output voltage of the motor will rise very rapidly. initially the resistor will conduct a lot of current, as the pulse rapidly decays the current will reduce to zero. You can empirically determine what resistor you will need. As it is a pulse load, not steady state You can use a smaller resistor.

Answer 4

I like the idea with the two diodes and the break resistor. In this circuit I try to avoid the losses across the diodes using MOSFETs. This can only work, if at no time both switches are activated, but I assume this is impossible in this machine.

^{simulate this circuit – Schematic created using CircuitLab}

1. Activating one direction switch will enable the MOSFET on the opposite side. The low pass filters R4/C1 and R5/C2 should protect the gates from short voltage spikes, where ever they might come from.
2. Releasing e.g. SW1 will activate the break resistor. There will be a short negative back EMF, that is clamped by the body diode of MOSFET M1 and the diode D2 can quickly tun off the negative path via M2.
3. I added the diodes to be sure that the negative path is no longer conducting if the opposite switch is activated while the motor is still rotating. Probably not needed.
4. C3 and C4 together with R1 form a snubber network to protect the contacts, this is more a question of EMI style.

I did not test this circuit, so there may be aspects I'm currently not aware of.

This sure would make the machine more expensive, I can understand why this is not implemented regulary.