Hot!PIC16F19176 Family Silicon Errata and Silicon Revision A3

Author
John Linq
New Member
  • Total Posts : 12
  • Reward points : 0
  • Joined: 2016/08/31 19:23:12
  • Location: 0
  • Status: offline
2019/04/11 20:10:39 (permalink)
0

PIC16F19176 Family Silicon Errata and Silicon Revision A3

 
In DS80000744A, there is an item [Program Flash Memory (PFM) Endurance],
In DS80000744D, there is an item [ADC2 Channel Switching],
Both of them are removed in the newer Silicon Errata, what does this removing mean?
The problem has been fixed, or it is not a problem?
 
For ADC2, the maximum recommended impedance for analog sources is 1k.
Usually, it is 10k.
This new specification brings some trouble to us, the Silicon Errata item [ADC2 Conversion]
[At the very beginning of the ADC conversion, the input signal may briefly be pulled to ground.]
makes the situation worse.
 
Does anyone know when and where to buy the Silicon Revision A3 chip?
 
Thanks for any inputs.
 
 
#1

14 Replies Related Threads

    acharnley
    Super Member
    • Total Posts : 248
    • Reward points : 0
    • Joined: 2016/05/01 06:51:28
    • Location: 0
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/12 09:11:56 (permalink)
    0
    Only 1K? Yuk! Can you move chip?

    I've been working with these and the PIC18 which is only $0.24 more per chip. I find it to be a lot more rounded, for example the ADC has a charge-pump feature which gets much better accuracy.

    Pulling the sampling cap to ground is usually to discharge before doing a second read on a different source, but sometimes you want to do the opposite and pull it high (which ever is closer to what you're expecting). That's another PIC18 feature.
    #2
    coffee critic
    Super Member
    • Total Posts : 350
    • Reward points : 0
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/12 09:35:14 (permalink)
    0
    Your sales engineer or FAE can get a lead time and order samples of a specific new revision.  Microchip Direct should also be able to handle these types of request longer term. 
     
    Did you check if increasing the TAD to 4us resolves the issue on A1? 
    post edited by coffee critic - 2019/04/12 09:53:55

    n_*$
    #3
    mlp
    boots too small
    • Total Posts : 709
    • Reward points : 0
    • Joined: 2012/09/10 15:12:07
    • Location: previously Microchip XC8 team
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/13 09:54:50 (permalink)
    0
    John Linq
    Both of them are removed in the newer Silicon Errata, what does this removing mean?
    The problem has been fixed, or it is not a problem?

    The items not appearing in Errata for newer silicon means that the problem has not been found in newer silicon.
    It might be that the problem still exists but nobody has hit the newer silicon hard enough to see it.
    It might be that the root cause has been addressed and the problem no longer exists.
     
    To know more detail you would need to talk to someone with a link to deep inside Microchip, and you're not likely to find them here on the forum. Talk to your FAE.

    Mark (this opinion available for hire)
    #4
    John Linq
    New Member
    • Total Posts : 12
    • Reward points : 0
    • Joined: 2016/08/31 19:23:12
    • Location: 0
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/14 19:28:24 (permalink)
    0

    Thanks for all the replies.


    > Only 1K? Yuk! Can you move chip?

    I usually use PIC16F1936/1939, and sometimes PIC16F1786.
    I checked several PIC MCU datasheets, and confirmed the maximum recommended impedance for analog sources is usually 10k.
    PIC16F19176 is only 1k.

    (I am a programmer, don't know much about electronic engineering.)
    It seems that, all the impedance for analog sources on our products are more than 10k.
    For ADC2 of PIC16F19176 Silicon Revision A1 on our board, the measured voltage is always lower than the actual voltage.
    the higher voltage, the higher loss.

    PIC16F19176 has three advantages, that is why I finally agree to use PIC16F19176 on my new project.
    1. Two UART.
    2. 12-bit ADC.
    3. Doubled RAM/Flash

    > Your sales engineer or FAE can get a lead time and order samples of a specific new revision.

    The Sales and FAE of our distributor push us to use PIC16F19176.
    They said PIC16F19176 is very cheap, cheaper than PIC16F1936, and compatible with PIC16F1936,
    they will surely provide all the needed support.

    I DON'T BELIEVE THEM, but have to use PIC16F19176.

    > Did you check if increasing the TAD to 4us resolves the issue on A1?

    I increased the TAD by changing ADACQL/ADACQH and also software __delay_us().
    But the improvement is almost none.

    > The items not appearing in Errata for newer silicon means that the problem has not been found in newer silicon.

    Some silicon issues are fixed, so only revision A1 is affected, revision A3 is not.
    But the mentioned [Program Flash Memory (PFM) Endurance] and [ADC2 Channel Switching] are just simply removed, that is why I can't understand.




    #5
    qhb
    Superb Member
    • Total Posts : 9619
    • Reward points : 0
    • Joined: 2016/06/05 14:55:32
    • Location: One step ahead...
    • Status: online
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/15 05:00:10 (permalink)
    0
    John Linq
    ...
    > Did you check if increasing the TAD to 4us resolves the issue on A1?

    I increased the TAD by changing ADACQL/ADACQH and also software __delay_us().
    But the improvement is almost none.

    That is NOT increasing TAD.
     
    #6
    mlp
    boots too small
    • Total Posts : 709
    • Reward points : 0
    • Joined: 2012/09/10 15:12:07
    • Location: previously Microchip XC8 team
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/15 08:08:46 (permalink)
    0
    John Linq
    Some silicon issues are fixed, so only revision A1 is affected, revision A3 is not.
    But the mentioned [Program Flash Memory (PFM) Endurance] and [ADC2 Channel Switching] are just simply removed, that is why I can't understand.

    Then talk to your FAE, or someone else sufficiently-well-connected inside Microchip. We (outsiders) don't know anything beyond what has been published.

    Mark (this opinion available for hire)
    #7
    coffee critic
    Super Member
    • Total Posts : 350
    • Reward points : 0
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/15 09:42:58 (permalink)
    +1 (1)
    As I said in my earlier post you need to increase the TAD not the acquisition time.  The longer TAD time will allow the sample and hold cap to recover some of the charge lost during the glitch.  Any improvement in the reading should tell you if the issue you are seeing is related to the errata.  If not you have a different problem.    

    n_*$
    #8
    John Linq
    New Member
    • Total Posts : 12
    • Reward points : 0
    • Joined: 2016/08/31 19:23:12
    • Location: 0
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/15 19:12:15 (permalink)
    0

    Thanks for all the replies.


    > That is NOT increasing TAD.

    Now, it is confirmed that I made some mistake.

    > We (outsiders) don't know anything beyond what has been published.

    I can understand this situation; but still want to get some experiences/knowledge from experts.
    Do you think PIC16F19176 Revision A1 is ready for production use?
    Do you think PIC16F19176 Revision A3 is ready for production use?
    When will you start to use a new designed MCU, Revision A5 or higher?
    (I am not a native English speaker, so....)

    Our procurement staff just told me that, Microchip doesn't have PIC16F19176 Revision A3 at the moment.

    > you need to increase the TAD.
    > Any improvement in the reading should tell you
    > if the issue you are seeing is related to the errata.
    > If not you have a different problem.

    I will try to increase the TAD. Thanks. Hope it is not a different problem.



    #9
    NKurzman
    A Guy on the Net
    • Total Posts : 17129
    • Reward points : 0
    • Joined: 2008/01/16 19:33:48
    • Location: 0
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/15 19:36:37 (permalink)
    0
    All of those Revs ARE Production Parts.  They have Errata as Published.  If the Errata does affect your design or the work around are not acceptable, Then you can't use them.  If you must use a Newer Rev Part, then you MUST check with You Parts Supplier to Insure you can get the Rev you need.
    If none of that is acceptable for you, Then you must choose a different part.
    #10
    John Linq
    New Member
    • Total Posts : 12
    • Reward points : 0
    • Joined: 2016/08/31 19:23:12
    • Location: 0
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/15 20:22:45 (permalink)
    0
     
    Thanks for the reply.
     
     
    I changed the ADCLK, and did get some improvement.
     
    (Input voltage is fixed.)
     
    ADCLK => ADC Value
    0x03 => 0xAF7
    0x05 => 0xAFF
    0x07 => 0xB01
    0x0F => 0xB01
    0x1F => 0xB01
    Change FOSC from 32MHz to 16MHz
    0x1F => 0xB01


     
    post edited by John Linq - 2019/04/15 22:22:10
    #11
    John Linq
    New Member
    • Total Posts : 12
    • Reward points : 0
    • Joined: 2016/08/31 19:23:12
    • Location: 0
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/15 22:17:52 (permalink)
    0

    Correction:

    Microchip only produced a few PIC16F19176 Revision A3;
    Thus Microchip Taiwan doesn't have PIC16F19176 Revision A3 at the moment;
    Microchip Taiwan would transfer PIC16F19176 Revision A3 from Microchip HQ,
    only if customers purchase a large volume.
     
     
    #12
    John Linq
    New Member
    • Total Posts : 12
    • Reward points : 0
    • Joined: 2016/08/31 19:23:12
    • Location: 0
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/16 00:41:38 (permalink)
    0
     
    This Table
    ADC CLOCK PERIOD (TAD) VS. DEVICE OPERATING FREQUENCIES
    is very different from the previous ones.
     
    My original configuration for TAD violates the minimum required TAD time.
    #13
    John Linq
    New Member
    • Total Posts : 12
    • Reward points : 0
    • Joined: 2016/08/31 19:23:12
    • Location: 0
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/16 23:30:43 (permalink)
    0
    I changed the ADCLK (TAD), and did get some improvement.
    However, the measured voltage is still lower than the actual voltage.
    I need Revision A3, but can't get it.
    #14
    coffee critic
    Super Member
    • Total Posts : 350
    • Reward points : 0
    • Status: offline
    Re: PIC16F19176 Family Silicon Errata and Silicon Revision A3 2019/04/17 10:12:57 (permalink)
    0
    Your best option is to contact your local Microchip sales office (not distribution) and discuss directly.  FAE or FSE  can usually request samples of a specific revision for development.  If nothing else they can get you a lead time.  It will help them if you are some what flexible on temperature rating and package type.     

    n_*$
    #15
    Jump to:
    © 2019 APG vNext Commercial Version 4.5