• AVR Freaks

Hot!PIC16F1508 ESD Protection for EUSART

Page: 12 > Showing page 1 of 2
Author
JDW
Junior Member
  • Total Posts : 101
  • Reward points : 0
  • Joined: 2009/07/23 23:46:02
  • Location: Japan
  • Status: offline
2019/08/28 01:29:56 (permalink)
0

PIC16F1508 ESD Protection for EUSART

I've programmed a 16F1508 (Vdd=3.3v) to communicate via EUSART with a fingerprint sensor module. Communication is logic level (HI=3.3V, LO=0V), 8-bit, Asynchronous, No Parity, 1 Start bit, 1 Stop Bit, 9600 baud. However, I've learned I need to move the sensor module off the main circuit board using a 1.5 meter cable. The PIC controller and external fingerprint sensor will be used in an automotive environment. I am not well versed in ESD protection and therefore could use advice from those of you who have practical experience. This sort of protection is important since the sensor will be attached via connector and cable length of 1.5m.

I have been considering use of either of the following two ESD Protection devices:
  • LittleFuse SP3014-02UTG (NOTE: This part would NOT connect to my 3.3v voltage rail)
  • Wurth 824011
Below/attached are 2 simple examples of how I think the ESD protection parts might be used in my circuit (although I am not sure if 360-ohm is best for the inline resistance).
 
NOTE: There will be more capacitance on my 3.3V that is shown above. I will have other 0.1uF ceramic caps here and there, along with two 47uF aluminum electrolytic low ESR (140m-ohm) caps as well.
 
My concern about using the Wurth part (or any of that design) is that an ESD surge would potentially lift my Vdd above the Absolute Maximum Rating of the 16F1508. With the LittleFuse part, that won't happen, but I am curious what the caveats are to NOT having its Zener tied to Vdd. Below/attached is the Maximum Ratings of the 16F1508.
 
post edited by JDW - 2019/08/28 01:31:18

Attached Image(s)

#1

23 Replies Related Threads

    JDW
    Junior Member
    • Total Posts : 101
    • Reward points : 0
    • Joined: 2009/07/23 23:46:02
    • Location: Japan
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/08/28 17:02:37 (permalink)
    0
    "Pretty please"?
    #2
    JDW
    Junior Member
    • Total Posts : 101
    • Reward points : 0
    • Joined: 2009/07/23 23:46:02
    • Location: Japan
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/08/29 18:40:52 (permalink)
    0
    "With sugar on top"??
    #3
    NKurzman
    A Guy on the Net
    • Total Posts : 18852
    • Reward points : 0
    • Joined: 2008/01/16 19:33:48
    • Location: 0
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/08/29 19:31:45 (permalink)
    0
    If you are concerned about the Rail you can add a TVS.  You can also add a feritte bead.
    Those Parts are for user ESD.  can people touch those points in a User environment.  (Or Production if production is not ESD protected.
    I hope your Auto 12V in is well protected.  Auto power can be harsh.
    #4
    JDW
    Junior Member
    • Total Posts : 101
    • Reward points : 0
    • Joined: 2009/07/23 23:46:02
    • Location: Japan
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/08/29 19:38:33 (permalink)
    0
    First of all, thank you for the reply.  I am a bit giddy (and relieved) someone made time to comment!  Thanks!
     
    I am aware of automotive electronics, load dumps, etc. as I have been designing for cars and trucks for years.  But this project is the first where I've needed to have a key component off the main board and therefore I need to think about ESD mainly.  As shown in the attachments in my opening post, a fingerprint sensor will be connected off the main board (which has my PIC and fully filtered switching power supply).  The sensor will attached to the main board via connector.  But human hands will do that, and no doubt some will induce static electricity into that connector, hence the need for ESD protection.  It's not as simple as "just add a TVS" as per the attachments in my previous post.  As you can see there are different approaches to dealing with the ESD.  And although those 2 attachments use regular diodes (in a single part, mind you), I am aware there are solutions that use TVS for ESD.  However, the two approaches shown in my attachments (in my opening post) are the most popular and common -- I must assume for a reason.
     
    And so, I am mainly curious which of the two approaches shown in my attachments is the best and why, keeping in mind I don't want my 3.3V voltage rail to rise too high, for obvious reasons.
     
    Thanks!
    #5
    JDW
    Junior Member
    • Total Posts : 101
    • Reward points : 0
    • Joined: 2009/07/23 23:46:02
    • Location: Japan
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/02 18:11:36 (permalink)
    0
    BUMP.
    #6
    ric
    Super Member
    • Total Posts : 27979
    • Reward points : 0
    • Joined: 2003/11/07 12:41:26
    • Location: Australia, Melbourne
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/02 19:26:39 (permalink)
    0
    Personally, I will never take a PIC signal directly off the board.
    Here's how I generate some "RS232-like" signals using some BRT devices (transistors with built in bias networks).
    Doing this with your "TX" signal could work well, because your USART has a control bit to invert the TX data, which cancels out the inversion caused by the external transistor. (SCKP in BAUDCON bit 4)
     
    Unfortunately the peripheral won't invert the RX signal for you. Some PIC18F devices have a "DTRXP" bit in BAUDCON bit 5 to do it. The PIC16F1947 datasheet mentions a bit by that name, but doesn't actually appear to have it.
     
    Failing that, you could use a a second BRT to reinvert the RX signal. It's not possibly to do the inversion in software, unless you bit-bang the entire RX part of the USART.
     
     
     
     

    Attached Image(s)


    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!
    #7
    ric
    Super Member
    • Total Posts : 27979
    • Reward points : 0
    • Joined: 2003/11/07 12:41:26
    • Location: Australia, Melbourne
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/02 19:31:39 (permalink)
    0
    n.b. If you had an unused CLC with both its input and output pins spare, you could use that as an inverter for the RX signal.

    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!
    #8
    JDW
    Junior Member
    • Total Posts : 101
    • Reward points : 0
    • Joined: 2009/07/23 23:46:02
    • Location: Japan
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/02 19:40:27 (permalink)
    0
    Thanks for your 2 replies, Ric.  I really appreciate it.
     
    Basically you are using an NPN BJT transistor in place of the ESD protection components I mentioned in my opening post.  And just as you show in your schematic, using an NPN with pull-up will invert the signal.  A transistor would not be superior to the ESD devices I mentioned but would be much better than no protection at all.  I will give that some thought.  Thanks again.
     
    Now, if anyone else who understands the ESD protection devices mentioned in my opening post would care to comment about them (please my closing paragraph in that opening post), I am certainly wanting to hear that too.
     
    Many thanks! 
    #9
    mpgmike
    Super Member
    • Total Posts : 448
    • Reward points : 0
    • Joined: 2014/01/23 17:27:06
    • Location: NJ
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/03 07:32:24 (permalink)
    0
    You might want to check out Microchip's Hot Swap devices:
     
    https://www.microchip.com...art.aspx?branchID=9036

    I don't need the world to know my name, but I want to live a life so all my great-grandchildren proudly remember me.
    #10
    JDW
    Junior Member
    • Total Posts : 101
    • Reward points : 0
    • Joined: 2009/07/23 23:46:02
    • Location: Japan
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/03 16:31:15 (permalink)
    0
    Thank you for the information about hot swap devices. However, they are expensive and overkill for my application. But most importantly of all, use of them does not give an engineering explanation to the question I put forth in my opening post, which is what I’m really seeking.

    Thanks.
    #11
    coffee critic
    Super Member
    • Total Posts : 392
    • Reward points : 0
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/04 10:18:43 (permalink)
    +1 (1)
    You could try putting a BAT54S schottky diode pair on each of the UART pins.  These parts are less than $0.02 in volume.  The this device with the series resistor should provide enough protection to keep the voltage with in 0.5V of either rail.   

    n_*$
    #12
    JDW
    Junior Member
    • Total Posts : 101
    • Reward points : 0
    • Joined: 2009/07/23 23:46:02
    • Location: Japan
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/04 17:44:16 (permalink)
    0
    coffee critic,
     
    Thank you for providing a specific part.  But that is actually a THIRD device added to the TWO I mentioned in my opening post that would require analysis.  It is the analysis I seek, not simply a "close my eyes and use what seems best after testing" approach.  I want to understand the theory behind which of the now THREE devices I should chose for my application and why.  
     
    Consider there is no Zener in the BAT54 part you proposed.  There is a Zener in the two mentioned in my opening post.  But one of the two doesn't allow the Zener to touch the main voltage rail, nor do the diodes in that device (LittleFuse) touch the main voltage rail either.
     
    So as I said in my opening post, I'd prefer the ESD device to not screw up my main voltage rail, and it would appear the LittleFuse device mentioned in my opening post would accomplish that, but I am curious as to the theory behind it to know the caveats.  My guess is that it would dump all the ESD energy into the Tx and Rx lines alone, leaving my Vcc voltage rail happy but inflicting certain consequences on my Tx and Rx lines.  It is the subject of those "consequences" I am concerned about and the reason I posted here.
     
    So again, if anyone has experience with these ESD parts and understands the different types and when to use them and when not to, I wish to hear from you!
     
    I must say it is rather depressing.  I actually wrote to Wurth several days ago but they never wrote back.  So I guess even the parts suppliers don't know.  Crazy.
    #13
    PStechPaul
    Super Member
    • Total Posts : 2813
    • Reward points : 0
    • Joined: 2006/06/27 16:11:32
    • Location: Cockeysville, MD, USA
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/04 23:39:09 (permalink)
    0
    Here is the datasheet for the Wurth device:
     
    https://katalog.we-online.de/pbs/datasheet/824011.pdf
     
    I don't have specific experience with either device, but both of them incorporate a TVS diode which clamps at 8-10 volts at currents of 1 to 8 (or 12) amps. The Wurth device has a higher current rating, and I would tend to prefer it to the Littelfuse device because it has a connection to the positive terminal of the TVS, which would likely be connected to Vdd. There would also be capacitance on this terminal, which would probably absorb fast transient surges more quickly than the TVS, although I haven't found a specification for operating speed, other than the test waveforms which are 5x50 nSec or 8x20 uSec.
     
    Ferrite beads are also often used for transient voltage protection and noise immunity. A small bead could be put in series with the TxD and RxD lines for normal mode rejection, and the external cable could have a clamp-on ferrite component for common mode protection.
     
    More info:
     
    https://www.onsemi.com/pub/Collateral/AND8181-D.PDF
     
    https://www.onsemi.com/pub/Collateral/AND8231-D.PDF
     
     

     
    #14
    JDW
    Junior Member
    • Total Posts : 101
    • Reward points : 0
    • Joined: 2009/07/23 23:46:02
    • Location: Japan
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/05 01:13:30 (permalink)
    0
    Thank you for the information, Paul.  Specifically, page 3 of the AND8231-D.PDF provides useful information.  I would not want the TVS approach since the clamping voltage is too high for my PIC.  So Table 2 options would be best.  The poor power rating excludes the Schottky at right in that table. The power rating of the Diode Array isn't quite as good as the Diode Array + TVS, but the clamping voltage is lower with the standard Diode Array, so I suppose I will go with that.  
     
    As to the ferrite bead, I'm not sure which size would be best or even if it would be needed if I use the Diode Array.  I do know that my Tx and Rx lines from the PIC work fine at 9600bps even if I use an inline 470-ohm resistor.  I've read it's best to assist the ESD diode device with an inline resistor, as such would help better protect the PIC I think.  The PIC's internal diode clamp can handle only 20mA according to the datasheet.
    #15
    coffee critic
    Super Member
    • Total Posts : 392
    • Reward points : 0
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/05 09:55:05 (permalink)
    0
    The value that the series schottky diode provides is that the low forward voltage clamps any negative pulse to something less than 0.5V VF of the ESD diode.  The proper connection is to connect pin 1 to Vss pin 2 to the signal pin and pin 3 to Vdd.  You absolutely need to have a current limiting resistor in place or you will blow out the diode on the first ESD hit.  Adding a ferrite bead would also limit the magnitude of the pulse but not necessary the energy. 
     
    You might also want to check how much ESD protection is built into the fingerprint sensor module.  I may already have sufficient protection that all you need is to make sure that the ground between the sensor board and the MPU board is solid.   

    n_*$
    #16
    JDW
    Junior Member
    • Total Posts : 101
    • Reward points : 0
    • Joined: 2009/07/23 23:46:02
    • Location: Japan
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/05 17:15:04 (permalink)
    0
    coffee critic,
     
    Thank you for your reply.  Yes, I understand the Schottky diode (Z1 in "Diode Array Plus TVS" on pg.3 of AND8231-D.PDF & Wurth 824011 in my opening post) has merit, but I must once again repeat the question put forth in my opening post which no one thus far has answered specifically...
     
    Please consider that the Schottky diode can be:
     
    1. Entirely missing, as in the "Diode Array" given on pg.3 of AND8231-D.PDF.
    2. Present with its Cathode connected to Vdd, as in "Diode Array Plus TVS" on pg.3 of AND8231-D.PDF.
    3. Present but NOT connected to Vdd, as shown in the LittleFuse part graphic in my opening post.
    4. Present with its Cathode connected to the Cathode of a diode whose anode connects to Vdd, like the Wurth 82400102. (Please Google that p/n, since links often prevent posts from being accepted in this forum.)
     
    I want to know what ESD part is best for my application and why.  That's really it.  
     
    But my fear about using parts with the Zener connected directly to Vdd is that any ESD surge will result in my Vdd rising above the Absolute Maximum Ratings of my PIC, and that is why I attached a graphic of those ratings in my opening post.  So to eliminate any danger of Vdd rising too high, I am guessing I should probably go with option 1. or 4. (in my list above), but I want to know the electrical caveats of doing that.  Once I am square on that, I can then begin to think about protecting (or not protecting) my fingerprint sensor side.  Therefore, any further thoughts and suggestions on this matter would be greatly appreciated.
     
    Thank you.
    #17
    PStechPaul
    Super Member
    • Total Posts : 2813
    • Reward points : 0
    • Joined: 2006/06/27 16:11:32
    • Location: Cockeysville, MD, USA
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/06 02:00:48 (permalink)
    +1 (1)
    I think there is really very little danger that a transient surge or spike will have enough energy to raise the voltage of Vdd beyond the 6.5 volts absolute maximum, because the capacitors will absorb it. The ESD test waveform, as shown in the application note above, is characterized by 8 kV source on 150 pF through 330 ohms. This is an energy of about 0.005 J. The peak current is 8000/330 = 24 amps, which diminishes to 8 amps in 60 nSec. If the energy is dumped through the Schottky diodes into the 5V bus, and the capacitance is 1 uF, the energy in it will be 12.5 uJ.If the entire amount of energy would be transferred to the 1 uF capacitor, it would reach 100V. A 400 uF capacitor would have 0.005 J at 5 volts. The additional 0.005 J would charge it to about 7 volts. So perhaps there could be a problem, but this is a very extreme case for protection against lightning strikes.
     
    The human body model (HBM) is much less severe, using 100 pF through a 1500 ohm resistor. But the voltage may still be 8 kV, which is still 0.003 J. The peak current is less, 8000/1500 = 5.3 amps.
     
    My analysis does not take into account the energy dissipated in the series resistance, which would be 194 kW peak. The energy is 0.005 watt-seconds, so the time for full dissipation would be in the order of 25 nSec. The voltage on a 1 uF capacitor charging at 24 amps for 25 nSec would be about 25 * 24 / 1000 = 0.6 volts.
     
    It might be worthwhile to do a simulation of this, or perhaps even perform the test with actual hardware. I'm not 100% sure of my calculations, either.

     
    #18
    JDW
    Junior Member
    • Total Posts : 101
    • Reward points : 0
    • Joined: 2009/07/23 23:46:02
    • Location: Japan
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/06 02:35:10 (permalink)
    0
    Thank you for the information, Paul.  I must also consider the fingerprint sensor attached to my PIC, which uses the same 3.3V voltage rail as the PIC.  That sensor had handle up to 6.0V max, a tad less than the PIC.
     
    I will have a couple low ESR aluminum electrolytic 47uF caps on that voltage rail, along with several 0.1uF ceramics, 0.01uF ceramic, and the output capacitor of the power supply will be two 10uF ceramics in parallel.  So there is a fair amount of capacitance.
     
    Lightning strikes are not a consideration as this will be inside a 12V or 24V vehicle.  It's basically the human body model as there will be a cable with connectors at either end.
    #19
    PStechPaul
    Super Member
    • Total Posts : 2813
    • Reward points : 0
    • Joined: 2006/06/27 16:11:32
    • Location: Cockeysville, MD, USA
    • Status: offline
    Re: PIC16F1508 ESD Protection for EUSART 2019/09/07 20:47:30 (permalink)
    +1 (1)
    Here is a simulation, which shows that the two Schottky diodes on the TxD line, one to GND, and one to Vdd, will keep Vdd and TxD under 6 volts. This is with a 1 uF X7R capacitor on Vdd.
     


     
    #20
    Page: 12 > Showing page 1 of 2
    Jump to:
    © 2020 APG vNext Commercial Version 4.5