Shop OBEX P1 Docs P2 Docs Learn Events
Options for a Failed? EEPROM — Parallax Forums

Options for a Failed? EEPROM

Prophead100Prophead100 Posts: 192
edited 2012-10-18 05:49 in Propeller 1
Question: Is there an easy way to verify/troubleshoot the function of an EEPROM on a protoboard? Secondarily, how difficult would it be to solder in a new thru-hole EEPROM elsewhere on the board to replace the main failed one and have the system default to it?

Background: Here is what the project looks like. http://forums.parallax.com/showthread.php?143083-Transparent-Prop-Clock-with-Sensors I just finished putting it together and somewhere along the line the EEPROM appears to have failed. The program loads fine in RAM (F10) but not in EEPROM (F11). The EEPROM (F11) load works fine on a second empty board and my trusty PropBOE. Only pins 0-8, 13,14 and 24 are soldered up with an RTC(0-2), BMP085 (i2C on pins 13/14), TAOS234(3-8) and DHT21(24). The EEPROM and thru-hole pins looks like new but replacing the EEPROM in its currently location could be difficult because its partially below (non-touching) the VFD display. I may be able to reheat the solder on the pins of the EEPROM but it would be tight. The project would work without the EEPROM but I would have to reload the program after every power failure so this might be worth fixing but I'd hate to have to tear apart the hole board to replace it.

Thanks Prophead100

Comments

  • WBA ConsultingWBA Consulting Posts: 2,934
    edited 2012-10-15 18:09
    You could easily add a DIP-8 EEPROM to the board since P28 and P29 are available using the holes around the prop circuit. To disable the onboard EEPROM, just snip off pins 5 and 6 so that it is no longer connected to the Propeller pins at P28/29.
  • frank freedmanfrank freedman Posts: 1,983
    edited 2012-10-15 18:37
    If you have access to an air solder station, you could use that for removal of the failed part. Then install the replacement with zephpaste or similar. The lowest Ayoue was about 150.00 from sparkfun last year.
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2012-10-15 19:08
    Prophead100 wrote: »
    Question: Is there an easy way to verify/troubleshoot the function of an EEPROM on a protoboard? Secondarily, how difficult would it be to solder in a new thru-hole EEPROM elsewhere on the board to replace the main failed one and have the system default to it?

    Background: Here is what the project looks like. http://forums.parallax.com/showthread.php?143083-Transparent-Prop-Clock-with-Sensors I just finished putting it together and somewhere along the line the EEPROM appears to have failed. The program loads fine in RAM (F10) but not in EEPROM (F11). The EEPROM (F11) load works fine on a second empty board and my trusty PropBOE. Only pins 0-8, 13,14 and 24 are soldered up with an RTC(0-2), BMP085 (i2C on pins 13/14), TAOS234(3-8) and DHT21(24). The EEPROM and thru-hole pins looks like new but replacing the EEPROM in its currently location could be difficult because its partially below (non-touching) the VFD display. I may be able to reheat the solder on the pins of the EEPROM but it would be tight. The project would work without the EEPROM but I would have to reload the program after every power failure so this might be worth fixing but I'd hate to have to tear apart the hole board to replace it.

    Thanks Prophead100
    The secondary part of your question is being answered but I wonder why your EEPROM is faulty? I suspect a bad connection etc rather than a chip failure. Can you verify with a meter that there is a connection from the Prop to the EEPROM and that the lines are in no way shorted to something else? The only other thing is to make sure the pullup resistor is connected to SDA at least. You can also use the voltage range to make sure the lines are high. Always do the absolute basics first and be methodical, you will be surprised how many don't do this and assume it is something complicated or make it so. Don't be hesitant to unsolder some things if you have to, you need to do what you need to do and the whole exercise will probably take less time than posting this thread.
  • Prophead100Prophead100 Posts: 192
    edited 2012-10-15 21:40
    Thanks everyone! I'll try the troubleshooting Peter suggests then move on to a replacement if that doesn't work.
  • frank freedmanfrank freedman Posts: 1,983
    edited 2012-10-16 00:12
    Peter Jakacki wrote: »
    The secondary part of your question is being answered but I wonder why your EEPROM is faulty? I suspect a bad connection etc rather than a chip failure. Can you verify with a meter that there is a connection from the Prop to the EEPROM and that the lines are in no way shorted to something else? The only other thing is to make sure the pullup resistor is connected to SDA at least. You can also use the voltage range to make sure the lines are high. Always do the absolute basics first and be methodical, you will be surprised how many don't do this and assume it is something complicated or make it so. Don't be hesitant to unsolder some things if you have to, you need to do what you need to do and the whole exercise will probably take less time than posting this thread.

    If as Peter suggests you may have a faulty connection, just a dab of flux and reflow the pins should eliminate that as an issue, just be quick enough not to heat the chip to much. Or one pin, cool, one pin,.... etc. fine tip and beware of solder bridges.

    One last item, if you need to remove a suspect chip and you have a replacement on hand, clip the pins at the body, lift the remains and then remove the pins individually. While this method wastes the chip, it is intended to preserve the board. In practice chips are fairly replaceable while many times the boards are not. Better to destroy the chip than a few $K worth of board. (Well not in this case but in other things commercial, medical etc.)

    Frank
  • Prophead100Prophead100 Posts: 192
    edited 2012-10-16 21:58
    The eeprom seems to trace out ok to the pin 28, 29, VSS and VDD and the pull-up seems to be on spec. Looks like the next step is to try another eeprom. Should be fun!
  • RaymanRayman Posts: 14,670
    edited 2012-10-17 04:17
    I agree with Peter, that it's hard to imagine the eeprom going bad suddenly...
    You may also want to make sure you have the very latest version of the Prop Tool, just to be sure...

    I would triple-check all the connections on all the pins and check for shorts between pins, etc...
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2012-10-17 07:38
    Prophead100 wrote: »
    Question: Is there an easy way to verify/troubleshoot the function of an EEPROM on a protoboard? Secondarily, how difficult would it be to solder in a new thru-hole EEPROM elsewhere on the board to replace the main failed one and have the system default to it?

    Background: Here is what the project looks like. http://forums.parallax.com/showthread.php?143083-Transparent-Prop-Clock-with-Sensors I just finished putting it together and somewhere along the line the EEPROM appears to have failed. The program loads fine in RAM (F10) but not in EEPROM (F11). The EEPROM (F11) load works fine on a second empty board and my trusty PropBOE. Only pins 0-8, 13,14 and 24 are soldered up with an RTC(0-2), BMP085 (i2C on pins 13/14), TAOS234(3-8) and DHT21(24). The EEPROM and thru-hole pins looks like new but replacing the EEPROM in its currently location could be difficult because its partially below (non-touching) the VFD display. I may be able to reheat the solder on the pins of the EEPROM but it would be tight. The project would work without the EEPROM but I would have to reload the program after every power failure so this might be worth fixing but I'd hate to have to tear apart the hole board to replace it.

    Thanks Prophead100

    Regarding the first part of the question... If it loads to RAM and runs, but doesn't load to EEPROM - it is a EEPROM failure. Trying to get into a failed EEPROM to test it may be near impossible. Resoldering all the leads and the pullup resistors might resolve the problem.

    Verify that the resistors have not been shorted to give a hard pullup (maybe a nasty solder bridge)- that would have rather nasty implications as it might damage the i/o pin on the Propeller and render the future replacement of the EEPROM for boot purposes rather impossible.

    Everyone has pretty well covered your replacement options, but it is always of first importance to recognize any permanent damage.

    And while it is hard to imagine an EEPROM suddenly failing, that is the nature of failures. But the foremost source of problems that I have are due to my own assembly and disassembly of boards. Solder bridges were quite common until I decided to look for them each and every time I did anything involving the soldering iron.
  • RobotWorkshopRobotWorkshop Posts: 2,307
    edited 2012-10-17 08:04
    Prophead100 wrote: »
    The eeprom seems to trace out ok to the pin 28, 29, VSS and VDD and the pull-up seems to be on spec. Looks like the next step is to try another eeprom. Should be fun!

    You shouldn't need a hot air station to replace that small EEPROM. There are a couple easy ways to do it. You can carefully use an exacto or utility knife to cut the leads on one side of the chip. Be careful not to nick the board itself or the traces. Once a side is free you can bend up the chip and break off the four leads on the other side. To clean off the remnants of the leads just use some fresh solder with the soldering iron and they will usually just stick to the tip of the iron and then wipe them off on the sponge. You can then remote the rest of the excess solder with solder wick.

    Since that chip is so small you can also add solder to one side to heat all leads on that side at once while lifting that edge of the chip. That way you don't have to use a knife to cut any of the leads.

    When installing the new chip just add solder to one pad and when heating that pad place the chip in position. If it isn't perfect just re-heat that pin and move the chip. The solder the remaining connections. If you get too much solder or a solder bridge don't worry since you can go back over with solder wick and it usually just leaves the perfect amount of solder on the leads.

    Robert
  • Prophead100Prophead100 Posts: 192
    edited 2012-10-17 08:45
    I'll pull of the display to get to the eeprom better tonight. Some triple checks then microsurgery....
  • frank freedmanfrank freedman Posts: 1,983
    edited 2012-10-17 09:50
    Speaking of microsurgery, I use a set of surgical loupes I got on ebay from a dental student. 2.5X at 18 inch working distance. Makes life much better once you get used to working at arms length rather than trying to cram all the tools and sh! Er, stuff under a big magnifying glass .5in off the work. Go with the flip up ones so that you are not constantly having to take them off or look sideways with them on. Cheaper than fixed. My dentist regrets going with fixed lenses.

    Good luck
    FF

    Prophead100 wrote: »
    I'll pull of the display to get to the eeprom better tonight. Some triple checks then microsurgery....
  • RaymanRayman Posts: 14,670
    edited 2012-10-17 13:31
    Here's how I would remove that tiny eeprom chip:
    Put a bunch of solder on both sides, so there's a glob on each end that covers all the pins.
    Heat one side to 720F and then quickly move to the other side and push.
    If you do it fast enough, you can push it off it's pads.
    Then, you can remove it and clean up.
  • RDL2004RDL2004 Posts: 2,554
    edited 2012-10-18 05:49
    You can find knock-offs of the Atten AT-858D on ebay for under $50 - might be useful to have one. I bought mine last year for $80, but it also includes a regular soldering iron. Does a good job on heat shrink too. Watch out because as far as I know the actual AT-858D is only available for 220v.
Sign In or Register to comment.