Shop OBEX P1 Docs P2 Docs Learn Events
Prop stopped booting? — Parallax Forums

Prop stopped booting?

eagletalontimeagletalontim Posts: 1,399
edited 2013-11-06 14:44 in Propeller 1
I have been using the Prop in my car for over a year now and this morning, I started my car and noticed the Prop would not boot up. I did forget to connect my removable power supply this morning, so I shut the car off, then connected the power supply and restarted the car. Still nothing :( I have a 2x16 parallax serial display, a 7 segment display, and a few other components that are connected to the Prop for sensor readings. The display nor the 7 segment display had any output except for the cursor in the top left of the 2x16 display. I attempted to reprogram the EEPROM and the Prop with the exact same code which showed successful according to the Parallax Prop Tool. Nothing seemed to work. Since some of the pins of the prop have a low voltage coming in from sensors, could this have possibly fried the Prop? The current is limited below the ratings of the pin before it hits the prop pin. Is there a maximum number of times the EEPROM can be read or the Prop can boot up? If the Prop Tool says everything was programmed but the prop still has no output, what could be a possible problem there? Please remember, this has been working for over 1 year and all that was changed was me accidentally turning the vehicle on without the power supply hooked to the Prop.

Comments

  • Mike GreenMike Green Posts: 23,101
    edited 2013-11-04 06:48
    There's no maximum times the Prop can boot up and no maximum times the EEPROM can be read. What's changed is that the power supply was not hooked up to the Prop when the vehicle was turned on. Microcontrollers are very very sensitive to voltages being connected to I/O pins without power being applied to the power pins. This is true for any digital chips and most analog chips not specifically designed for this. The protective diodes on the I/O pins are rated for only 50uA and voltages in excess of 4V (for a 3.3V part like the Propeller) can be instantly fatal. How do you have the Propeller protected from the sensors?
  • eagletalontimeagletalontim Posts: 1,399
    edited 2013-11-04 07:14
    Each pin's circuit is different since some sensors are 5V output, and some are 3.3V. Any 5V output from the sensor is separated from the prop with a 4.7K resistor. If it is a 3.3V input, it is separated with a 150ohm resistor. The RPM is a 5V square wave and is put through a 470K and 150K voltage divider circuit before it hits the prop. That is the only sensor that does not go through an external IC chip.
  • Mike GreenMike Green Posts: 23,101
    edited 2013-11-04 07:29
    If the 3.3V sensors are powered and the Propeller is not, you can still get into trouble because Vdd for the Propeller is 0V and the input voltage is 3.3V. That will burn out the protective diodes and may damage other Prop-internal circuitry as well. 5V sensors can also burn out the protective diodes since about 1mA can flow through the 4.7K resistors when Vdd is 0V and the diodes are only rated for 500uA. The 4.7K resistors limit the current to 500uA only when the Prop's Vdd is greater than 2.5V.

    Basically, these measures protect the Prop adequately only when the Prop is powered. You need more protection if the Prop can possibly be unpowered when its inputs are powered.
  • eagletalontimeagletalontim Posts: 1,399
    edited 2013-11-04 07:34
    Would using a different / higher resistor value be the proper way to protect the prop pins in case of a power failure to the prop? On the final circuit, the power supply is not removable.
  • eagletalontimeagletalontim Posts: 1,399
    edited 2013-11-04 10:58
    Also, since the power supply was not hooked up, no sensor IC's were powered so the only input was the RPM signal.
  • eagletalontimeagletalontim Posts: 1,399
    edited 2013-11-04 17:08
    Ok, I am finally back home and am doing some testing. So far, I have found that the Protoboard is getting the correct power and that the Prop is actually sending a signal to the EEPROM on reboot (Pressing the Reset button). Still don't have an output from any pin yet. Will attempt to program an unused pin to go HIGH and see if that works. I would hate to have to unsolder the prop if it is not dead. Are there any other test that I can do before declaring the prop dead?
  • kwinnkwinn Posts: 8,697
    edited 2013-11-04 17:38
    You could write and download a simple spin program that toggles each I/O pin in sequence and use a scope, meter, or led/resistor to see if any of them are working.
  • jmgjmg Posts: 15,173
    edited 2013-11-04 18:22
    eagletalontim wrote: »
    Are there any other test that I can do before declaring the prop dead?

    Are you able to read the EEPROM ?
  • eagletalontimeagletalontim Posts: 1,399
    edited 2013-11-04 18:28
    I hooked my O-scope up to the eeprom and was able to get a signal in and out of the prop. I just decided to replace the prop so I removed the old one and put a new one on and now it is working. Now, come to find out, my ADC may have been the issue. I am checking this out right now. Not sure if it got an input greater than 5V to the ADV which caused it to fail and maybe dump excessive power back to the prop? Not quite sure just yet.
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2013-11-04 18:39
    eagletalontim wrote: »
    I hooked my O-scope up to the eeprom and was able to get a signal in and out of the prop. I just decided to replace the prop so I removed the old one and put a new one on and now it is working. Now, come to find out, my ADC may have been the issue. I am checking this out right now. Not sure if it got an input greater than 5V to the ADV which caused it to fail and maybe dump excessive power back to the prop? Not quite sure just yet.

    The 5V will leak back into the 3.3V rail and depending upon the current sourced and used it can raise the 3.3V rail quite a bit. During reset the current used drops to a very low level and in this condition the supply volts shoots up until it starts to play havoc with the Prop's substrate. I have had issues where this has happened and it stopped the Prop from resetting although it was not damaged. Either way just limit the current that can flow or sometimes I just put a very efficient "Zener" across the 3.3V supply by using two green LEDs in series as they will trip at 3.6V and not before (3.3V Zeners are poor clamps).
  • eagletalontimeagletalontim Posts: 1,399
    edited 2013-11-04 18:56
    When the power supply was removed, the 5V from the ADC input was being driven by the vehicle's ECU. On the ADC, I have the Vin wired directly to the ECU and the VCC wired to the 5V rail of the prop protoboard. The Vref of the ADC is tied directly to the 5V rail on the protoboard as well. Should the Vref or the Vin be wired differently to prevent this from happening again? The output of the ECU never exceeds 5.06V which is within specs of the ADC08831 Vin range. Maybe the Vin at 5V from the ECU and the ADC Vcc at 0V caused the ADC to be damaged and when I hooked up the power supply board to the protoboard, the ADC dumped all 5V to the Prop pin and fried it?
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2013-11-04 19:03
    eagletalontim wrote: »
    When the power supply was removed, the 5V from the ADC input was being driven by the vehicle's ECU. On the ADC, I have the Vin wired directly to the ECU and the VCC wired to the 5V rail of the prop protoboard. The Vref of the ADC is tied directly to the 5V rail on the protoboard as well. Should the Vref or the Vin be wired differently to prevent this from happening again? The output of the ECU never exceeds 5.06V which is within specs of the ADC08831 Vin range. Maybe the Vin at 5V from the ECU and the ADC Vcc at 0V caused the ADC to be damaged and when I hooked up the power supply board to the protoboard, the ADC dumped all 5V to the Prop pin and fried it?

    It's not really clear how you've hooked it up. Can you switch to "graphics mode" vs "text mode" as I am sure it will put us in the picture. If you only have rough drawings then just CamScan (or photo) that and post it.
  • eagletalontimeagletalontim Posts: 1,399
    edited 2013-11-04 19:28
    Ok, here is a schematic of how I had it hooked up. I have now switched the 4.7K resistor with a 10K resistor to hopefully prevent this from damaging the Prop, but I am not sure if it is still enough. That is the last thing I need is for the board to get fried while turned off.....
    ADC-Failed.jpg
    1024 x 409 - 51K
  • eagletalontimeagletalontim Posts: 1,399
    edited 2013-11-04 20:01
    Did the ADC fry because the VCC was 0V and the Vin was somewhere between 1V to 5V?
  • jmgjmg Posts: 15,173
    edited 2013-11-04 20:48
    eagletalontim wrote: »
    Did the ADC fry because the VCC was 0V and the Vin was somewhere between 1V to 5V?

    Has the ADC also failed ?
    Another failure mechanism in vehicles is ESD events.

    The currents possible depends on the ECO output impedance, usually a good idea is a RC filter on the ADC in (10k/10n+), to limit both Power-off flows, and ESD currents too.
  • eagletalontimeagletalontim Posts: 1,399
    edited 2013-11-05 18:33
    I read somewhere in the spec sheet of the ADC08831 that adding additional capacitance to the Vin could cause the reading to be inaccurate and cause issues. This is why there is no additional RC filter on the Vin. There is a decoupling cap on the ADC's VCC and Ground pins though.

    Come to find out in the end, the flux that I used when soldering the prop to the protoboard was not the correct flux....yeah, i cheaped out and found some plumbing flux laying around. Apparently, this flux has some resistance to it and was "shorting" a few pins causing the prop to not properly boot up. When I soldered on the new prop, there must have been some left over flux from the previous time and it also "shorted" some pins. I used my hot air station to heat only the flux up and blow it away from the pins. It has helped and allowed me to get it working again. I think it is time to invest in some denatured alcohol and some proper flux. I do now also have some solder paste which I will use next time. Probably should not cheap out on the flux with small components like this. Lesson learned!
  • jmgjmg Posts: 15,173
    edited 2013-11-05 18:47
    eagletalontim wrote: »
    ...yeah, i cheaped out and found some plumbing flux laying around.

    Yikes, plumbing flux !
  • Mike GreenMike Green Posts: 23,101
    edited 2013-11-05 21:13
    Plumbing flux is very corrosive. Make sure to wash away any flux residue or the residue will absorb moisture from the air over time and corrode your PCB traces and solder joints in the area.
  • eagletalontimeagletalontim Posts: 1,399
    edited 2013-11-06 04:29
    Does anyone have a preferred flux pen they like that does not cost an arm and a leg? There are so many of them and I am unsure which to pick. I will be soldering some 0603 resistors as well, but hopefully with an IR toaster if I can locate one that does not have to be modified to solder a board.
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2013-11-06 05:57
    eagletalontim wrote: »
    Does anyone have a preferred flux pen they like that does not cost an arm and a leg? There are so many of them and I am unsure which to pick. I will be soldering some 0603 resistors as well, but hopefully with an IR toaster if I can locate one that does not have to be modified to solder a board.

    You don't really need flux, just the solder paste, and you don't really need an IR oven, I get by quite nicely with a regular cheap toaster oven with top and bottom elements. I preheat to max and set an electronic timer to beep me after 4 minutes. I place the boards on some folded aluminum foil to cushion and avoid hot spots which is placed on the metal tray. The flux does come in handy afterwards for fixing solder bridges. Withdraw the tray gently (don't bump it) after 4 minutes and place on a cooling tray. If you are doing double sided then I've never had a problem doing one side first then when it's cooled I put the other side on and back in the oven for another 4 minutes. But in this case I am even more gentle with removing the tray.

    There are plenty of solders and fluxes around but I don't know which places you can buy from to give you much advice but I use a 10cc flux gel syringe and this solder paste although I use a manual dispensing gun as the thumb plunger gets a bit painful after a while.

    As for 0603 resistors they can get a bit small and fiddly but I find it easier to use quad resistor arrays as they are about the same price as a single resistor yet they are not much bigger than an 0805 and very easy to handle.
  • eagletalontimeagletalontim Posts: 1,399
    edited 2013-11-06 14:15
    I only have one board that has the 0603 resistors and there are 3 of them total on this board. The actual size of the entire board is 7/8" tall by 5/8" wide so not much room for larger components. I was able to solder an 0603 cap on a previous board with a fine tip soldering iron.

    The toaster oven I have bought for testing does not have a temperature setting that I can select 400 degrees. I have no way to read the temperature of the oven, so I am worried that I could accidentally overheat some of the more sensitive components. I payed $10 for the oven at family dollar :p
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2013-11-06 14:44
    eagletalontim wrote: »
    I only have one board that has the 0603 resistors and there are 3 of them total on this board. The actual size of the entire board is 7/8" tall by 5/8" wide so not much room for larger components. I was able to solder an 0603 cap on a previous board with a fine tip soldering iron.

    The toaster oven I have bought for testing does not have a temperature setting that I can select 400 degrees. I have no way to read the temperature of the oven, so I am worried that I could accidentally overheat some of the more sensitive components. I payed $10 for the oven at family dollar :p

    Spend another dollar or two and get an oven thermometer, I have a stainless steel one that I hang on the rack inside so I can see if it's at maximum which seems to take 3 minutes to start to reflow the solder and I leave it in for 4 minutes total. If you shine a light into the oven while it is baking you can observe when reflow starts to occur, it's unmistakable. So i just add up to another minute as this reflow may not be even.

    As for soldering iron tips I find that the small chisel nose tips are much better even with fine pitch components, as long as you have flux, everything will flow smoothly :)

    Here's a quick link I just found that seems to be useful.
Sign In or Register to comment.