 |
By: Greg (offline) on Monday, November 19 2007 @ 02:08 PM PST (Read 6136 times)
|
|
|
Greg |
| Greg |
|
I finished my assembly of the first board last night.
When I powered it up, all I get is solid illumination of roughly 1/3 of the LEDs, no dimming, no motion sensing, in any quadrant. Roughly 2/3 remain dark.
I checked the polarity and placement of everything, the positioning of the chips, the integrity of the joints, and found no assembly errors.
After soldering the chips, I did think they became quite warm after flipping the board and touching them. I even planned to solder/cool/solder the next set. Could I have fried the chips (all or one)? Any idea what would produce this kind of malfunction?
Greg
|

Apprentice
Status: offline
Registered: 11/19/07 Posts: 8
|
|
|
|
|
 |
By: Windell (offline) on Monday, November 19 2007 @ 03:09 PM PST
|
|
|
Windell |
| Windell |
|
Hi Greg,
It sounds like there are (at least) two different things wrong. It would helpful if you could describe what pattern / what banks the lit LEDs are in; that could help point to the root causes. Even better: take a picture of it running so that I can get a better idea of what's going on.
The lack of motion sensing over the whole board indicates that there is a systematic problem with the power supply. Two places to start looking: First, make sure that R115, R116 are the correct type of component and that they have good solder joints. Secondly, it is possible that one (or more) of the chips is not hooked up correctly or is (yes) actually dead. These chips are pretty tough; I've never killed one by soldering. However it is possible, especially if you hold the soldering iron to any given pin for more than a couple of seconds. The most likely chip issue is that one of the middle pins on either side of one of the chips is not soldered correctly-- pay careful attention to those points.
As I noted in another forum post, a problem with the power supply on the board can easily result in *half* the LEDs not lighting up. Since only a *third* or yours are lighting up, I suspect that there is a second problem as well: An LED is backwards in one or two of the banks that should otherwise be lit, by the pattern of lit banks that you have. I know that you said that you checked the polarity; just to be sure (and for everyone else reading this) the flat sides of the LEDs should match up to the flat side of the symbol on the board.
Another tactic that you could play at this point is to go ahead and assemble the next board, with a lighter touch on the soldering iron so that you can have a little more confidence in what you are doing. Try not to expose the chip to the soldering iron for more than about 1/2 second at a time. Once you have a board that works, it might be a little easier to debug the one that you have already built.
Windell H. Oskay
drwho(at)evilmadscientist.com
http://www.evilmadscientist.com/
|

Evil Scientist
 Status: offline
Registered: 06/15/06 Posts: 1932
Sunnyvale, CA
|
|
|
|
|
 |
By: Greg (offline) on Monday, November 19 2007 @ 09:10 PM PST
|
|
|
Greg |
| Greg |
|
Power supply it was, well, mostly.
I resoldered the R115 & R116 and the switch wire (I think it was the switch wire...it's just temporary, so I did not take care when soldering). This made the LEDs fade as they should, but many where still dark.
The second problem was bad solder joints. I turned up the temp on my iron (it's new to me) and briefly touched every joint.
Now all is good, except for IR operation. All four IR LEDs are in correct polarity, notch to flat, long lead in square hole.
Any ideas?
Greg
|

Apprentice
Status: offline
Registered: 11/19/07 Posts: 8
|
|
|
|
|
 |
By: Windell (offline) on Tuesday, November 20 2007 @ 01:05 AM PST
|
|
|
Windell |
| Windell |
|
Proper IR operation requires, among other things, that the chips are in correctly (you've got that now-- good), and that both the IR LEDs and the black phototransistors are in place properly. Resistor R415 also must be in place and properly soldered. You should go through and double check all of those connections.
Really, there are two components to the IR working: there are the IR sources and the IR sensors. You can test whether the IR sensors are working by turning on or off an incandescent light source that casts light on the panel. Even if the sensors are not soldered yet, you can test whether the IR sources are working by looking at them with a digital camera; you may be able to see them lit up, continuously.
Windell H. Oskay
drwho(at)evilmadscientist.com
http://www.evilmadscientist.com/
|

Evil Scientist
 Status: offline
Registered: 06/15/06 Posts: 1932
Sunnyvale, CA
|
|
|
|
|
 |
By: Greg (offline) on Tuesday, November 20 2007 @ 09:42 PM PST
|
|
|
Greg |
| Greg |
|
Again, excellent advice!
We looked more closely and found it was barely responding (in the dark), but was, non the less. So we compared the digital camera image of board #2 and found a faint light coming from the IR led's (compared to #2)
It was R415, but not the solder joint. I somehow managed to put a 10m resister in there...no wonder it was a little dim!
BTW, board #2 was perfect!
|

Apprentice
Status: offline
Registered: 11/19/07 Posts: 8
|
|
|
|
|
 |
By: Windell (offline) on Tuesday, November 20 2007 @ 09:53 PM PST
|
|
|
Windell |
| Windell |
|
Yup 10 M load resistor on the LEDs-- that'll do it! 
I'm glad to hear that things are working better-- let us know if you need help with anything else.
Windell H. Oskay
drwho(at)evilmadscientist.com
http://www.evilmadscientist.com/
|

Evil Scientist
 Status: offline
Registered: 06/15/06 Posts: 1932
Sunnyvale, CA
|
|
|
|
|