P2ES Rev B red LED & Could not find a P2 on port /dev/cu.usbserial-...

I had some success with Chip's new Spin2 interpreter and PNut's changes to support Spin2... After running several VGA example programs from PNut_v34a, unplugging from my MacBook and replugging into a host Raspberry Pi for power (and back to the MacBook again for re-uploading changes), I've got the red LED lit up and get the dreadful "Could not find a P2 on port /dev/cu.usbserial-..." message from loadp2 or PNut.

I'm fairly certain that I made an effort to unplug the USB cables before plugging or unplugging the Accessory VGA or HDMI boards, to make sure no power was on the pins. But, who knows...

The board shows up on the MacBook as: cu.usbserial-DN43WMDD in the /dev directory, so I know that the FTDI chip is working.
When I also plug the AUX USB into a 5v power source, I am able to use PNut or loadp2 to upload programs. The red LED goes OFF.
Note: the jumpers for LDO VIN 5v & VDD 1V8 are in-place. The jump for ACC HDR 5V is off.

So, what component is most likely blown (voltage regulator? something on the P2 chip itself?)? How do I check? Where to look for voltage, etc...?

Thanks for any insight!
dgately
«1

Comments

  • VonSzarvasVonSzarvas Posts: 1,869
    edited 2020-02-12 - 20:22:05
    Here's an image of the suspect parts.

    U401 is FTDI : FT231XS-R
    U601 is ON Semi : NCP380HMUAJAATBG

    Do you see the Green ON LED light up when connecting the PC USB port ?
    If yes, then likely U601 output mosfet has blown
    If no, it's possible the issue is one of the FTDI chip outputs, which could be tested further or possibly worked-around by an FTDI setting. (or by rebooting the PC, as rogloh points out)


    I'm assuming here that you can upload and run programs when connecting both USB cables?

    Another possible situation is that one of the P2 I/O groups is pulling down one of the rails. You could check that by powering by both USB's, then measuring all the 3.3V outputs (one per edge header), to ensure they are all very close to 3.3V. And whilst your measuring, check the voltage between the 5V_ACC jumper and one of the test posts whilst powering from PC USB, then from AUX USB.



    481 x 537 - 35K
  • dgately, also try rebooting your Mac, just in case. I've had several USB problems on my Mac with FTDI at various times and a reboot often fixed it. But your problem may be different...
  • dgatelydgately Posts: 1,200
    edited 2020-02-12 - 22:21:30
    VonSzarvas wrote: »
    Here's an image of the suspect parts.

    U401 is FTDI : FT231XS-R
    U601 is ON Semi : NCP380HMUAJAATBG

    Do you see the Green ON LED light up when connecting the PC USB port ?
    If yes, then likely U601 output mosfet has blown
    If no, it's possible the issue is one of the FTDI chip outputs, which could be tested further or possibly worked-around by an FTDI setting. (or by rebooting the PC, as rogloh points out)

    I'm assuming here that you can upload and run programs when connecting both USB cables?
    VonSzarvas, rogloh... Yes, rebooted my MacBook... same result!

    Yes, when I plug a cable from the P2's PC USB to my MacBook (or any 5v source), the blue LED lights up, dimly for about 4 seconds, then the green & red LEDs light up, as the blue LED goes dark. If I then plug a 5v source into the AUX USB, the red LED goes dark and the green LED remains lit (and the P2 becomes loadable, once again).

    Yes, when both USB ports are connected (ACC to a 5v source, PC USB to a Mac or RaspberryPi), I can upload P2 binaries without problems.
    Another possible situation is that one of the P2 I/O groups is pulling down one of the rails. You could check that by powering by both USB's, then measuring all the 3.3V outputs (one per edge header), to ensure they are all very close to 3.3V. And whilst your measuring, check the voltage between the 5V_ACC jumper and one of the test posts whilst powering from PC USB, then from AUX USB.
    I'll attempt these tests now!

    Thanks,
    dgately

    LEDs.jpg
    288 x 370 - 44K
  • I'd try with the VGA board removed... I think someone had an issue with their VGA board drawing too much power...
  • PublisonPublison Posts: 11,287
    edited 2020-02-12 - 22:12:52
    Rayman wrote: »
    I'd try with the VGA board removed... I think someone had an issue with their VGA board drawing too much power...

    That was probably @Reinhard that had a problem with the audio amp causing the problems with drawing too much power. I believe it was determined the amp was defective.

    https://forums.parallax.com/discussion/171131/question-about-a-v-board-solved#latest
  • Another possible situation is that one of the P2 I/O groups is pulling down one of the rails. You could check that by powering by both USB's, then measuring all the 3.3V outputs (one per edge header), to ensure they are all very close to 3.3V. And whilst your measuring, check the voltage between the 5V_ACC jumper and one of the test posts whilst powering from PC USB, then from AUX USB.
    3.3v:
    With both USB's powered, I see 3.28 (+ or - .02v) on every 3.3v output on the edge headers
    With just the AUX USB powered, I see basically the same as above, 3.28...
    With just the PC USB powered, I see negative readings of around -1.4v

    5v:
    With both USB's powered, I see 5.1v at the righthand ACC HDR 5v jumper pin when ground probe is at the ground test point. When I test across the jumper pins, I get a voltage reading, starting from 3.5v and steadily decreasing, to less than 2v.
    With just the AUX USB powered, I see 5.1v at the righthand ACC HDR 5v jumper pin when ground probe is at the ground test point. When I test across the jumper pins, I get a voltage reading, starting from 1.2v and steadily decreasing, to less than .6v.
    With just the PC USB powered, I see .803v at the righthand ACC HDR 5v jumper pin when ground probe is at the ground test point. When I test across the jumper pins, I get a voltage reading, starting from -2v and steadily increasing, to 1.2v.
  • Rayman wrote: »
    I'd try with the VGA board removed... I think someone had an issue with their VGA board drawing too much power...
    Still had the problem after all accessory boards removed.

  • evanhevanh Posts: 8,965
    edited 2020-02-12 - 22:47:40
    dgately wrote: »
    ...
    With just the PC USB powered, I see .803v at the righthand ACC HDR 5v jumper pin when ground probe is at the ground test point.
    Same fault as the last person had. An I/O pin group has gone short. The LDO for that group is working its guts out fighting the short and this trips the PC-USB 500 mA limit, causing the red fault indicator and low volt readings. Whereas AUX-USB is stronger and can overpower the struggling LDO so that everything else, except the shorted pin group, comes up fine.

    Only question now is, which I/O pin group has gone short?

  • evanh wrote: »
    Only question now is, which I/O pin group has gone short?

    Except @dgately already measured all the Vxxx pins and got 3.3V.

    One last place to check for healthy 3.3V is from the Common rail. That can most easily be measured from the two flash-IC pins highlighted in red. Again, keep the black probe on one of the ground posts.


    Another test... If you have the Accessory Boards, you could run the LED matrix test code to visually check all the I/Os are working.

    https://github.com/parallaxinc/propeller/blob/master/resources/Accessory Test Code/test_LEDMATRIX.spin2

    Set the basepin at line 8 to whichever header you are testing/plugging into.
    Consider swapping the commented lines at 50 and 51 to keep the leds lit for quicker testing.



    This is an interesting situation, as the fault LED operating suggests U601 is working as it should. The voltage readings follow what @evanh explained, which might lead to a P2 issue. But your 3.3V readings suggest all the P2 groups are fine. The common rail or LED test might give a final clue. Otherwise, based on current results, I'm leaning toward U601 needing a swap-out. Without searching for the datasheet on that part, there's probably an output mosfet that's blown, with some current still passing through a body diode.
    960 x 533 - 65K
  • I have a guess. VGA board is identical to A / V board. I also had a problem with that. Most likely the amplifier is drawing too much current. I will only get the final proof when the new board has arrived. That should be the case in a few days.
    When the board is unplugged, a USB connection should work and all voltages should be ok.(?) is it?
  • dgately wrote: »
    Rayman wrote: »
    I'd try with the VGA board removed... I think someone had an issue with their VGA board drawing too much power...
    Still had the problem after all accessory boards removed.

    oh sorry I see that to late.
    can you measure the current consumption and compare this with an "working" P2ES ?
  • ReinhardReinhard Posts: 384
    edited 2020-02-13 - 12:12:44
    As reference:
    I remove the Jumper LDO VIN 5V and measure here the current, without any plug in boards: 1.25mA
  • VonSzarvas wrote: »
    One last place to check for healthy 3.3V is from the Common rail. That can most easily be measured from the two flash-IC pins highlighted in red. Again, keep the black probe on one of the ground posts.
    Ah, right, the "Common_VIO". That's a ninth LDO. Yes, a short on that rail would also give the same effect.

  • VonSzarvas wrote: »
    One last place to check for healthy 3.3V is from the Common rail. That can most easily be measured from the two flash-IC pins highlighted in red. Again, keep the black probe on one of the ground posts.
    1. With only the PC USB powered, I see around 21.0mv (yes, millivolts)...
    2. With both PC USB & AUX USB powered, I see 3.28v
    3. With just AUX USB powered, I see 3.28v, as expected
    4. If after having both USB's powered, I remove the AUX USB power, I see 1.6v on those pins. A reset, sets them back to the millivolt reading.
    Another test... If you have the Accessory Boards, you could run the LED matrix test code to visually check all the I/Os are working.

    https://github.com/parallaxinc/propeller/blob/master/resources/Accessory Test Code/test_LEDMATRIX.spin2

    Set the basepin at line 8 to whichever header you are testing/plugging into.
    Consider swapping the commented lines at 50 and 51 to keep the leds lit for quicker testing.
    Yes, I had already done that test yesterday... The LED matrix lights up pixels as it should. Of course, when the LED matrix is plugged into P56-P63, there are two rows that are not consistent (P62 & P63 are obviously rx & tx to the Terminal). I expected that outcome. Obviously, P62 & P63 are working as I'm able to load programs and converse serially with the P2.
    This is an interesting situation, as the fault LED operating suggests U601 is working as it should. The voltage readings follow what @evanh explained, which might lead to a P2 issue. But your 3.3V readings suggest all the P2 groups are fine. The common rail or LED test might give a final clue. Otherwise, based on current results, I'm leaning toward U601 needing a swap-out. Without searching for the datasheet on that part, there's probably an output mosfet that's blown, with some current still passing through a body diode.
    U601's location is going to make swapping it out a difficult process for me :neutral: At least I'm able to load programs onto the FLASH chip and then run off of the AUX USB power. It's not a total loss, just a slight inconvenience. I'm more interested in the "Why it happened?".

  • Reinhard wrote: »
    As reference:
    I remove the Jumper LDO VIN 5V and measure here the current, without any plug in boards: 1.25mA
    1. With PC USB powered, I see 0.0mA
    2. With just the AUX USB powered, I see 1.31mA
    3. With PC USB & AUX USB powered, I see 14.38mA
  • With all three jumpers (ACC, LDO, VDD) removed, does powering up from PC-USB still give a red LED?
  • evanh wrote: »
    With all three jumpers (ACC, LDO, VDD) removed, does powering up from PC-USB still give a red LED?
    Yes, well... After the board has been sitting without any power (hours since last powered), powering only the PC USB results in only the blue LED being lit, for about 30 seconds. At that point I powered the AUX USB to get the green LED lit. Unplugging the AUX USB and the red LED lit up. Unplugging and powering just the PC USB results in the red LED being lit.

    Further attempts at powering the PC USB without the jumpers results in the red LED lighting up.



  • That's odd. The red LED shouldn't come on without the green LED. The green LED indicates the FT321 is good to go and has enabled the USB switch (U601). The red LED indicates the USB switch is enabled but in fault shutdown.
  • evanh wrote: »
    That's odd. The red LED shouldn't come on without the green LED. The green LED indicates the FT321 is good to go and has enabled the USB switch (U601). The red LED indicates the USB switch is enabled but in fault shutdown.

    Indeed. Without the green LED the PC has not approved the connection and may not be enabling power to the output of more than a few mA's, depending on design of the PC. (Some don't bother with USB power controls).

    Have you tried a different computer or at least different USB port during your diagnostics ? And a new USB cable ?

    There could be other reasons for the Green LED failing to light up, such as bad FTDI chip or a short/resistance on the 5V rail which drops voltage so much that the Green LED can't light.

    If you are familiar with using FT_Prog, you could make a little change:

    CBUS3 : Set to TriState (Disables USB power, so use AUX for power and USB for programming).

    CBUS3 : Set to Drive Low (Force USB power on, handy to test if U601 is working and also help determine if the PC or FTDI chip is the issue). Note that the PC may still be restricting the amount of current that can be sourced- using an external powered hub would get around that.

    CBUS3 : Set to PWREN# (Revert to factory setting!)


    There's still a few variables here to say "what caused it", but for now probably the best bet to set CBUS3 as Tristate, and carry on using your board this way if trying a new usb cable and different PC (or externally powered hub) don't help.
  • dgatelydgately Posts: 1,200
    edited 2020-02-14 - 16:25:18
    evanh wrote: »
    That's odd. The red LED shouldn't come on without the green LED. The green LED indicates the FT321 is good to go and has enabled the USB switch (U601). The red LED indicates the USB switch is enabled but in fault shutdown.
    Sorry! I should have said that the green LED is lit when the red LED is lit (tunnel vision in my thinking about just the red & blue LEDs)...

    And, I've tested multiple cables, different USB ports, a USB adapter with multiple ports, etc...

    I will test on my TeamCity Builder iMac in a moment, as well!

  • dgately wrote: »
    Reinhard wrote: »
    As reference:
    I remove the Jumper LDO VIN 5V and measure here the current, without any plug in boards: 1.25mA
    1. With PC USB powered, I see 0.0mA
    2. With just the AUX USB powered, I see 1.31mA
    3. With PC USB & AUX USB powered, I see 14.38mA

    This is very,very strange. Did you a measure like this.
    Remove the LDO VIN 5V Jumper and instead the jumper you have an ampmeter?
    Powered with PC USB.
    1024 x 683 - 410K
  • Reinhard wrote: »
    dgately wrote: »
    Reinhard wrote: »
    As reference:
    I remove the Jumper LDO VIN 5V and measure here the current, without any plug in boards: 1.25mA
    1. With PC USB powered, I see 0.0mA
    2. With just the AUX USB powered, I see 1.31mA
    3. With PC USB & AUX USB powered, I see 14.38mA

    This is very,very strange. Did you a measure like this.
    Remove the LDO VIN 5V Jumper and instead the jumper you have an ampmeter?
    Powered with PC USB.
    Yes...
    current.jpg
    2271 x 1150 - 852K
  • dgately wrote: »
    evanh wrote: »
    That's odd. The red LED shouldn't come on without the green LED. The green LED indicates the FT321 is good to go and has enabled the USB switch (U601). The red LED indicates the USB switch is enabled but in fault shutdown.
    Sorry! I should have said that the green LED is lit when the red LED is lit (tunnel vision in my thinking about just the red & blue LEDs)...

    And, I've tested multiple cables, different USB ports, a USB adapter with multiple ports, etc...

    I will test on my TeamCity Builder iMac in a moment, as well!
    Okay, that's good. Next thing to check is the PC-USB 5 Volts I guess. Probably the best place to measure that is at C401, it's the most exposed capacitor right behind the PC-USB socket.

    It should measure a full 5 Volts.

  • Whoa! I just read this thread. Any indication what failed? Does the P2 seem to be okay? Did something else fail?
  • Prop2 seems fine. He can still program it when powered from AUX-USB (Both USB cables connected).
  • cgracey wrote: »
    Any indication what failed?
    Possibly the USB power switch IC (U601). Just trying to narrow that down, as to whether there is any current drawn through it. I suspect it has no strength so is not delivering.

  • evanh wrote: »
    cgracey wrote: »
    Any indication what failed?
    Possibly the USB power switch IC (U601). Just trying to narrow that down, as to whether there is any current drawn through it. I suspect it has no strength so is not delivering.

    Thanks, Evanh.
  • dgatelydgately Posts: 1,200
    edited 2020-02-15 - 13:53:20
    evanh wrote: »
    Okay, that's good. Next thing to check is the PC-USB 5 Volts I guess. Probably the best place to measure that is at C401, it's the most exposed capacitor right behind the PC-USB socket.

    It should measure a full 5 Volts.
    I get 5v at the center of the red circle in this image:
    P2_RevB_USB-Power-SectionC401.png
  • Seems certain U601 output fet has blown. That's an ON Semi part number : NCP380HMUAJAATBG

    Need to change that part, or as a workaround, set CBUS3 to TriState to disable the use of U601, and use the Aux-USB for power and PC-USB for programming.

    If you would like to send the board back to get that U601 swapped, please email your order number to support@parallax.com and they will get you an RMA reference to return the faulty board. Repair turn around should be pretty quick- maybe 1-2 days.
  • Thanks all... I'm glad we've got a good idea of what component was affected. Wish I knew what action caused the problem (I had the ACC 5v jumper cap remove, so I know it wasn't a short to the header pin's 5v).

    No hurry as I am able to program and run using the PC and AUX, store the program to FLASH and just power through the AUX for my tests.

    I'll check into support@parallax.com when I get a chance!

    dgately
Sign In or Register to comment.