Shop OBEX P1 Docs P2 Docs Learn Events
Random Propeller failure? — Parallax Forums

Random Propeller failure?

ThricThric Posts: 109
edited 2012-01-25 22:46 in Propeller 1
So I was working with my prop this afternoon and for some reason it decided to just stop working. It was functioning properly for weeks, but it seems as if it changed its mind.

Here's what I went through:
At first I thought it was just a low battery, so I switched that out (9 volt battery). That didn't do anything.

I thought that maybe I broke a lead off my crystal (as it has happened before), so I tested the connection with a multimeter and then tried downloading a program that would light an LED without a clock set. But now nothing will download and the crystal is secure.

When I connect my programming cable the propeller tool recognizes the com port but if doesn't find the propeller connected to it.

I've tested the voltage on my board and its regulating at 3.3volts, which is confirmed by my GPS (which is fine) and power led.

I'm on a custom designed board which has been working properly for a while now, so I'm clueless on what went wrong.
The most recent thing that I did to my board was replace my series one xbee module for a xbee pro 900 module. And I've been turning that on and off for the past few days so it isn't that I just added it and it stopped. I've even left no xbee module in place and it still won't function properly :(

Is there something else that I can check?
I'm testing to see if it works by: 1- seeing if the pre-programmed (EEPROM) will flash my led and 2- see if I can download a new program.
Help is greatly appreciated as I'm starting to panic >.<

If you need to know more, I'll gladly reply!

Comments

  • SONIC the HedgehogSONIC the Hedgehog Posts: 321
    edited 2012-01-24 18:22
    I bet it's your EEPROM. The first time I did mine and it worked, I needed to move my EEPROM, and a pin was bent out. I checked for it and the power led was on, but the spin tool didn't recognize the propeller, it said no propeller chip found.
  • SONIC the HedgehogSONIC the Hedgehog Posts: 321
    edited 2012-01-24 18:25
    And I'm not positive, I haven had mine a month, but I did find out that the spin tool does not find the propeller if you don't have the EEPROM. What is the size of it? Maybe if it's like more than a mega byte the propeller doesn't recognize it or something( not like I would know, but never hurts to try.....)
  • SONIC the HedgehogSONIC the Hedgehog Posts: 321
    edited 2012-01-24 18:45
    If you did blow a propeller, then they sell them at fry's for 12.99 DIP, but I didn't see any of the other packages though.
  • ThricThric Posts: 109
    edited 2012-01-24 18:49
    Well the EEPROM I'm using is a 64kbyte and I've been able to program to it before everything decided not too. Also I'm using SMD so no pin is bent that I can tell. So I think thats one to list to check off, unless I overwrote to it? but that would be a few million times so I don't think so :-/ .
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2012-01-24 18:50
    Even without an EEPROM the Prop will still respond to the downloader and load a program into RAM. Even without a crystal the Prop will do the same. Having eliminated these possibilities then just look at the basics and check the programming cable that it does pulse the reset line and that data is being received from the downloader etc. Also check how much current the Prop is drawing if you hold the reset low as this should be <1ma or so depending upon BOE and the state of the I/O voltages. If it is drawing >10ma then you may have a faulty Prop.

    Once again however a photo is worth a thousand questions and guesses.

    EDIT: also make sure that you do not have any 5V inputs on the Prop that are not suitably current limited. Disconnect these to check the basics
  • Daniel HarrisDaniel Harris Posts: 207
    edited 2012-01-24 19:12
    One thing that the Propeller requires for downloading is that the computer be able to toggle the serial port's DTR (or RTS if configured in the Propeller Tool) line (which is connected to the Propeller's reset pin).

    Peter Jakacki has a good point - a photo of your circuit might help troubleshooting.
  • Cluso99Cluso99 Posts: 18,069
    edited 2012-01-24 20:34
    I presume you have tried powering the pc down. Just occasionally I have this problem (Windoze XP). The prop is a very robust chip apart from the pll and that is not your problem. There is so little required to get the prop communication. We need pics and perhaps circuits.
  • ThricThric Posts: 109
    edited 2012-01-24 21:15
    @Cluso99 I just powered down my laptop (and waited for the half an hour update :-/ ) and it doesn't seem to help; the propeller is still not visible to the prop tool, but the comport is. I'm running windows 7 and I sure hope the prop is fine :P .

    @Daniel Harris- I'm not sure how I can check the rst line (no scope) but the programming error hasn't been an issue before now so I think that it is being driven just something with the prop is no working? This is just a wild guess because my computer can connect to the FTDI chip so the next chip in line would be the propeller.

    @Peter Jakacki- I've disconnected all the periphials that I can (mainly just 4 pwm cables to servos), still nothing. I'm not sure how I would go about measuring the current draw of the propeller, how would you do this?

    I'll try posting my schematic and some pictures tomorrow (getting late now), electronics just seem to fail around me after I make a break through >.< .
  • prof_brainoprof_braino Posts: 4,313
    edited 2012-01-25 09:16
    As far as I know, only a REAL pro can cause the prop to fail. I've only noticed a couple verified failures on the forums, and my props always turn out to be fine.

    Last week I had a situation where the prop tool would intermittently report "No propchip detected" even though the virtual comport was reported present.

    I would unplug the USB cable, plug it back in, and it would find the prop and I was able to reprogram it. but the SECOND attempt or so, prop not found.

    The SPIN program was the same one that worked fine a couple weeks ago, and remains unchanged.

    Now it works fine again, and I did NOT do anything to fix it. During my diagnois process, I unplugged all the cables and replugged them. HA! A clue!

    I think its a combination of Windows update doing something unusual, and the fact that the ATX power supply on the PC is always giving power to the USB ports.
  • Nick McClickNick McClick Posts: 1,003
    edited 2012-01-25 10:30
    The only way I've fried my Propeller to the point it won't talk to my computer is with too much current - I was running a bunch of LED's without any current limiting directly from the Prop. The chip put up with it for a while, but eventually it stopped working.

    Is the chip warm to the touch?
  • Shawn LoweShawn Lowe Posts: 635
    edited 2012-01-25 11:10
    I am glad this thread is here, because I have a PPDB with a prop that is having the same issues. I swear when I first got the board I recognized the Prop chip, but now it wont with nothing hooked to the Prop. Perhaps this thread will help me with my problem as well (problem is probably between my ears)
  • ThricThric Posts: 109
    edited 2012-01-25 11:29
    @Prof_braino- I tried several scenarios with unplugging the usb connection and downloading several tries as well as the order I power my device (usb first then power or power then usb) no difference it seems.

    @Nick McClick- I powered my board up and left it for a few minutes and thankfully there is no magic smoke and the prop doesn't feel hot at all.

    There was a program in the EEPROM that I was booting from but now it won't boot up from there, everything seems to point towards a failed propeller it seems, yet at the same time it doesn't.

    I've attached my schematic, sorry to all the non-eagle users, but the schematic is too big to view with just a screen shot.
    sch
    sch
    BOARD 0.1.sch
    242K
  • Cluso99Cluso99 Posts: 18,069
    edited 2012-01-25 12:12
    Do make sure you are using the latest version of the PropTool (IIRC v1.3). I have less problems with prop or port not detected than the previous version. I have windoze XP.

    My only dead prop is the one I fried with my IR soldering station. I actually have not tried it, but since the smoke appeared and the package is blistered, I figure it's dead!
  • RaymanRayman Posts: 14,848
    edited 2012-01-25 12:14
    It does sound like it's dead, but it's hard to tell for sure without an oscilloscope...

    Do you have another Prop board you can try with your computer? Just to make sure your computer isn't the problem...

    BTW: Perhaps the only think I like about the DIP package is that it's easy to swap out when something like this happens...
  • ThricThric Posts: 109
    edited 2012-01-25 14:10
    @Rayman- the only reason why I dislike the DIP package is that once soldered, I find through hole such a pain to remove >.< . With SMD you can just blow the chip off with some hot air, which is what I may end up doing today with my board.

    I'll report back to see whats happened.
  • LeonLeon Posts: 7,620
    edited 2012-01-25 15:33
    I use sockets.
  • RaymanRayman Posts: 14,848
    edited 2012-01-25 15:56
    I too always use sockets with DIP, just so I can replace the chip quickly...
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2012-01-25 18:10
    No scope? Use leds!

    It's always handy to include status leds on boards and the simplest one is to add one to the SCL line tied high. This way you can see the Prop trying to clock the EEPROM after a reset plus you can just flash the led manually for a status signal when there is no I2C access. Of course the same could be said for the SDA line too but the SCL is more useful.

    EDIT: I can't really think of any boards I've designed that don't have some kind of status led on them. Sometimes I add a large value resistor from the supply/ground to feed the led so that if there is power (and nothing else) it will at least glow dimly, then when the Prop is running it will work as a normal status led. A quick visual indication helps solve many problems quickly,
  • RaymanRayman Posts: 14,848
    edited 2012-01-25 19:14
    I was wondering if LEDs could help in this situation...

    I think the pulse on the RES line would be too fast too see...

    But, maybe you could see activity on the TX output line after the Prop gets reset...
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2012-01-25 20:18
    I thought the Prop doesn't try to communicate unless it receives the correct header so the receive line into the Prop would need to be monitored to make sure (assumptions leads to frustration). It's also quite possible that you will see the reset led blink due to the eye's persistence of vision (or more accurately it's temporal integration/summation response) as along as it is bright enough and perhaps if you cupped your hand around the led to shield out surrounding light. You would be surprised how short a pulse your eye can perceive given the right conditions. Now that I have said this I have to go off and do some tests in this regard.
  • jmgjmg Posts: 15,183
    edited 2012-01-25 22:46
    Rayman wrote: »
    I think the pulse on the RES line would be too fast too see...

    :LEDs (plus your eye ball's native integration ) work quite well for narrow pulses.
    Choose a brighter LED if you know the pulse is narrow.
    I will be hard for the average Windows+USB PC to flip a RES line too fast to see (<< 1ms), and any sensible delay (5-50ms) is very easily visible.
Sign In or Register to comment.