Page 2 of 2

Re: RBBB issue with I2C

Posted: Tue Jul 12, 2011 5:07 pm
by mndless
Here are a few links to pictures... camera will only focus, close in, to a limit....

http://i585.photobucket.com/albums/ss30 ... bbb003.jpg

http://i585.photobucket.com/albums/ss30 ... bbb002.jpg

Used my reading glasses to enlarge this one... actually worked...

http://i585.photobucket.com/albums/ss30 ... bbb004.jpg

Re: RBBB issue with I2C

Posted: Tue Jul 12, 2011 6:20 pm
by scott-42
Sorry I should have specified the part of the board I really cared to see. Link to Album

This is the back of the analog pins. You can see a pinkish area around the pins where there is a ground plane.
http://i.imgur.com/H79qD.jpg

This is the top of the analog pins. Again you can see a slight outline of the ground plane surrounding the pins. You can also see the dark area through the board edge (very bottom of the picture) which is the back of the ground plane on the bottom.
http://i.imgur.com/b9X9H.jpg

You really need a magnifying glass if you are to check the ground plane around all of the pins to make sure it is not touching any of them (except GND of course). The reason I was asking you to check the resistance values between the GND/5V/A5 pins was that I suspect a short between the ground plane and the power.

The only other thing I could think of was that the diode (D1) was put in backwards but that doesn't explain why you would see any power on A5 with no chip installed.

Re: RBBB issue with I2C

Posted: Tue Jul 12, 2011 6:57 pm
by mndless
Minor destructive investigation in progress now... ;)

Took higher resolution and cropped inward some...

Image

http://i585.photobucket.com/albums/ss300/mndless/11.jpg larger view

Image

The pins are clean and shinny, just bad reflection in this picture... don't think my camera is going to allow a good enough close up...

Image

Interesting note... after removing the socket, as shown in the pics above, I still get the +/- 5v between grnd and A4 and A5....

Re: RBBB issue with I2C

Posted: Tue Jul 12, 2011 8:48 pm
by scott-42
A5, along with all the other analog pins, don't connect to anything other than the Mega pin directly next to it. For voltage to appear on A5, it has to have a short on the board to something and the only thing close is the ground plane. This is why I asked you to check the resistance between A5 and GND to see if it is shorted together.

Re: RBBB issue with I2C

Posted: Fri Jul 15, 2011 9:21 am
by mndless
So.... I take it that regestering a resistance value is a bad thing.... ;)

Re: RBBB issue with I2C

Posted: Fri Jul 15, 2011 10:28 am
by scott-42
If you have a resistance other than ∞ (infinity) from A5 to GND, then yes you have a problem. Either the solder joints are really bad and you leaked past the solder stop or the board is bad.

Have you had a chance to measure it and can you report the values?

Re: RBBB issue with I2C

Posted: Fri Jul 15, 2011 11:02 am
by mndless
Yep... my old multi meter wast recording a value, so I ordered a new one... got it last night, and sure enough... it does register a value, it was bouncing around beetween KH and MH.. .big numbers, but resistance was there...

I think I know what happend.... classic newb mistake.

I didn't go buy any new solder, I had some flux chore solder already... it wasn't adhereing as well as I liked, so I carefully, applied some additional flux to a few pads.. in hind sight, I bet it was not the right kind and obviously must have some electrical properties. It appears that If even small amounts of this flux melts and runs down into this area around the A4 and A5 pins, your shorted..., I guess the alcohol isn't sufficient to clean and/or leach it back out??...

Simple and smart thing would have been to just go buy some new solder.... oh well... great learning experience!

And thanks for the great help figuring out my goof... I'm sure there will be more :)

Re: RBBB issue with I2C

Posted: Sat Jul 16, 2011 7:12 pm
by scott-42
I'm glad you got it figured out. :D

I guess it was just time for the multi-meters to go out as one of mine also decided it was done. :( I also had to order a new one this week. Though mine was really abused in my younger years and I had fried the Amp reading portion of it a long time ago. Who knew checking 110V in the Amp leads position would arc weld the leads while blowing the fuse and frying the board. :o :oops: