Loud Pop = BAD! Right?

[Shawn]

I had just cancelled a print, and moved the bed away from the extruders,
then was waiting for things to cool down some. A few minutes later I
hear all the fans stop spinning, followed by a loud POP! and then the
fans spinning up again. But now there is no LED or control button
light. I'm guessing I just blew a capacitor.

I've already sent a note to MBI support.

My question is if there is anything I can do to get back to operational
status? I have an event on Saturday where I kinda sorta need a
functional printer.

I consider this a prime example to the corollary to Murphy's Law. The
law is "if anything can go wrong, it will". The corollary is "at the
most inopportune time". So, I'm not really surprised. Disappointed,
yes, but not really surprised. :)

Any tips on returning to an active state are appreciated!!

Shawn

[Laird Popkin]

I had exactly this happen to me, which was caused by a component blowing out on the main board, which required a replacement board. MBI support was great, but it took time to get the board FedExed. One thing I would recommend is labeling and photographing all of the connections before removing the board, as the diagram on the web site isn't completely clear as to what goes where. Swapping the board is pretty easy with that info.

[Jetty]

This happened to me too and has happened to a few others on this
forum. Check the 5V regulator (near the power switch) on the
Mightyboard
and see if you have a piece missing.

[Shawn]

Thanks. I pulled the motherboard late last night but couldn't see
anything that was obviously wrong. (other than an excessive amount of
solder IMO at a couple of points). I'll take another look today in
better light, and take a picture of the board/connectors too.

New parts are on their way. This is why MBI is getting a great
reputation - decent product AND great customer support.

Shawn

[PropellerScience]

If it is the 5V regulator I've got plenty floating around the house so
I'm good. I can't see anything on the schematic that would prompt me
to get another board from Makerbot. Of course, my board will never
burn out for any reason. Even if I spilled Pepsi on it!
http://vimeo.com/20995501

[Jetty]

> If it is the 5V regulator I've got plenty floating around the house so
> I'm good.

Wasn't that easy. When the 5V reg blew, so did a bunch of other stuff
on the board at the same time, including the 1280.

[RocketSled]

Collateral damage is common when power circuitry fails.  They can fail short and feed the unregulated input on to the regulated rail.

[PropellerScience]

It's been my experience the it's usually the downwind components that
cause a regulator to fail in the first place. It doesn't matter what
blows up, I do this stuff for a living, and my home workshop has
everything I need, including a stereo microscope for the small stuff.
If it's beyond small I have my trusty Balplan:
http://youtu.be/iAYJN7wwBOI

[Shawn]

UGH! Replacement parts arrived. Promptly blew again! Crap. This time
when trying to unload the right extruder. No movements, just unloading.
Sighs.

Of course the first thing I checked was that I had everything hooked up
right. I matched everything to my "before" pictures. So, I'm fairly
sure I had it all correct - except for the extruder thermistors which
were initially reversed. Identified this by the wrong extruder
indicating heat on doing an upload. Cooled everything off, shut down,
swapped the thermistors, and tried again. Got the expected behavior and
promptly moved to unloading the filament that is currently in there.
Then the blow.

I'm not supposed to have this running for the Calgary Maker Faire it
seems. :( (yep, MBI trouble ticket has been updated...)

Shawn

[Shawn]

er, typo... swapped thermistors were identified by doing an UNLOAD, and
watching the wrong extruder indicate heat....

[Jetty]

Check the X/Y/Z end stop cables for shorts. I believe MBI suggest a
faulty X cable can cause it.
But potentially any of the end stops could cause it if they short
internally. Also, check for shorts on your RGB LED
leads. Did you get new stepper drivers with your board? They were
probably nuked when your last
board died.

[Shawn]

New X-stop cable, though I may have had it in X-Min instead of X-Max.
But that would mean no cable on the X-Min at the time of failure, so no
chance of shorting there. The heads were approx centered...

This is the first I've heard of possible stepper damage...

[Jetty]

> New X-stop cable, though I may have had it in X-Min instead of X-Max.
> But that would mean no cable on the X-Min at the time of failure, so no
> chance of shorting there.

Wrong slot wouldn't do it.

Regarding the stop cable, they recommended to me to replace my X Stop
too
cable which I did. I pulled the old cable apart to see if I could
find any fault in it.
It was 4 conductors, 3 sheathed, and one bare. The bare one has
shrink on the end
at the connector, so it's not obvious it's bare inside until you pull
the connector apart.
There was no fault in the cable.

However I could see how it could happen on some bots, the cable has
one point of
failure, if one of the sheaths breaks through flexing or rubbing, you
get a short to
GND (potentially 5V to GND).

Obviously your X is okay, so next check the other stop cables. Z in
particular has
2 points where moving parts can rub against the cable and wear it over
time.
Top left hand corner of the back Y rod, and against the center of the
Y rod where the cable
exits the X stop.

When the regulator blows it tends to put 24V on the 5V line on the
board momentarily, which
will take out much of the board. I attempted to repair my board when
the reg blew, and it powered
up kind of, but the ATMega 1280 had been blown when the reg blew.

The A4982 on the stepper boards takes logic inputs
which are rated to 5.5V, so although the stepper drivers will handle
35V output to the steppers, they
won't handle 35V on the logic side.

So the fact that the 2nd board blew almost immediately could be due to
nuked stepper
drivers being placed on a new board, something up with the 24V power
supply (check it's
24V) or a short or intermittent short in something external to the
board.

The last 2 are fairly easy to check. (FYI steppers themselves you can
exclude, as the A4982
is protected from shorts on the output side, and the thermocouple you
can ignore too.)

I concluded that my issue wasn't due to the X stop cable but was due
to the FETS driving
the RGB LED wearing down the inverters over time due to gate
capacitance. Can't be
100% sure, but it seemed reasonable from the damage I saw on the
board and the schematic.

But ultimately, MBI will have a better idea of what's going on with
your bot because they have
a lot more information about failed boards due to the larger sample
set.