0
\$\begingroup\$

I am currently building an Rc car which I designed myself. Yesterday I started testing the whole system, including four motors (one in each tire), 5 servos (MGN995, 4 ontop the suspention, one steering), and a small 9g servo for the rear spoiler. The motors are controlled by a speedybee 60A ESC, and a Arduino Mega is the brain of the whole thing. To supply power, I have a 8A step down converter, converting the 12V from the LiFe Batteries (2x 12V 20A discharge) to 7V for the whole system. Lastly, I have a Radiomaster r88 Receiver for controlling the whole thing, and a Adafruit BNO055 sensor to eliminate roll with the suspention servos.

When I throttled up the motors (750kv, 6.5A stall current) the receiver started giving weird numbers. It should put out a PWM signal from 1000 to 2000, but suddenly there was something like 13000 and 6, so I told the firmware on the Arduino to disregard any of theise bad numbers. Every time one of the ununsual numbers would appear, the motors would throttle down, the next receiver value would be valid again, then the motors would throttle up (to the commanded value), drawing to much of something, then confusing the receiver into giving bad numbers, motors throttle down etc. This sometimes would end in a oscillation (thottle up, throttle down, throttle up, etc). The only thing stopping this, was me cutting the power to the system.

After some testing and adjusting the firmware, the step down module made a loud bang, and something there "exploded" (there was something similar to an explosion) ontop of the capacitors. I immidiatly cut power. What the heck just happened there?

I ripped out the step down module, but after further examination there is visually nothing wrong with it. I replaced it with a fresh one, but after turning everything back on, there was just nothing (exept the power bank powered Arduino). I examined the receiver and servos individually, and it turns out that 5/6 servos died, and the receiver. However again, there was nothing visually wrong with any component.

Now I could go ahead and replace all this stuff, just for everything to explode again. What could be the cause of this?

One theory I have, is that the ESC had interduced lots of electrical noise, and the step down module did't like that. I read somewhere online that converters don't like when the input voltage fluctuates fast. Is this possibility?

I am a student in 13 grade, and I have 6 hours of Mecanical/Electrical engineering at school. Most of my knowledge from the internet. So there might be a super stupid answer that I just don't know.

Thanks for you help!

Edit: Currently the servos are not connected, since im changing them out right now

enter image description here enter image description here

enter image description here

\$\endgroup\$
4
  • 2
    \$\begingroup\$ Anyone will ask you for at least a sketch of the system, without which it is impossible to even hazard an answer. \$\endgroup\$ Commented Oct 4 at 8:37
  • \$\begingroup\$ A picture of your setup will help quite a bit. \$\endgroup\$ Commented Oct 4 at 8:47
  • 1
    \$\begingroup\$ aside: i would avoid using dupont connectors, or any other loose push-on connectors, with something that moves around. Those don't hold very strongly and i can imagine one popping off in the worst place and time, in a way that could be dangerous from a human on/or fire safety viewpoint. At least hotglue them all down before using... \$\endgroup\$
    – dandavis
    Commented Oct 4 at 18:20
  • \$\begingroup\$ @aside yes! This is only for prototyping. I want to switch to two Arduino nanos, and solder everything. Thank you for the tip! \$\endgroup\$
    – Vaynexx_
    Commented Oct 5 at 18:11

1 Answer 1

0
\$\begingroup\$

UPDATE: I have looked at your added pictures, thank you for putting them in.
As I suspected, you have many wires longer than necessary running all over, which besides voltage drops also induces unwanted currents and voltages in them, as well as noise, leading to unpredictable behavior and possible failure.
Remember this: short and neat wiring is not only nicer to look at, it is also better for the circuits and their proper and reliable functioning.
Bent or coiled wires will pick-up or radiate more unwanted currents and voltages than the straight wires.

A separate piece of advice regarding switching power supplies, especially buck converters:
because they (and typically their single switching transistor inside) are the only "line of defense" or separation between the destructively high input voltages for the typically sensitive lower-voltage circuits down the line from them, and because they frequently fail "short", a good practice is to use a linear voltage regulator after them which can withstand the full input voltage without failing, and thus protect the lower voltage circuits.
Buck converters are typically used to provide the minimum voltage needed for those linear regulators, while the linear regulators provide the stable and far less noisy voltage to the more sensitive circuits or components.
You could also use a fuse at the input or the output of a buck converter, combined with an SCR "crowbar" which would make a short and burn the fuse out if the buck output voltage goes above safe limits.

If the input voltage is above the absolute maximum that the 5V circuits can withstand (usually is) and the step-down converter gets shorted, it will most likely burn them out.
Large and fast fluctuations in input voltage can affect the step-down converter, but not if it has a large enough capacitor on its input AND output.
Regardless of a powerful battery supplying it, large loads can introduce significant voltage ripples, especially if there is a length of wire involved to produce significant voltage drops.
A way to isolate such drops is to place a diode in series with the step-down converter, and then a decently sized capacitor to smooth out any ripples at its input.

Additionally, due to large and fast current changes in your wires, it is possible you have a significant "ground loop" problem where voltage spikes get induced in your "ground" wires, so even if the +V supply is kept regulated, the problem "comes up" from the ground which is supposed to be zero, but it can go up to sometimes unpredictable values.

\$\endgroup\$
12
  • \$\begingroup\$ What do you mean by "and the step-down converter gets shorted"? You mean that 12V goes in the 5V circuit? \$\endgroup\$
    – Vaynexx_
    Commented Oct 4 at 9:05
  • 1
    \$\begingroup\$ @Vaynexx_ That would be a very possible failure mode for most buck converters. \$\endgroup\$
    – winny
    Commented Oct 4 at 9:10
  • \$\begingroup\$ Why do they get shorted? Too much current? I have looked at the buck converter with my oscilloscope and the chip is dead \$\endgroup\$
    – Vaynexx_
    Commented Oct 4 at 9:14
  • 1
    \$\begingroup\$ @Vaynexx_ The high side switch transistor can fail for many reasons: overvoltage, over current, too much power dissipation (temperature too high), manufacturing defect,... If it shorts out then the input voltage will run right trough it, the inductor becomes a piece of wire at DC, straight to the output... \$\endgroup\$ Commented Oct 4 at 9:28
  • 1
    \$\begingroup\$ @Unimportant uhh, ok, that makes sense. Thanks! \$\endgroup\$
    – Vaynexx_
    Commented Oct 4 at 9:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.