Flipsky FSESC 4.12 won't power on

I have 2 Flipsky FSESCs like the one seen in this picture . A couple of days ago, one of them got disconnected during a ride and after re-connecting it to power, it won’t power on. No LED lights, nothing. I probed the 5V rail and I’m getting 5.1V, but when I probe the 3.3V rails, I’m getting 0V. The fact that I’m seeing 5V leads me to believe that the DRV8302 chip is working as it should be by stepping down my 6S pack(25.2V) voltage to 5V. The VREG (579-TC2117-3.3VDBTR) on the back of the FSESC, however, is heating up and I’m pretty sure it’s dead.

I assume that once I replace the VREG, it should power right up and things will work as expected. But, in the case that it doesn’t, I’m wondering if I should be keeping an eye out for any other components that could be causing the issue. Any help or insight is greatly appreciated.

Bonus question: Does anyone know why the act of an ESC disconnecting mid ride would have caused that?

4 Likes

Update: I learned that if a VESC based ESC is disconnected from the battery during a ride, and you attempt to brake, regenerative braking will kill the ESC because the power has nowhere to go.

I began buzzing out the FSESC with a multimeter and realized there’s a short between the 3.3V rail and GND on the entire FSESC. I desoldered the VREG and the short remains. I’m waiting on a hot air reflow station to arrive in order to desolder the MCU. Once the MCU is removed, I’ll check on the short and report.

2 Likes

This is spooky, I had the exact same thing happen today to the exact same FSESC, one of my bullet connectors fell out during a ride. I’m seeing the same as you report about the 5V and 3.3V rails however my 3.3V reads 0.2V, not sure how relevant that is.

One key difference is that both of my FSESCs are bricked, I have them set up over CAN with traction control turned on. I’m thinking this could be why both have been affected when I braked, how were yours set up?

I’m just considering buying replacements and holding onto the the dead ones to repair sometime in the future. If you can post updates on how you get on with your fix it would be greatly appreciated, my experience of attempting surface mount soldering in the past has been just a bit scarring…

Good luck!

UPDATE 2: I removed the 3.3VREG, but the short still exists between 3.3V and GND :man_facepalming: The next logical step (I think) is to remove the MCU with a hot air reflow station and check to see if the short is gone.

Noooo! I’m so sorry to hear that this happened to you too. It’s a bummer that this type of thing happens. I’m so torn on whether or not to buy higher end ESCs or if I should stick with these because they’re so cheap!

I was also seeing 0.2V when the voltage regulator was on the FSESC. I think this is somewhat expected, but not totally sure yet.

Mine were set up the same way minus the traction control. I’ve read that if the CAN BUS connector becomes disconnected and the FSESC is powered on, it can brick. It sounds to me like yours bricked when one FSESC got disconnected. I’m hoping it’s that and not a huge current influx when you were braking. Do you have an ST-LINK V2 programmer to try to fix them?

Good luck guys, did not thought vesc could be killed that way.

1 Like

Update 3: I got my reflow station in and removed the MCU, but the short between 3.3V and GND is still there.

At this point, I think I’m going to try using a benchtop power supply to feed the FSESC 5V @0.25A to see if I can measure a voltage drop across one of the passive components. (resistor, diode, capacitor etc.) This is my first time using this method to hunt down the root cause of the short.

If anyone has any tips or tricks, they’re greatly appreciated. I have a basic multimeter and craptacular DSO138 oscilloscope that I can put to use.

I had a similar issue: my FSESC 4.20 died during a ride but the 5V rail is shorted to ground instead of 3.3. I found another thread with someone who had a dead diode on the 5V switching regulator, my diode was reading as a short in both directions (though this could have been caused by something else), but after some reflowing the 5V was still shorted to ground. I also killed 2 of the replacement diodes I tried to install by overheating them, and lost a damn tiny SMD capacitor by desoldering it, so I’m at worse than square one. Had just ordered a unity anyway so I’m ok.

I hadn’t thought about the issue of disconnecting during a ride though, that could easily have been the cause of my death because I had a janky setup with the wires taped and near my foot.

Your logic with the bench supply sounds good!

1 Like

Update 4: Good riddance, I finally found the short!

.
The CANBUS chip (SN65HVD232DR) let the smoke out and failed short. If you look at the picture, you’ll notice that there’s a tiny spec on the “3” in the 232 on the IC. That is where it let the smoke out. It never raised any red flags during my visual inspection, but running my fingernail over it to try to remove it, let me know it was raised up and felt like a bump on the chip.

Here’s what I learned:

  1. There are only a few ICs on the FSESC or any VESC based ESC that interact with the 3.3V power rails: (STM32F4 MCU, 3.3VREG, the CANBUS Transceiver)
  2. ICs are likely the culprits, so check those first, and be sure to check for tiny specs where they may have let the smoke out. The next likely suspects would be any semiconductors like diodes. Lastly, check passive components like resistors, capacitors, etc.
  3. I knew very little about troubleshooting or fixing a circuit before I started, but I feel comfortable diving in head first now. There are only so many failure points, you’re bound to find the problem component.

My work isn’t done yet. I still need to make sure none of diodes, caps, or resistors are dead and failed open instead of having failed short. Most importantly, I need to solder a new MCU, CANBUS Transceiver, and 3.3VREG back to the circuit and pray that I’m able to flash a bootloader to the new blank MCU chip.