Shop OBEX P1 Docs P2 Docs Learn Events
Cannot Detect Propeller on a Custom-made Board — Parallax Forums

Cannot Detect Propeller on a Custom-made Board

HylianSaviorHylianSavior Posts: 6
edited 2011-09-07 03:21 in Propeller 1
Hello all,
I am having trouble detecting my Propeller chip on a board that I have created. This being my first time designing a custom PCB, surface mount soldering, and using the Propeller, there are a lot of things that could have gone wrong.

I am using an FTDI breakout to communicate with my Propeller, switching the reset signal to RTS as per http://www.ladyada.net/make/ybox2/progcable.html. However, when I connect only Vss and the Rx/Tx lines without Vcc, my power LED lights up. Is this a function of the Propeller or a mistake of my own? I've looked all over with a multimeter and I can't seem to find any shorts.

I have attached my schematic here in case anyone wants to see it:
schematic.jpg


Thanks in advance,
HylianSavior
schematic.jpg
1024 x 1164 - 87K

Comments

  • ElectricAyeElectricAye Posts: 4,561
    edited 2011-09-05 16:13
    Hylian,
    Welcome to the forums.

    If your LED is lighting up, have you measured the voltage that the LED is experiencing and then tried to follow that voltage to see where it is coming from? Also, is the voltage across the LED continuous or some sort of digital signal?
  • idbruceidbruce Posts: 6,197
    edited 2011-09-05 16:18
    Hylian

    If you have a regular PropPlug, you could test your board with that instead, but I would imagine that you don't have one.

    Bruce
  • idbruceidbruce Posts: 6,197
    edited 2011-09-05 16:25
    Hylian

    You could also try using your modified cable on a board that is known to be in working condition.

    Bruce
  • Mark_TMark_T Posts: 1,981
    edited 2011-09-05 16:27
    HylianSavior wrote: »
    Hello all,
    I am having trouble detecting my Propeller chip on a board that I have created. This being my first time designing a custom PCB, surface mount soldering, and using the Propeller, there are a lot of things that could have gone wrong.

    I am using an FTDI breakout to communicate with my Propeller, switching the reset signal to RTS as per http://www.ladyada.net/make/ybox2/progcable.html. However, when I connect only Vss and the Rx/Tx lines without Vcc, my power LED lights up. Is this a function of the Propeller or a mistake of my own? I've looked all over with a multimeter and I can't seem to find any shorts.

    I have attached my schematic here in case anyone wants to see it:
    schematic.jpg


    Thanks in advance,
    HylianSavior

    Alas if you have connected RX/TX and not Vdd then you will have probably exceeded the abs max specs for the Prop chip many times over - the inputs must not exceed Vdd by more than 0.5V at any time....
  • $WMc%$WMc% Posts: 1,884
    edited 2011-09-05 16:31
    The USB cable has 5volts on it. Be-careful with this.
  • idbruceidbruce Posts: 6,197
    edited 2011-09-05 16:34
    Walt

    I was just reading all this....
    First get yourself a FTDI cable, make sure its the 3.3V version!

    Actually pretty interesting... I learned something new today.

    Bruce
  • $WMc%$WMc% Posts: 1,884
    edited 2011-09-05 17:10
    Hello Bruce
    '
    I was referring to a regular USB cable and jack.A bridged solider joint could put 5volts from this cable where you wouldn't want it.
    '
    I believe the FTDI chip is powered by the host USB 5volt supply,It is at least on the PropProDEV board.
  • idbruceidbruce Posts: 6,197
    edited 2011-09-05 17:15
    Walt

    According to the information that he linked to... The intended cable has a FTDI chip embedded within the cable.... That was a new one for me.

    Bruce
  • $WMc%$WMc% Posts: 1,884
    edited 2011-09-05 17:24
    I didn't look at the link sorry. I thought it was board mounted.
    '
    This is a new one for me too!
  • HylianSaviorHylianSavior Posts: 6
    edited 2011-09-05 18:12
    Thanks for your replies, everyone.
    I can confirm that I am using the 3.3V version of the FTDI breakout I am using.
    The Rx, Tx, and Reset lines all seem to be held high at 3.3V.
    I have also checked that the Rx/Tx and Reset lines all reach the correct pads on the Propeller with a multimeter.

    I looked at all the lines with a scope while trying to identify with Propeller Tool, and it looks like all the lines get pulled low at the same time (or about, since I can only check one at a time). This seems to imply that I have a short somewhere or my Propeller chip is broken. I have searched for a short with a multimeter for hours, so I'm led to believe that the problem lies somewhere else.

    Another clue may be that I had previously soldered the Propeller chip upside down and plugged it in before desoldering and putting in a new Propeller correctly. Some of my other ICs may have gotten fried, although I'm not sure how since they should've been only getting power.
  • LeonLeon Posts: 7,620
    edited 2011-09-05 19:04
    BOEn shouldn't be floating, connect it to Gnd.
  • HylianSaviorHylianSavior Posts: 6
    edited 2011-09-05 19:45
    Leon wrote: »
    BOEn shouldn't be floating, connect it to Gnd.

    Oh, that may be the problem, thanks. Now to figure out a way to connect it... I have the Propeller in a QFN package.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-09-05 19:52
    (What about bypass caps on/for all the Prop V_cc pins? There aren't any indicated. Just putting that out there. Keeping things understated as my wont. OK?)
  • Zap-oZap-o Posts: 452
    edited 2011-09-05 19:56
    Just solder one side of a small wire on the propeller and the other to GND. I cant stress the need to use solder-wick it makes soldering surface mount ic's a breeze.
  • HylianSaviorHylianSavior Posts: 6
    edited 2011-09-05 21:06
    PJ Allen wrote: »
    (What about bypass caps on/for all the Prop V_cc pins? There aren't any indicated. Just putting that out there. Keeping things understated as my wont. OK?)
    My voltage regulator has capacitors, which should be fine.
    Zap-o wrote: »
    Just solder one side of a small wire on the propeller and the other to GND. I cant stress the need to use solder-wick it makes soldering surface mount ic's a breeze.
    Problem is, I have the QFN package- no leads. I used a skillet and solder paste to solder it in. I'll probably find a way to bridge BOEn and GND since the pins are right next to each other.
  • Mike GreenMike Green Posts: 23,101
    edited 2011-09-05 22:05
    Your voltage regulator may have capacitors, but that doesn't help the Propeller chip. You really need a bypass capacitor of around 0.1uF (100nF) across each pair of Vdd/Vss pins on the Propeller, mounted as close as possible to the chip. If your setup worked, maybe you could get away without them or some of them, but when your setup doesn't work, the lack of these capacitors overshadows most other causes.

    The FTDI chip can provide a little bit of power through the Tx line to the Propeller when the Propeller isn't powered and vice-versa. What happens is that the Tx signal feeds through the I/O pin's protective diode to the Vdd bus on the chip, then out the Vdd pins on the chip to the power LED. The other way, if you unplug the USB cable and power the Propeller, then try to transmit from the Propeller to the PC via the FTDI chip, the Tx line from the Propeller provides a little bit of power to the FTDI chip, enough to sometimes falsely cause resets of the Propeller. One solution is just not to try to transmit to the FTDI chip if USB is disconnected.
  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2011-09-06 08:17
    My first thought was that any soldering issues on a QFN package will be nearly impossible to detect on inputs to the Propeller chip. So assuming the layout is all correct (minus the BOEn), then you could still have issues under the package that you cannot easily determine.
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2011-09-06 11:23
    This thread could help to troubleshoot a bad connection...

    http://forums.parallax.com/showthread.php?134326-IC-Characterization-Conceptual-DEMO
  • HylianSaviorHylianSavior Posts: 6
    edited 2011-09-06 12:25
    OK, thanks for the help, everyone!
    It looks like I'll be making a revision 2 of my board.
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2011-09-06 19:49
    As a design tip you can use the same I2C lines as the EEPROM rather than dedicate a pair as you have. As Mike mentioned those bypass caps need to be close to the package but I find you don't need all of them if you have a well designed tight power and ground layout. Consider allowing for a load resistor of around 1K or so (could be the power LED) across the 3.3V supply to prevent phantom power problems and also this helps during reset when the Prop draws very little current and the supply rail could float up if the inputs have been overdriven. I find this VDD float a problem with the whole reset process and I have seen it prevent the Prop from being reprogrammed.

    Have you tried a cheap toaster oven rather than a skillet? I fold aluminum foil to about 4 layers thick and use this as a tray for the pcbs and a flat length of metal (2U rack front panel) for the "pizza shovel". Put a little upwards fold on the front edge of the foil and this makes it easier when you go to remove the "tray". When you paste up the QFN apply the paste to the bottom of the QFN rather than the pcb as you will find this is far more reliable. QFN pads for manual assembly should be longer than the recommended lands and protrude out so you can get to these with a fine tip of necessary. Don't be afraid to smudge the solder paste across the pads as I find this helps to eliminate floating pins and the little bit of solder in-between always draws towards the pins and pads during reflow anyway.
  • LeonLeon Posts: 7,620
    edited 2011-09-07 03:21
    I don't have an problems soldering QFN parts with my Metcal soldering station. I extend the pads as Peter mentioned, and drag-solder them using plenty of gel flux.
Sign In or Register to comment.