"Blown" Teensy 3.6 with external voltage derived from LM3671 3.3V Boost converter...

Status
Not open for further replies.

infovore

New member
"Blown" Teensy 3.6 with external voltage derived from LM3671 3.3V Boost converter...

I've got a custom-circuit board for a project with a Teensy 3.6 at its heart. It's powered from a 3.7V Lipo battery.

Previous iterations of the board fed the 3.7V into the VIN pin, and let the Teensy regulate its own voltage. However, to avoid voltage droop, I redesigned it with an LM3671 3.3v boost converter coming from the battery's 3.7V VCC pin, so as to supply a consistent 3.3V until the battery could no longer supply anything.

Here's my schematic for the boost converter -

Screen Shot 2021-06-25 at 14.22.24.pngScreen Shot 2021-06-25 at 14.22.24.png

- and it connects to the 3.3v VCC pins on the Teensy.

(The Teensy has no connections whatsoever to its VIN - there's an external USB socket connected up to the battery charger, and the D+/D- lines broken out to the Teensy, but there's not a backflow through the Teensy's own USB port, because it's not used at all.)

I saw a brief glimmer of a working circuit, but then the prototype ceased functioning. A continuity tester tells me there's now continuity between VCC and GND (on the Teensy 3.6), and about 100mv pd is seen between VCC and GND. Spinning up another board with no teensy attached and I see 3.3V from the converter with a battery connected.

I'm slightly lost as to what could be causing this the moment a Teensy's attached - my gut feel is an onrush of current - but no prior thread on directly powering VCC led to the outcome I was looking for. One answer might be "the current draw from the rest of the circuit that's on that VCC line is higher than the Teensy would like" (there's a low-power TPA2012 speaker amplifier, for instance) - but I'm now deferring to more experienced EEs.
 
Is the current handling of the inductor high enough? If it saturates its bad news. Did you go with
one of the ones listed in the datasheet? And follow the datasheet example layout?

Its not a boost converter.
 
Is the current handling of the inductor high enough? If it saturates its bad news. Did you go with
one of the ones listed in the datasheet? And follow the datasheet example layout?

Its not a boost converter.

The inductor part I used is 2.2uH, with a 1550-1600ma rated current, which is inline with what the datasheet suggests - "A more conservative and recommended approach is to choose an inductor that has a saturation current rating greater than the maximum current limit of 1150 mA."

I'll be honest and say no, my layout isn't quite the layout suggested within the datasheet, in particular with regards to ground planes. Whilst I'd expect this to have an impact on eg, noise, would it also have an effect on the saturation of the inductor? (I'm using similar positioning, fat traces, etc).

And yes, you're right, my bad: it's a DC-DC converter.
 
Everything about a 2MHz converter is crucial, capacitors, inductors, layout - I suggest just use the parts and layout
from the datasheet verbatim. This chip generates very fast edges (< 10ns) with high currents, so can easily induce
large voltages into nearby traces if the layout is poor.

The suggested components are particular sizes - this affects the layout and the self-resonant frequencies of the
parts and so you should use that size of component (as close a match to one of those listed as possible in all
parameters).
 
Status
Not open for further replies.
Back
Top