Shop OBEX P1 Docs P2 Docs Learn Events
Why will this NOT WORK?!?!?!?! — Parallax Forums

Why will this NOT WORK?!?!?!?!

MicrocontrolledMicrocontrolled Posts: 2,461
edited 2010-05-31 03:43 in Propeller 1
I have a Gadget Gangster Boss board with a Prop chip hooked up in the standard fashion exept that I am using 2 AA cells for power. The AA's are hooked up directly to the power and ground lines so the power portion of the board is not used. I have a 5Mhz Xtal. I have an SD card slot soldered to the right of the Prop chip. I had to cut 2 ground lines on the board because pin 7 on the SD card slot is on the ground line at the end of the board. The SD is hooked up to pins 0..3 in the usual way, and they are all pulled up to power with 10k's. I also have a headphone jack directly soldered to the chip with no amplifiers or filters. The problem is: Nothing will run. I have a Prop Plug jimmied up to the correct pins and it will detect the chip and download programs to it, but none of them will run. I at first thought I had the SD wired wrong so I ran a serail program to see if I could get anything on the serial line. Nothing. I replace the Xtal with an identical 5Mhz I got off a protoboard. Still nothing. I load on a simple BLINKING LED program.
CON

  _clkmode = xtal1 + pll16x
  _xinfreq = 5_000_000

PUB Blink

dira[noparse][[/noparse]26]~~
repeat  
  outa[noparse][[/noparse]26]~~
  waitcnt(clkfreq/5 + cnt)
  outa[noparse][[/noparse]26]~
  waitcnt(clkfreq/5 + cnt)



STILL NOTHING!! Then I comment out the clock functions


PUB Blink

dira[noparse][[/noparse]26]~~
repeat  
  outa[noparse][[/noparse]26]~~
  waitcnt(clkfreq/5 + cnt)
  outa[noparse][[/noparse]26]~
  waitcnt(clkfreq/5 + cnt)



THEN the LED starts blinking. What is up?? Is there something wrong with the EEPROM? It's the only componant that I haven't replaced! I've swaped out 3 different prop chips but nothing happens! I've even tested to see if the prop chips were bad! Is it something in my wiring that could be doing this?? What could be wrong????

Thanks,
Micro

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Check out my new website!!

Use the Propeller icon!! Propeller.gif

Follow me on Twitter! Search "Microcontrolled"
«1

Comments

  • hover1hover1 Posts: 1,929
    edited 2010-05-27 18:59
    Sounds more like a PLL problem, (I don't see how it would be the EEPROM).

    I believe running the PLL will draw more current, and the AA batteries may not be producing enough voltage. You are at 3.0 volts at best. Some rechargables will give you 2.4 volts total.

    2.7 is minimum VDD.

    Jim
  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-27 19:10
    I forgot to mention that I tried powering it off the PPDB's power supply just a while ago, and that didn't work ether. I was suspecting it was a PLL problem as well, but I can't see what is causing it.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • hover1hover1 Posts: 1,929
    edited 2010-05-27 19:37
    If you ran the 3 propellers off the PPDB power supply, I guess you don't have a PLL problem.

    Are the Boss Boards regulators isolated from the input power. I'm not familiar with the board. I have to get a schematic.

    Me's ponders some more.
  • W9GFOW9GFO Posts: 4,010
    edited 2010-05-27 19:56
    There is a thin trace that runs vertically along the right side of the Boss Board that if damaged can cause big problems. It probably is not the issue here but there is likely a scenario in which it could be.

    Rich H

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    The Simple Servo Tester, a kit from Gadget Gangster.
  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-27 20:03
    Oh shoot. I cut a thin trace because I thought that it lead to ground on the plane that the SD DO pin is soldered to. What does it do?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • hover1hover1 Posts: 1,929
    edited 2010-05-27 20:08
    I guess some pictures would help at this point. Does the Boss Board integrate all the decoulping caps needed by the Prop? The lack of them would spell PLL problems.
  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-27 20:10
    I just soldered the thin trace on the right back together and it still doesn't work on the blinking LED program with the clkmode enabled.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • jazzedjazzed Posts: 11,803
    edited 2010-05-27 23:05
    Did you try your LED program with _clkmode = RCFAST?
    That should give you a 12MHz system clock and will work without the PLL.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    May the road rise to meet you; may the sun shine on your back.
    May you create something useful, even if it's just a hack.
  • Nick McClickNick McClick Posts: 1,003
    edited 2010-05-28 00:12
    I bet it's a blown PLL. I think the classic symptom is 'works when you don't use the crystal, freezes when you use the pll'. Add ceramic caps to NO10 / EF10 for decoupling.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Propeller Forums RSS Feed!

    Gadget Gangster - Share your Electronic Projects
  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-28 00:13
    OK, i just tried it with RCFAST and the LED blinked. However, I am running an SD card and sound driver off this, so I think the Xtal is needed here. What do I do now?

    Thanks,
    Micro

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • jazzedjazzed Posts: 11,803
    edited 2010-05-28 01:19
    I'm afraid you need to bury that chip unless you have a slower application for it.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    May the road rise to meet you; may the sun shine on your back.
    May you create something useful, even if it's just a hack.
  • hover1hover1 Posts: 1,929
    edited 2010-05-28 01:32
    Add decoupling caps. Try new Prop chip.

    Jim
    Microcontrolled said...
    What do I do now?

    Thanks,
    Micro

  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-28 10:40
    I will add decoupling caps. Thanks.
    Its funny that you think the chip is bad, because I just took it out of the boss board and put it in the PPDB and it worked with full clock, playing music off the SD card. It must not be the chip, but the configuration of the boss board.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • jazzedjazzed Posts: 11,803
    edited 2010-05-28 11:59
    Very surprising that the chip works at full speed on the PPDB.
    Maybe the power/ground pin connections are not very good on the other board.
    Do you have a volt/ohm meter (DVM or other)? Radio Shack has cheap meters.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    May the road rise to meet you; may the sun shine on your back.
    May you create something useful, even if it's just a hack.
  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-28 15:00
    I'm getting a solid 3.3V across the Propeller voltage in pins with a voltmeter. This is really weird. I'll heat up the iron and get the decoupling caps on. Do you use 1000uf or what??

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • jazzedjazzed Posts: 11,803
    edited 2010-05-28 15:16
    The point of using a meter is to measure the resistance (with power off) between the ground plane and the two Propeller ground pins (same can be done from 3.3V to both Propeller power pins). Just for the sake of curiosity what are the decoupling cap values? You might also try another crystal.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    May the road rise to meet you; may the sun shine on your back.
    May you create something useful, even if it's just a hack.
  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-28 15:30
    I just put in a 10uf decoupler and a new Xtal. No luck.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • jazzedjazzed Posts: 11,803
    edited 2010-05-28 16:56
    Microcontrolled said...
    I just put in a 10uf decoupler and a new Xtal. No luck.
    Typically a 0.1uf (or less) decoupler is used across the Xtal side power pins to keep Xtal noise out of the power system. It is unlikely that decoupling has anything to do with your problem with a 5MHz Xtal. Follow other's recommendations on Tantalum polar capacitors for over clocking - I don't do that because it just invites trouble.

    Are you sure all Propeller grounds and power pins have good connections? There should be 0 ohms between the ground and both propeller ground (pins with power off).

    What else do you have connected? Can you post pictures of front and back of the board?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    May the road rise to meet you; may the sun shine on your back.
    May you create something useful, even if it's just a hack.
  • localrogerlocalroger Posts: 3,452
    edited 2010-05-28 22:23
    Sounds like a problem with the crystal or its wiring; next I'd try running the Prop off the crystal at 5 MHz without using the PLL. That will tell you whether it's the oscillator or PLL giving trouble. If the chip works at full speed in the PPDB the PLL's should be OK.
  • hover1hover1 Posts: 1,929
    edited 2010-05-28 22:26
    Picture time! It's not fun guessing how you have it wired [noparse]:)[/noparse]

    Jim
  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-28 22:36
    OK, OK, I'll post pics in a while, just let me test the resistance. Thanks guys! smile.gif

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-28 22:42
    About 2/10 of an ohm across the power lines. My meter is cheap and not exact, though.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • hover1hover1 Posts: 1,929
    edited 2010-05-28 23:16
    When one says across the power lines, that normally means across + and -, 2/10 ohm would indicate a short. Do you mean ground to ground, and power to power continuity?
  • HarpritHarprit Posts: 539
    edited 2010-05-29 01:38
    Keeping in mind that I am not an electronics guy..

    Two AA batteries provide minimal inrush current at start up.
    The thing is not starting up right is a strong possibility.
    At 80 MHz you are asking the chip to run at its highest current demand
    Even though you measure 3.3 volts, that may not be the case at start up
    Try starting out at 5 MHz then 10 then 20 then 40 and then 80 to see what happens
    Adding about 100 mfd across the battery might do the trick if the problem appears a the higher freqs only
    Put a scope across the batteries at start up and see if you see a spike down.
    Try it with a more powerful regulated power supply

    Harprit

    ·
  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-30 01:25
    Here are some pics of the board.
    P5260639.JPG
    P5270642.JPG
    P5270648.JPG
    P5270652.JPG

    @Harprit: I have been powering it off the PPDB power at the time, so that should not be a concern.
    @hover1: I don't trust my multimeter that much. I think that the meter is a little off.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • hover1hover1 Posts: 1,929
    edited 2010-05-30 01:34
    I don't see a single capacitor on the board.
    Where is the crystal mounted, I don't see it it the picture.

    Jim
  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-30 23:27
    The crystal is mounted under the board, and it IS connected to the right pins. I've checked over and over. The caps where added at the 2 circle marks under the top power regulator spot. I added them after the picture was taken. No luck on this so far.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • jazzedjazzed Posts: 11,803
    edited 2010-05-31 01:46
    Microcontrolled said...
    The crystal is mounted under the board, and it IS connected to the right pins. I've checked over and over. The caps where added at the 2 circle marks under the top power regulator spot. I added them after the picture was taken. No luck on this so far.
    So is starting with an empty Boss board (or something else) an option? Doing small things one at a time is often more effective than doing it all at once. Sometimes in a fit of enthusiasm, adding things like the SD connector can cause problems for example. BTW, if you add a .1uf cap like I mentioned before CLOSE TO THE CHIP, there will be less noise (that you can't see with a meter) on the power line. There is a lot of lead inductance there so it might help but I doubt it.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    May the road rise to meet you; may the sun shine on your back.
    May you create something useful, even if it's just a hack.
  • MicrocontrolledMicrocontrolled Posts: 2,461
    edited 2010-05-31 02:24
    If I had another boss board or had any intention of buying more, I would have just done that after this thread reached 2 pages. I have a limited supply of DIP 40 connectors, (I got them free at UPEC, but they are $1.50 a piece) and no extra SD card slot or any source to buy them from. So I would like to use the current boss board if possible. I don't really see what the purpose of decoupling caps are, but I'm sure that you need them.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Check out my new website!!

    Use the Propeller icon!! Propeller.gif

    Follow me on Twitter! Search "Microcontrolled"
  • Dr_AculaDr_Acula Posts: 5,484
    edited 2010-05-31 03:43
    I'm still trying to work out which tracks have been cut. But in the meantime, I'm presuming this is just a protoboard on the other side. If so, there do seem to be a few components missing. A reset button? Capacitors? "I don't really see what the purpose of decoupling caps are, but I'm sure that you need them." It isn't just decoupling. Also power supply caps. Maybe it's overkill but I have 1000uF on the input to regulators, 470uF on the output, 22uF tantalum somewhere near the propeller, 0.1uF between the two power pins near the xtal, and another 0.1uF across the other two power supply pins.

    Ok, you are battery powered but there will still be startup surges etc. So of the above 5 caps you could leave out the 470uF one.

    Can you post an updated photo of both sides of the board?

    Also, if you are trying to scrimp and save on sockets and they are $1.50 apiece, check out www.futurlec.com/SockIC.shtml where you can see 40 pin dip sockets are 15c. Futurlec take a couple of weeks for orders to arrive, but shipping is only $4 for small orders. Buy some other things while you are there. There are certain components I use lots of - 1k resistors, 10k resistors, 470uF caps, 22uF tantalum caps, 0.1uF bypass caps. The 0.1uF caps are 6c each for instance www.futurlec.com/CapCerMono.shtml

    Small steps like jazzed says. For instance, once you do get the thing working, when you add that sd card I'd be tempted to put a 22uF tantalum and a 0.1uF right across the sd card power pins.

    Addit: while the hot melt glue gun is warm, maybe some glue where the wires go on to the board, and also the wires on the back of the switch. Lots of flipping the board over when soldering and that is where the wires will break off.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    www.smarthome.viviti.com/propeller

    Post Edited (Dr_Acula) : 5/31/2010 3:52:22 AM GMT
Sign In or Register to comment.