Shop OBEX P1 Docs P2 Docs Learn Events
How many Prop chips have YOU burnt out? — Parallax Forums

How many Prop chips have YOU burnt out?

GeeksGoneBadGeeksGoneBad Posts: 100
edited 2012-08-16 13:31 in Propeller 1
I've gone through three in the past few days and I'm wondering if that's an unusually high number when trying to learn? I'm hoping I'm on par and not a complete doofus LOL

How many did y'all burn out when goofing around?
«13

Comments

  • Heater.Heater. Posts: 21,230
    edited 2012-07-27 12:15
    None. They are tough little buggers.
    Unless you are not wiring up all your power pins correctly or majorly abusing the other pins.
  • g3cwig3cwi Posts: 262
    edited 2012-07-27 12:15
    <0.1. I destroyed an IO pin.

    Three in a few days sounds a little unfortunate. What kind of goofing around were you doing?

    Cheers

    Richard
  • LeonLeon Posts: 7,620
    edited 2012-07-27 12:22
    One. I had a nearly invisible short on a PCB between two regulator pins that put 9V onto the 3.3V supply.
  • HShankoHShanko Posts: 402
    edited 2012-07-27 12:39
    @ Leon,

    OUCH!

    @ GeeksGoneBad

     Yes, you are doing something MAJOR bad. Is this on a breadboard, wrong wiring, power wire backwards, etc.? I have never lost one. Maybe lucky, or careful. A miswire can only take a short moment after power is applied to cause something to GO.
  • PaulPaul Posts: 263
    edited 2012-07-27 13:32
    I found 5 bad BS2P24's yesterday. Now that's a chuck of change! Can't remember any Props going bad.
  • mindrobotsmindrobots Posts: 6,506
    edited 2012-07-27 13:41
    Sorry, none here. I do much more software work than hardware work...while my code is bad it isn't bad enough to destroy a chip....so far! :lol:
  • PublisonPublison Posts: 12,366
    edited 2012-07-27 13:46
    Easy to do without decoupling caps. Have you tried running them at internal oscillator, ( no xtal , no PLL) ?

    No caps can render the PLL circuit inoperable. Non reversible.
  • average joeaverage joe Posts: 795
    edited 2012-07-27 13:46
    I will agree with everyone that burning up 3 chips is strange. I have not burnt up a single prop, nor damaged a pin yet. I'm sure since I just posted this I'm due to burn one up now. Double check your supply lines, remember to connect BOTH sides of the chip's supply rails! I've heard of others burning chips up this way. Other than that, these chips are VERY tough. I've accidentally abused the chip in various ways such as over 8volts on the supply rail, 5volt inputs to pins, pins shorted to ground, etc. These were momentary faults, I'm sure the propeller couldn't handle SUSTAINED conditions that I listed.

    More info would be helpful. A picture is worth 1000 words, but a schematic is worth millions...
  • Oldbitcollector (Jeff)Oldbitcollector (Jeff) Posts: 8,091
    edited 2012-07-27 14:01
    Pooched the PLL on one that I know.. Never fried one completely.

    Did cut some traces on a protoboard late one night. Felt stupid for working on a project at 3am, half-asleep. :)

    OBC
  • Duane C. JohnsonDuane C. Johnson Posts: 955
    edited 2012-07-27 14:26
    4, as I recall, over a years time out of 21.

    Clearly my fault always.
    I am a hardware builder, mostly on plugboards.
    All were executed when high voltage got to the wrong place.

    For what I do this seems about right. Part of doing business.

    BTW, once something is built I haven't had any failures.

    The 3 in a row, not good, unplug the chip and test before putting
    another back, Otherwise it can get expensive.

    Duane J
  • SapiehaSapieha Posts: 2,964
    edited 2012-07-27 16:08
    Hi.

    None --

    And I uses extensive Over-clocking.
    Test all PCBs I made for Bill Henning -- that needs often change place of Propeller from one PCB to other.
    And Over-clocking in wide spectra of frequency.
  • Mike GMike G Posts: 2,702
    edited 2012-07-27 17:02
    None --

    My workhorse -- PROPELLER DEMO REV C 2006 -- runs like a champ.

    The On/Off switch is a takes a tender touch though.
  • LevLev Posts: 182
    edited 2012-07-27 18:08
    I've killed or maimed two, by neglecting to connect up all of the grounds and Vdd's. Doh!
  • User NameUser Name Posts: 1,451
    edited 2012-07-27 18:24
    Seven volts momentarily on Vss were the cause of both Prop deaths here. Both were the result of careless probing. In all other respects, the Prop has been a battle tank.
  • BitsBits Posts: 414
    edited 2012-07-27 18:29
    I have burned up 3.

    Considering I engineer about 10 new products a year, I would say I am overdue. :innocent:
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-07-27 23:29
    No Propeller chips have me burnt out yet. Why do you ask? :)

    -Phil
  • msrobotsmsrobots Posts: 3,709
    edited 2012-07-27 23:39
    Phil - you somehow missed the point. But good to know anyways.

    Do I bother you pushing that AUTODOC-thing of you here in those foras? I really like it ...

    and, hmm, never burnt a prop. Stepped on one and bend all those pins.

    Enjoy!

    Mike
  • TylerSkylerTylerSkyler Posts: 72
    edited 2012-07-28 00:44
    One, still brings tears to my eyes...
  • g3cwig3cwi Posts: 262
    edited 2012-07-28 01:30
    Losing a prop would be like losing a friend. No, worse: eight friends.
  • computer guycomputer guy Posts: 1,113
    edited 2012-07-28 03:13
    None. But that's because I bug everyone on the forum before trying anything new.
  • Cluso99Cluso99 Posts: 18,069
    edited 2012-07-28 03:37
    Most likely 1 - overcooked on IR heater - the top bulged and cracked.

    I am amazed that I have not killed others as I have done some weird things on my prototypes.

    Like Sapeiha, I almost always overclock to 104 MHz.
  • Oldbitcollector (Jeff)Oldbitcollector (Jeff) Posts: 8,091
    edited 2012-07-28 11:48
    computer guy wrote: »
    None. But that's because I bug everyone on the forum before trying anything new.

    Now that's cheating! :)

    OBC
  • HumanoidoHumanoido Posts: 5,770
    edited 2012-07-28 13:41
    Zero. In two years with over a hundred chips, only one went dead on its own.
  • frank freedmanfrank freedman Posts: 1,983
    edited 2012-07-28 16:10
    GeeksGoneBad wrote: »
    I've gone through three in the past few days and I'm wondering if that's an unusually high number when trying to learn? I'm hoping I'm on par and not a complete doofus LOL

    How many did y'all burn out when goofing around?

    0, but not for lack of trying.....

    Maybe when the curve tracer comes off the back burner.
    FF
  • zoopydogsitzoopydogsit Posts: 174
    edited 2012-07-28 16:28
    None.
    I've been amazed they just run and run. I've left them out running complex code for months between code changes and seen amazing reliability. I've gotten them to do a lot of complex and amazing things. I make my own boards and having worked with TTL and various uProcs over the years then I'm comfortable designing, building, debugging and troubleshooting things. I've had my share of problems to troubleshoot like the common "no propeller chip detected on any port", but this is a relatively simple problem to troubleshoot using a logic probe. An example the wife was cleaning our bedroom and accidentally knocked a running platform off the bedside table, still ran but couldn't be detected for loading code, turned out to be a hairline fracture in a surface mount cap. I have had props that appeared to crash, but it turned out to be bad power supply circuits, not the prop. When I thought this was an issue I got it to run the most basic code to flash a led to give me confidence that the chip had basic functionality before pushing up the code again, turned out when you have all 8 cores running it uses more current and stretches the power supply leading to the problem.

    I've read earlier in the forum that in some rare circumstances folks have killed the internal PLL by not using decoupling caps on the power rails. They could still get it to load code and run on the internal clock, just not the external XTAL. I suspect that they probably didn't have a strong ground track between all of the prop's ground pins and may have had a set floating. The learning from this is four fold. 1. Connect all power pins. 2. Connect them together. 3. Use decoupling caps on all input power pins. 4. Understand how to take the prop back to the minimum baseline for troubleshooting.

    In general, if you give it strong clean 3.3v, use decoupling caps on every set of power pins, have a good solid ground and don't over voltage or overdrive (too much drain current) the I/O pins, then they seem to be auite indestructible and amazingly stable workhorses (8 of them).

    This may sound stupid, but what does dead mean? How do you know it's dead?
    (if this is "propeller not found on any port then this may not be a bad prop and can be a fault in the circuitry for reset handshaking and the prop replying).

    Please tell us know what you did.
  • jazzedjazzed Posts: 11,803
    edited 2012-07-28 19:16
    I've popped a few QFNs with a soldering iron. Not buying anymore of those.
  • Martin HodgeMartin Hodge Posts: 1,246
    edited 2012-07-28 20:43
    Just found the first bad Prop a couple days ago. Bad PLL it would seem. This chip was in a normal production run that had no other problems on the board.
  • Toby SeckshundToby Seckshund Posts: 2,027
    edited 2012-07-29 07:53
    My "9 Volt test" did for about 32 pins ie one Prop, the EEPROM .went too, but the SD card cooled down and worked again.

    This Prop was the fastest one too, D'oh!
  • prof_brainoprof_braino Posts: 4,313
    edited 2012-07-29 08:03
    None, but I found out that if we connect 5 volts to Vdd instead of Vin on the quickstart, they still work but get REALLY hot. Goofy how that works out, eh?
  • prof_brainoprof_braino Posts: 4,313
    edited 2012-07-29 08:09
    GeeksGoneBad wrote: »
    I've gone through three in the past few days and I'm wondering if that's an unusually high number when trying to learn? I'm hoping I'm on par and not a complete doofus LOL

    Can we ask exactly what your set up was to yield so many in a short time? Not that I want to duplicate your results, I would like to know what to avoid. :)

    And it would be helpful to arrive at your "doofus" evaluation request. ;)
Sign In or Register to comment.