Just let the smoke out of two Maple r5's

LeafLabs Garden » Support » Maple Hardware Support

Just let the smoke out of two Maple r5's

(9 posts) (5 voices)

Started 3 years ago by tmbomber
Latest reply from tmbomber

tmbomber
Member

Ok, I've a bit of a problem...

I just smoked my second Maple board and I'm at a loss as to why. I had a uSD shield connected to the boards:

https://www.sparkfun.com/products/9802

The other hardware I had connected was: three LED's with 330 ohm resistors to three different pins (D3, D6, & D7, other end of the LED's tied to ground; a switch tied to D5 and a 15K ohm pull up resistor to 3.3V, other side of the switch tied to ground; and a 6N138 tied to D0, wired the same way as Sparkfun's MIDI input on their MIDI breakout board. Power was provided by a 7.5v wall transformer.

The second Maple failed connected to a separate set of hardware. (I have two uSD shields wired the same.

I don't think the button I added could have been the problem. I wasn't pressing the button when the second board died, and a 15K pullup to 3.3V should be ok going to a digital input pin.

Three LED through 330 ohm resistors from a digital output to ground shouldn't be a problem, either.

The uSD shield has a level shifter on it between the processor and uSD card. I cut pin D8 from the bottom of the uSD shield and jumpered the now open pin to D10 to get the software to work. I've been reading uSD cards for a couple days now and it seemed to work. Could the level be the problem? Are Maple DIO pins 5V compliant?

I'm going to spend some time investigating this. If I can't find what went wrong can I send the two boards in and have someone take a look to see what I did?

Posted 3 years ago #
tmbomber
Member

hmmmm.... IC3 is dead on one of them... that's a 3.3V regulator that supplies VDDA. Which would normally be the 5V output to the shield header...

http://www.sparkfun.com/datasheets/DevTools/Arduino/microSD_Shield-v13%20Schematic.pdf

The uSD shield doesn't use the 5V pin the uSD drive and the level shifter use the 3.3v pin.

btw, the level shifter is a 74HC4050D...

hmmmmm...

more testing tomorrow

Posted 3 years ago #
crenn
Moderator

Update: After taking a figurative lashing from the comments section we're begging for mercy by updating the microSD Shield board. All new microSD shields have a voltage converter chip on-board that converts the Arduino 5V signals to 3.3V signals in accordance with SD specifications. No more 'blown up' SD cards!

I think that is part of the clue to what could have gone wrong (I don't believe that could cause the smoke to be let out. Nor the level shifter.) however I'm not sure. I believe the shield has been changed from that update, however the changes haven't been reflected in the schematic.

The changes to the SD card shield, do you think that could have caused any shorts to anything? Could you draw up a schematic of everything you had hooked upto the Maple? Could the SD card be the issue?

Some of the Maple pins are 5V tolerant, however, the uSD shield uses a buffer to put the signals down to 3.3V anyway, so this should not be an issue.

Posted 3 years ago #
feurig
Member

First the regulators on the stock maples are a bit on the whimpy side although nothing should have touched the analog regulator, so I second crenns not sure why that would have happened. Second, since the sdcard is 3.3v, native you would be almost better wiring it up (there are only 6 wires) directly than paying for a relatively expensive product which is designed for a 5v processor.

Bummer about the r5s. RIP.

Posted 3 years ago #
Anonymous
Unregistered

The wimpy power supply on the Maple is the reason why I bypass everything and dump 5V or 3.3V directly to the Vin pin. All of my shields will thus draw power directly from my external power supply and leave the Maple's power supply to only a handful of things.

I'm so paranoid about blowing this board. Hell, I ran a 20x2 VFD directly from the Arduino Mega and Uno and never melted them down, and that was 400-500mA at 5V draw. :/

Posted 3 years ago #

tmbomber
Member

Hooo Boy...

http://leaflabs.com/docs/hardware/maple.html#power-regulation-on-the-maple

Specifically:

These voltage regulators nominally take an input of up to 16V. In addition,
while the maximum continuous output current for the board is 250mA, if you
are powering the board off higher voltages the amount off current it can
supply goes down, due to the regulators needing to dissipate the extra
power. So if you are powering the board off 12V, the max current is about
40mA at room temperature. In general (again, at room temperature) the max
power dissipation (PD) for the chip is about .37W, and output current =
PD/(Vin-Vout). For exact max current calculations, please refer to the
datasheet linked above.

If you are planning to draw a lot of current from the Maple board, it is
necessary to provide input power as close to 3.3V as possible. Powering
the microcontroller circuitry and LEDs on the board alone takes approximately
30mA, so if you are powering the board with 12V that leaves only 10mA (at best)
available for powering any user circuitry. Attempting to draw more than 10mA
runs the risk of shorting out the power regulators and bricking your board.

I'm thinking I overcurrented the regulators...

I'm going to try pulling them off the bricked board and putting some beefy
ones on the uSD shield. (It's not like I can damage it more)

I'll be trying that next week after the holiday.

Posted 3 years ago #

tmbomber
Member

Ok, the smoke's all back in the boards now. Or rather, I've removed the blown regulators and added some LM1117-3.3's to the uSD shield. The boards are working fine and I'm driving three LED's without damaging anything.

I'm thinking maybe y'all might consider upgrading those regulators a might...

Posted 3 years ago #
mbolivar
LeafLabs

I'm thinking maybe y'all might consider upgrading those regulators a might...

solved in maple 2. we feel your pain. glad you managed to get all the smoke.

Posted 3 years ago #
tmbomber
Member

Maple...2????

I'm thinking I need to hang around the developers forum a bit more...

In the interim, I've two more Maple boards I placed on order before smoking my current two and I'd rather not have to perform the same repair on them.

Is there a recommended best practice for driving multiple LEDs? I was thinking I could set the pins up as "OUTPUT_OPEN_DRAIN" and using the pins to sink rather source power.

Anybody know off hand the maximum current a single pin can sink? Or the total the chip can sink from multiple LEDs?

Posted 3 years ago #

RSS feed for this topic

Reply

You must log in to post.