 |
By: Anonymous: mpechner () on Sunday, December 27 2009 @ 01:16 PM PST (Read 3306 times)
|
|
|
Anonymous: mpechner |
| Anonymous: mpechner |
|
On mine, I am at the first test point and nothing lights.
I expect a pin off the CPU to be going hi/low every so often. My logic probe is showing pins either at high or low. Where high is 5V and low is anywhere from mV to about 1.2V.
I am hoping the BD written on the cpu is BulbDial and not BaD.
All solder points look good. All parts inserted correctly.
But I'll check again.
|
|
|
|
|
|
 |
By: Windell (offline) on Sunday, December 27 2009 @ 03:13 PM PST
|
|
|
Windell |
| Windell |
|
>On mine, I am at the first test point and nothing lights.
First make sure that pin 1 of the chip (lower left corner) is high. Double-check the soldering of the quartz crystal and of C1 and C2, as well as the places that they connect to the chip, pins 9 and 10. Make sure that C1 and C2 are the 18 pF caps, the ones with the black stripes. Also carefully look in the whole area around the chip for possible solder bridges or other missing connections. You may also want to verify all of the power/ground connections for the chip starting with the circuit diagram.
> I expect a pin off the CPU to be going hi/low every so often.
They do go high or low, but only at a *very* fast rate. In operation, up to six LEDs may be visibly lit at any one time, but in practice-- as a charlieplexed circuit --only one LED is ever lit at a given instant. So while one might guess by looking at the clock that a certain pin goes DC high or low to turn on an LED, it turns out that you can only see this with an oscilloscope or high-speed logic analyzer-- probably not with a multimeter or DC logic probe. 1.2 V sounds like a reasonable reading for a signal that's on about 1/3 of the time.
>I am hoping the BD written on the cpu is BulbDial and not BaD.
Yes; we mark them up to indicate that they are preprogrammed. (BD or BW, for standard or monochrome-- the only difference is in the "factory default" white balance setting.)
Windell H. Oskay
drwho(at)evilmadscientist.com
http://www.evilmadscientist.com/
|

Evil Scientist
 Status: offline
Registered: 06/15/06 Posts: 1932
Sunnyvale, CA
|
|
|
|
|
 |
By: Anonymous: mpechner () on Sunday, December 27 2009 @ 04:20 PM PST
|
|
|
Anonymous: mpechner |
| Anonymous: mpechner |
|
My bad. I had resisters in R12 and R13.
All perfect. Works great.
|
|
|
|
|
|
 |
By: Anonymous: Dan () on Wednesday, January 06 2010 @ 07:58 PM PST
|
|
|
Anonymous: Dan |
| Anonymous: Dan |
|
I've tested everything mentioned above and am about to pull up the schematic. This is my first soldering project though, so I am suspicious of myself.
Should most things on the board be connected to the 'top' of the 3 prongs on the power adapter? My multi-meter beeps, but only for a second, when I measure the connectivity of that to basically everything on the board.
One thing I noticed is that my µ-processor doesn't have any letters on it. Is there any chance that means it wasn't pre-programmed?
|
|
|
|
|
|
 |
By: Windell (offline) on Wednesday, January 06 2010 @ 10:48 PM PST
|
|
|
Windell |
| Windell |
|
 I've tested everything mentioned above and am about to pull up the schematic. This is my first soldering project though, so I am suspicious of myself.
Flaky solder joints are easy to come by, especially when you're just getting started. It only takes one, in the right place, for things to not work correctly. Check your solder joints visually-- make sure that they look like clean and shiny wet droplets of mercury, wetted to both surfaces that they are supposed to attach to. You need to make sure that power is getting to the AVR and that its connections to the clock parts (crystal and caps) are well established.
 Should most things on the board be connected to the 'top' of the 3 prongs on the power adapter? My multi-meter beeps, but only for a second, when I measure the connectivity of that to basically everything on the board. Be *very* careful poking around with that, especially when the power is on.
A better approach is to use the existing, labeled test points on the board. To test continuity, first UNPLUG your clock and let it sit for a minute. Place one of your leads to the GND_IN point on the right side of the board-- this is a better place to test ground connections. Yes, a lot of things are connected to ground, the schematic shows you which ones. If you find things that are connected to ground that should not be, that probably means that there's an accidental solder bridge somewhere.
 One thing I noticed is that my µ-processor doesn't have any letters on it. Is there any chance that means it wasn't pre-programmed? Possible but unlikely. The markings are often faint and small pencil marks; if you've been very careful not to smudge the top surface of the chip, you may see them.
Windell H. Oskay
drwho(at)evilmadscientist.com
http://www.evilmadscientist.com/
|

Evil Scientist
 Status: offline
Registered: 06/15/06 Posts: 1932
Sunnyvale, CA
|
|
|
|
|
 |
By: Anonymous: Dan () on Sunday, January 10 2010 @ 12:42 PM PST
|
|
|
Anonymous: Dan |
| Anonymous: Dan |
|
Thanks for all the great advice! I definitely only "poke around" with the multimeter when there isn't power... I'm not that naive . But, I'm glad you're checking.
I visually inspected all of the solder points on the blue board three times, even under a scope, but didn't see anything that looked bad. So, I put it down and worked on other things for a few days. Today, I picked it up and it lit! So, obviously, a bad solder joint then.
I cleaned up the board in a few places anyway, but still can't find anything that looks wrong. Finished the assembly, it's still lighting, so until it stops working again I think I'm going to ignore the problem. I'm fully prepared to disassemble it again when the time comes.
|
|
|
|
|
|