Printer issues (recently upgraded to sailfish)

[CarryTheWhat]

I'm not sure if these issues are related, but they both showed up at about the same time, a few days after upgrading to sailfish.

The small LCD on the front of the Replicator appears to be obstructed somehow (pictures attached). When viewed directly, all you can see is a matrix of black rectangles, but you cannot make out what is being displayed. But when viewed from above at a steep angle, you can just make out the display.

The other problem is much more serious. No matter what it is I'm printing, the printer is always stalling mid-print. When I left the printer off all night, this happened 20-30 minutes into the print -- but eventually, it is happening on every print before the raft is even finished (and sometimes before the raft even begins). The print just freezes where it is, but no error is reported. On the LCD, I can still pause the print, but when I resume, it returns to it's previous position, and continues to wait. RepG continues to count upwards in Elapsed time, but Est done does not change. It looks like RepG has crashed somehow... or suddenly stopped communicating.

Immediately when the printer stalls, then repG starts to spit out these errors:

[10:23:32] Read timed out. 
[10:23:32] Read timed out; retries remaining: 5 
[10:23:34] Read timed out. 
[10:23:34] Read timed out; retries remaining: 4 
[10:23:38] Read timed out. 

Has anyone encountered either of these problems?

[Dan Newman]

Sounds like your MightyBoard is ill. You can try redownloading the firmware to
the board and see if that resolves the issue. However, it's likely a problem
with the electronics themselves. You can contact MBI support to see if your
board is failing.

Dan

> --
> You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
> To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/AIZTw0Ty6roJ.
> To post to this group, send email to make...@googlegroups.com.
> To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.
> <again1.jpg><again2.jpg>

[Dan Newman]

On 1 Nov 2012 , at 4:07 PM, Dan Newman wrote:

> Sounds like your MightyBoard is ill. You can try redownloading the firmware to
> the board and see if that resolves the issue. However, it's likely a problem
> with the electronics themselves. You can contact MBI support to see if your
> board is failing.

BTW, basis for suspecting an ill board is that the LCD display is malfunctioning.
The firmware download process writes the firmware and then reads it back to verify
that it was written correctly. So, unless you simply didn't notice a warning from
RepG, the firmware write should have been a complete success or a complete failure
but nothing in between.

Dan

[CarryTheWhat]

Interesting.. Yeah there were no problems uploading the new firmware. And it continued working fine for a while, I'm not sure if they are related problems. In the past, I have had the same LCD display issue, and i think that board eventually fried.. But it did not cause the communication interrupt that result in failed prints.

I am reverting the firmware now, but I don't see how that will help the LCD.

In case it matters; when I start up the Replicator, the LCD displays as normal.  But over the course of 2 or 3 minutes, it gradually gets darker and darker.

[Dan Newman]

On 1 Nov 2012 , at 4:14 PM, CarryTheWhat wrote:

> Interesting.. Yeah there were no problems uploading the new firmware. And
> it continued working fine for a while, I'm not sure if they are related
> problems. In the past, I have had the same LCD display issue, and i think
> that board eventually fried.. But it did not cause the communication
> interrupt that result in failed prints.
>
> I am reverting the firmware now, but I don't see how that will help the LCD.
>
> In case it matters; when I start up the Replicator, the LCD displays as
> normal. But over the course of 2 or 3 minutes, it gradually gets darker
> and darker.

Sounds like you have another board going bad. From various reports, they
are susceptible to issues from problematic endstop cables. Did MBI have you
replace them when you replaced the board? (There's been a couple of reports
in this forum of folks receiving a new board AND a new X axis endstop cable.)

At any rate, there's nothing in the firmware to slowly fade your LCD display
or to write lots of black rectangles on it. And if there's one problem (display),
then there could just as likely be some other issues (e.g., USB comms).

Dan

[CarryTheWhat]

Ok so that was pretty well timed. 10 minutes later, the same voltage regulator exploded, as before (pictured)

Yeah before, they gave me both a new board and an X axis endstop cable, which I replaced (but was kind of hard because of how perfect the cables are expected to kink in just the right way).

Well, this thing is out of commission for a while.....

[Test]

I just had my voltage regulator blow again as well, even after the replacement cable. 3rd time now. 

Support has asked me to ship the whole Rep1 back so they can dig through it. I doubt that will be quick, particularly given the storm, but hopefully they learn something useful. 

From what I gather, the voltage regulator can fail for any number of reasons. It is just the weak link on the board. Others have mention that 24v to 5v leaves a lot of heat to dissipate. 

Has anyone looked at a Rep2 board to see if they might have fixed this issue? It must be costing a fortune in replacement boards and the Rep2 just makes it that much more problematic. 

Sent from my iPhone

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/vDruZidzdWIJ.

To post to this group, send email to make...@googlegroups.com.
To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.

<again3.jpg>

[Gary Crowell]

I believe the Rep2 has a switching regulator for 5V.

Gary

--
----------------------------------------------
Gary A. Crowell Sr., P.E., CID+

Linkedin  Elance  KE7FIZ

[Gary Crowell]

Actually, I've heard conversations of this sort many times...

engineer:  "Ok, we've got a good design for the 5V switching regulator; it'll cost about $2 for parts."

manager:  "A switcher?  Can't we just use a regular regulator."

engineer:  "A linear regulator?  Yeah, it'll only cost about 50 cents, but it'll get really hot..."

manager:  "But it'll work right?  Why spend the extra buck and a half?"

engineer:  "Sure, but the lifetime will be shorter and if there's any cooling problem..."

manager:  "Fine then.  Linear regulator it is."

Sigh.

[Jon Daniels]

This happened to me a few days ago.  Same regulator blown apart...  Waiting for a replacement board.

[RocketSled]

On Saturday, November 3, 2012 5:01:55 PM UTC-4, Jon Daniels wrote:

This happened to me a few days ago.  Same regulator blown apart...  Waiting for a replacement board.

I deal with the reliability of electronics in my day job (I make SSDs).

If you're blowing Regulators repeatedly, there is likely to be something else wrong to blame for the problem.  The existing regulator may not be as reliable as it is possible to be, but it clearly is reliable enough or there would be lots more people experiencing failures (it's all about the statistics of reliability).  Two failures in a row are long odds from a component standpoint, but high probability if there's some other contributing factor.

Without going back to look at the schematics to see where 5V goes, if it goes off-board (like, for instance to the limit switches) you might want to inspect your wiring harnesses for exposed wire/short circuits. If 5V goes off the motherboard and that's where the problem is, your new board is just going to blow like the old one...

[Clinton Hoines]

Curious is there any advantage to putting a small heat sink on the regulator? Like this guy a pc ram heatsink. http://www.memoryexpress.com/Products/MX5173

Would that extend the lifetime of the regulator if there is no issue at all, preventative maintenance?

 

Clinton

[RocketSled]

On Saturday, November 3, 2012 6:07:10 PM UTC-4, Clinton Hoines wrote:

Curious is there any advantage to putting a small heat sink on the regulator? Like this guy a pc ram heatsink. http://www.memoryexpress.com/Products/MX5173

Would that extend the lifetime of the regulator if there is no issue at all, preventative maintenance?

Without a doubt.  All electronic components have a probability of failure, usually called the "FIT rate".  The probability is governed by the design of the circuit, and the "Arrhenius Equation".  Arrhenius basically says that failure rate increases exponentially with temperature. 

So components have a nominal probability of failure that is always non-zero.  The longer they operate, the more likely they are to fail.  Higher operating temperatures accelerate that time to failure (TTF).  So conversely, lower operating temperatures retard the TTF.  It's why electronics like a computer have fans. 

But... Semiconductors have a maximum allowable temperature, TJmax (maximum junction temperature, typically 125ºC), above which the probability of failure quickly approaches 100%.  So if power dissipation causes junction temps to go above the maximum (like, if you have a dead short) no amount of heatsinking is going to be enough.  

The upper limit in performance for a microprocessor is usually dictated by TJMax.  As the clock speed increase, the processor does more "work" per unit time, which means it dissipates more power per unit time, which means it gets hotter.  That's why a really good heat sink allows you to overclock to much higher frequencies than the part is rated to operate (I run my Intel i7 3930 at 4.3GHz, but it's only rated for 3.2GHz).  Extreme overclockers immerse their entire PC in a refrigerated non-conductive liquid like mineral oil, because that gives them about the most-effective heat sink it's possible to make. 

There are exceptions, but for most electronics, cooler is almost always better.