• AVR Freaks

Hot!MCLR capacitor (RC) required for PIC24FJ64GA705 ?

Page: 12345.. > >> Showing page 1 of 6
Author
user2x
Super Member
  • Total Posts : 429
  • Reward points : 0
  • Joined: 2011/02/10 20:43:36
  • Location: 0
  • Status: offline
2019/10/08 20:57:55 (permalink)
0

MCLR capacitor (RC) required for PIC24FJ64GA705 ?

Hello,
 
I would like to use MCLR only for ICD / debug connection but I also do not want spurious resets caused by EMC.
 
The datasheet shows a RC network on the MCLR pin. 
Is that actually needed for proper device reset operation this device reset itself otherwise (POR) at power up? 
 
I am considering a 4k7 pullup on that pin and also route it to the ICD connector.
 
What say ye?
 
Regards
X
 
 
 
 
 
#1

105 Replies Related Threads

    ric
    Super Member
    • Total Posts : 28677
    • Reward points : 0
    • Joined: 2003/11/07 12:41:26
    • Location: Australia, Melbourne
    • Status: online
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2019/10/08 21:12:52 (permalink)
    5 (1)
    You don't need the capacitor, just a pullup.
    The built in power on delay is fine.
     

    I also post at: PicForum
    Links to useful PIC information: http://picforum.ric323.co...opic.php?f=59&t=15
    NEW USERS: Posting images, links and code - workaround for restrictions.
    To get a useful answer, always state which PIC you are using!
    #2
    user2x
    Super Member
    • Total Posts : 429
    • Reward points : 0
    • Joined: 2011/02/10 20:43:36
    • Location: 0
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2019/10/08 21:29:13 (permalink)
    0
    Great, I was hoping it would be that. 
     
    The difference I found is that on this CPU version, the MCLR cannot be disabled with the CONFIG bits.
     
    Regards
    X
     
    #3
    user2x
    Super Member
    • Total Posts : 429
    • Reward points : 0
    • Joined: 2011/02/10 20:43:36
    • Location: 0
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 14:37:08 (permalink)
    0
    It has been a while since this thread was started.
    But, I now have 9 boards out of 30 that appear "Dead"
     
    Well, not completely dead as they program and verify fine with IPE but do not seem to run the code.
    The voltages and state of MCLR seem fine.
     
    I have started a Microchip case and they came back yesterday saying that the problem is likely caused by the capacitor not being fitted to the MCLR line.
     
    I find that hard to believe but do not have any other explanations apart from perhaps soldering issues (QFN) either.
     
    What do you think?
    post edited by user2x - 2020/09/09 14:39:22
    #4
    NKurzman
    A Guy on the Net
    • Total Posts : 18976
    • Reward points : 0
    • Joined: 2008/01/16 19:33:48
    • Location: 0
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 15:18:37 (permalink)
    4 (1)
    Which PIC?
     
    #5
    Antipodean
    Super Member
    • Total Posts : 1927
    • Reward points : 0
    • Joined: 2008/12/09 10:19:08
    • Location: Didcot, United Kingdom
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 15:36:55 (permalink)
    5 (1)
    NKurzman
    Which PIC?

     
    The one in the subject line presumably, seeing it is the OP continuing the thread.
     
    To the OP i could quite well imagine the capacitor is required, as this will hold the PIC in reset while the internal voltage regulator stabilises, which I suspect is where your problem lies. Not all PICs will start from an unknown power ramp up period reliably.
     

    Do not use my alias in your message body when replying, your message will disappear ...

    Alan
    #6
    NKurzman
    A Guy on the Net
    • Total Posts : 18976
    • Reward points : 0
    • Joined: 2008/01/16 19:33:48
    • Location: 0
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 15:47:17 (permalink)
    0
    Reading is Fundamental.
    I have never used a Cap on a PIC.
    But if you are correct then they may need it.
    This chip does not appear to have a Built in Power on delay in the configuration.
    May be the BOR could help.
    #7
    NorthGuy
    Super Member
    • Total Posts : 6350
    • Reward points : 0
    • Joined: 2014/02/23 14:23:23
    • Location: Northern Canada
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 16:32:30 (permalink)
    0
    When you program the PIC, it executes a program called "programming executive". It is possible to program without programming executive, but it's rather slow, so I don't think Microchip tools would do this. Programming executive runs just the same as any other program. Therefore, your PICs can run.
     
    It is hard to figure out what is the problem with your PICs - not enough information. But you can investigate. Start from something simple, like a LED blinker running from FRC without PLL. Program it to bad PICs and see if some of your bad PICs can blink LEDs.
     
    #8
    user2x
    Super Member
    • Total Posts : 429
    • Reward points : 0
    • Joined: 2011/02/10 20:43:36
    • Location: 0
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 17:39:57 (permalink)
    0
    Hello everyone,
     
    Yes, the PIC is the same as in the subject except it is the smaller footprint QFN version PIC24FJ64GA702.
     
    I do not have more information to share at the moment except that in 6 out of 6 cases, heating the CPU with a heatgun, resolved the problem for at least long enough to do pass testing.
     
    Replacing the CPU also fixed the issue in a further few boards.
     
    And for that reason, I have been chasing manufacturing defects for a while but have now hit a dead end as the manufacturer has come back with Xray pictures showing no issues with the soldering of the QFN. The reflow oven profiles are also within the correct specification.
     
    That reset capacitor is the only thing that stands out.  It was omitted because of severe space constraints and because I myself as well as the initial replies here seemed to agree it is not required.
     
    I am going to investigate further...
     
    #9
    user2x
    Super Member
    • Total Posts : 429
    • Reward points : 0
    • Joined: 2011/02/10 20:43:36
    • Location: 0
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 17:56:12 (permalink)
    0
    Update:
    A 100nF capacitor has been fitted from MCLR to GND.  The pullup is 4k7.
     
    The chip still appears as dead as before. 
     
    The only somewhat relevant Errata I can see is
    6. "The main BOR may not function on some devices"    This is given for Silicon revision A3.
    I am not sure if this can be relevant for this problem nor what silicon revision these chips are.
     
     
     
     
     
    post edited by user2x - 2020/09/09 18:06:17
    #10
    ric
    Super Member
    • Total Posts : 28677
    • Reward points : 0
    • Joined: 2003/11/07 12:41:26
    • Location: Australia, Melbourne
    • Status: online
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 18:07:33 (permalink)
    0
    NorthGuy
    ...
    It is hard to figure out what is the problem with your PICs - not enough information. But you can investigate. Start from something simple, like a LED blinker running from FRC without PLL. Program it to bad PICs and see if some of your bad PICs can blink LEDs.

    +1
    I suspect something in your program is running the chip out of spec, and only some of the chips can do it.
    Try a very basic "flash an LED" program first.
     
     

    I also post at: PicForum
    Links to useful PIC information: http://picforum.ric323.co...opic.php?f=59&t=15
    NEW USERS: Posting images, links and code - workaround for restrictions.
    To get a useful answer, always state which PIC you are using!
    #11
    NorthGuy
    Super Member
    • Total Posts : 6350
    • Reward points : 0
    • Joined: 2014/02/23 14:23:23
    • Location: Northern Canada
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 18:09:08 (permalink)
    0
    Since replacing the chip requires heating too, this simply looks as bad soldering. May be not necessarily the PIC itself, but something in a vicinity.
    #12
    user2x
    Super Member
    • Total Posts : 429
    • Reward points : 0
    • Joined: 2011/02/10 20:43:36
    • Location: 0
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 18:12:06 (permalink)
    0
    Understood.
    I need to get them back from the factory and do exactly that.
     
    The code runs basic 8MHz from internal oscillator. No PLL.
     
    Are we now sure about the capacitor requirement (or not)?
     
     
    #13
    user2x
    Super Member
    • Total Posts : 429
    • Reward points : 0
    • Joined: 2011/02/10 20:43:36
    • Location: 0
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 18:40:17 (permalink)
    0
    NorthGuy
    Since replacing the chip requires heating too, this simply looks as bad soldering. May be not necessarily the PIC itself, but something in a vicinity.


    Yes, but they Xrayed them and found no issues. I am not sure how reliable that method actually is.
     
     
     
     
     
    #14
    NorthGuy
    Super Member
    • Total Posts : 6350
    • Reward points : 0
    • Joined: 2014/02/23 14:23:23
    • Location: Northern Canada
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 20:11:07 (permalink)
    0
    user2x
    Are we now sure about the capacitor requirement (or not)?



    If you install the capacitor and it doesn't solve the problem then the lack of the capacitor cannot cause the problem.
    #15
    user2x
    Super Member
    • Total Posts : 429
    • Reward points : 0
    • Joined: 2011/02/10 20:43:36
    • Location: 0
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/09 20:13:24 (permalink)
    0
    NorthGuy
    user2x
    Are we now sure about the capacitor requirement (or not)?



    If you install the capacitor and it doesn't solve the problem then the lack of the capacitor cannot cause the problem.


    yes but that is on a sample of one only.
    #16
    Jerry Messina
    Super Member
    • Total Posts : 554
    • Reward points : 0
    • Joined: 2003/11/07 12:35:12
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/10 03:50:53 (permalink)
    0
    If you think it's a power on timing issue then GND the MCLR line, power up the target, and release MCLR.
    If it still doesn't run then a capacitor isn't going to solve anything.
     
    #17
    Antipodean
    Super Member
    • Total Posts : 1927
    • Reward points : 0
    • Joined: 2008/12/09 10:19:08
    • Location: Didcot, United Kingdom
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/10 04:57:39 (permalink)
    0
    user2x
    Update:
    A 100nF capacitor has been fitted from MCLR to GND.  The pullup is 4k7.
     The chip still appears as dead as before. 
     

    What value resistor is shown on the datasheet? I have a feeling it is around 10x that typically.
     
    user2x
    The only somewhat relevant Errata I can see is
    6. "The main BOR may not function on some devices"    This is given for Silicon revision A3.
    I am not sure if this can be relevant for this problem nor what silicon revision these chips are.
     

     
    Doesn't the programmer tell you the chip revision?
     
    user2x
    NorthGuy
    Since replacing the chip requires heating too, this simply looks as bad soldering. May be not necessarily the PIC itself, but something in a vicinity.

    Yes, but they Xrayed them and found no issues. I am not sure how reliable that method actually is.
     

     
    X-ray can be a good indicator, but is also reliant on the pads properly tinning. The pad may not tin but the ball will deform while molten as the rest of the balls flow onto their pads due to the way the chip settles.
     
    Jerry Messina
    If you think it's a power on timing issue then GND the MCLR line, power up the target, and release MCLR.
    If it still doesn't run then a capacitor isn't going to solve anything.
     


    This would be a good starting point to ensure that startup timing is the source of the problem.


     

    Do not use my alias in your message body when replying, your message will disappear ...

    Alan
    #18
    user2x
    Super Member
    • Total Posts : 429
    • Reward points : 0
    • Joined: 2011/02/10 20:43:36
    • Location: 0
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/10 14:25:41 (permalink)
    0
    Hmm, the plot thickens (He said, scratching his beard).
     
    UPDATE:
    Grounding the MCLR and also trying a capacitor did not fix anything.
     
    I created a test program that flashes a LED.  It WORKS!
    I added all my code back in and moved the LED flash operation down in the code to see where it stops working.
    It never did and the whole code then basically worked as per expectation.
     
    The original .hex file is now back in and it also works fine.
     
    I am now wondering the programmer itself can be to blame for the problems.
    Perhaps it does not always program the chips correctly. Is that even possible?
     
    Unfortunately, the only board I had with this problem is now ok so I can't do anything else to it. 
    Unless, it is indeed some sort of dry joint that will return later.
     
    I have 5 boards (previously dead) coming back today, with CPU chip replaced...
     
    Regards
    X
     
     
     
     
    #19
    Antipodean
    Super Member
    • Total Posts : 1927
    • Reward points : 0
    • Joined: 2008/12/09 10:19:08
    • Location: Didcot, United Kingdom
    • Status: offline
    Re: MCLR capacitor (RC) required for PIC24FJ64GA705 ? 2020/09/10 15:37:42 (permalink)
    0
    user2x
    ...
    I am now wondering the programmer itself can be to blame for the problems.
    Perhaps it does not always program the chips correctly. Is that even possible?
     


    Definitely a possibility. From the tone of previous posts it sounds like you are having the PCBs assembled by a board house. Do they also program them then test? If so I wouldn't put it [past the programming operator to be a bit loose and fast with the operation resulting in erroneous programming.
    But that wouldn't explain one not operating correctly after you programme it.
     

    Do not use my alias in your message body when replying, your message will disappear ...

    Alan
    #20
    Page: 12345.. > >> Showing page 1 of 6
    Jump to:
    © 2020 APG vNext Commercial Version 4.5