1
\$\begingroup\$

I am using a DC1601B-C buck-boost converter in an automotive circuit to power some various electronics that I have added. It worked fine for a couple months, but under a less-than-normal load of 2.25A (when it should be able to output up to 20A) the input power pin caught fire.

In my circuit I also have a SLD8S17A TVS diode attached between the input power and ground to block the load dump. The diode was still functional after the fire so I don't believe the fire was caused by load dump.

In the DC1601B-C datasheet, the test instructions say not to "hot plug" the board. It also says to start power with the J1 jumper in the "OFF" position and switch it to the "ON" position after input power was applied. I had the board soldered into the circuit and left the jumper in the "ON" position, then applied power by turning a key switch with the load already attached.

I tried to research what hot plugging was and all I could find was hot plugging devices into a computer while the computer was still on, which I am not doing, so I am not sure what hot plug means in relation to a voltage converter.

My question is, was I "hot plugging" the converter by having it soldered into my circuit with the load attached before the power was turned on? And a follow up question is what else could have caused the input line to blow/what other protection circuitry am I missing?

As you can see in the photos below, the regulator caught fire only at the input power pin.

How can I prevent the DC1601 demo board from catching on fire?

enter image description here enter image description here

\$\endgroup\$

1 Answer 1

Reset to default
2
\$\begingroup\$

The problem is probably inrush current, and DC to DC converters can source a lot of current. On startup it could pull the upstream power supply down and that would cause it to source even more current. (like if the sourcing supply was lower than 12V, then the boost feature would kick in).

I've had DC DC converters burn up their inputs before, usually it was a transient, the electronics got pulled low briefly which caused the DC DC converter to source even more current. Too much current was pulled through copper and things burn up.

The way to avoid this is with soft starts and current limits (which should be used especially when DC DC converters are connected downstream from each other). The DC1601 has a current limit and a soft start, which should be used every time as the manual suggests.

Its also possible that the crimp could have played a role in this, as extra resistance can expend a lot of power. (10mΩ @ 20A would be 4W, 100mΩ would be 40W. 40W is a problem)

\$\endgroup\$
6
  • \$\begingroup\$ So for the DC1601 I am assuming the soft start is having the jumper 3 in the SSO position and the current limit is jumper 2 in the CCM position like the document suggests. Since I had these jumpers in the correct positions, is there something else I should be doing to avoid this issue? \$\endgroup\$
    – orange1234
    Aug 26, 2020 at 18:14
  • 1
    \$\begingroup\$ Making sure the load does not short circuit, which is possible. Sometimes electroinc loads on startup can get into a metastable condition where both transistors are on and there is a lot of current being drawn. Sometimes this condition can bring down the power supply upstream from an electronic load. Proper power supply sequencing is needed to ensure that electronic loads never reach this state. \$\endgroup\$
    – Voltage Spike ♦
    Aug 26, 2020 at 19:00
  • 1
    \$\begingroup\$ Think the Answer goes to the right direction... it was certainly more than just 2.25A and not just some ms of inrush current... the resistance of the crimp-connector/skrew-terminal is a "hot" candidate. We also can not put aside a possible manufacturing fault of the PCB. --> have you tried if the IC's still work? is one of them short circuit or now open load? \$\endgroup\$
    – schnedan
    Aug 26, 2020 at 19:02
  • \$\begingroup\$ @schnedan I have not tried to see if they still work since I worried about how extensive the damage appears to be to the PCB. As you both suggested I will see if there is a better solution than the crimping terminal to connect the wires. \$\endgroup\$
    – orange1234
    Aug 26, 2020 at 19:57
  • 1
    \$\begingroup\$ there is nothing wrong with a propper crimp connection. they are often prefered for high power stuff in automotive, industrial, rail,... and so on applications. But you need proper tools to do em right. also with skrew terminals you need some lockwasher (with spikes) to prevent it from getting loose \$\endgroup\$
    – schnedan
    Aug 26, 2020 at 20:09

Your Answer

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

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