First two LEs built & working, but what's up with the data link light?

[richagem]

I've successfully built my first two LEs (woohoo!), with the second one going much faster than the first <grin>.  I don't have all the power cords plugged in for the various channels, but I created a test sequence in Vixen that steps through all 32 channels and all indicator LEDs light up as expected.

The only thing that's a little funky is that on one of them (the first one I built), the Data Link LED is lit even if the LE isn't plugged in to the dongle (both wireless & terminate jumpers are closed).  In spite of it being lit, the channels seem to respond fine, and when I bring the 2nd LE into the picture, both LEs seem to respond fine whether the one with the lit Data Link LED is the first in the series (terminate jumper removed) or the last (terminate jumper closed).

I've checked my connections on the back and everything looks clean.  I've swapped the 485 chips between the two boards, but the behavior doesn't change.

Any thoughts on what else I should look at?

Rich

[RJ]

Some Rs485 chips behave that way and we do not know why. Normally changing the one on the right of the data in jack will solve it.

RJ

[dmoore]

If the problem is in the 485 chip and he swapped the chips between the two controllers, wouldn't the problem follow the bad 485?  He seem to indicate that he swapped them and nothing changed.  Could that be a board problem?

[richagem]

Correct, the behavior doesn't follow the chip(s) to the other board...

[RJ]

Yes I was worried you change just one and got the other one.

If that is the case is there a chance you have the led solder in backwards?

RJ

[sirloinofbeef001]

That is the same problem I have..(if its a problem) A couple of my boards do that. I checked the solder joints and I changed the led's as well.The testing comes out fine tho.. Is it a big deal???

[richagem]

I verified the LED is in the right way...

[RJ]

No it does not harm anything but I do not understand why some of the 485 chips do not follow their specs.


Now oh his it does not seem to be the chips from what I so I am curious. Post some good clear close ups of both sides of the board in question.

RJ

[richagem]

Here are some pictures.  Let me know if they're clear enough & are showing the right parts.  I can also put higher resolution ones out on the web to look at.

Thanks,
Rich

[RJ]

Weird!  It all looks good to me. anyone see anything I missed?

RJ

[deanathpc]

Maybe a cold solder joint somewhere.  Although they all look good.  There are times though unless you look at 'em via a magnifying glass you won't see 'em.  I'm honestly not sure.

I know I have a couple now that the data link led dims depending on what i'm doing.  I think it might have started out as on whether i had something plugged in or not though.  I can't remember.  Never hindered any function though.

[Dennis Cherry]

My build of 11 LE's last year had your same issue on some of the LE's. I had 5 of the 22  485 chips that acted that way.  It did not affect the data at all.

Putting a scope on the output of a normal 485 chip without a DMX signal shows the output is high or at the 5 volt level. on the 5 chips that would turn the DATA LED on the output would be low or < .5 volts.

Next adding the DMX signal to the LE to both chips did not make a difference in the DMX signal pattern.

As to the LED dimming slightly that would be normal on the 485 chips that keep the DATA LED on. that's because adding the DMX signal now is turning the LED off and on very fast with the 512 channel data stream 44 times a second.

[rm357]

In the third picture - I think pins 7 & 8 on the 485 chip may be shorted by the traces on the board making contact.

There should be a small gap between the trace that goes to pin 8 and pin 7.

On second look, it looks like 3 and 4 may be shorted as well...

RM

[kauthier]

Very nice solder work...although it looks like you've got a few cold solder joints in the third pic of the back side.   Looks kind like a blob rather than a smooth flow.  Hard to tell from a picture tho.

Kevin

[sirloinofbeef001]

In the third picture - I think pins 7 & 8 on the 485 chip may be shorted by the traces on the board making contact.

There should be a small gap between the trace that goes to pin 8 and pin 7.

On second look, it looks like 3 and 4 may be shorted as well...

RM

I'm not sure about this, but if you look at the sockets for the led display, dont they look the same? Very clean board tho..