3v LED not lit

[deplanche]

Assembling 2 LEs.  Tested both before installing the chips.  Both 5v and 3v LEDs were on for both.  Installed the chips.  Now 3v doesn't light up on one no matter what.  The other will occassionally.  I know it is a chip issue, but before I start pulling all of them, is there a certain one or ones that are more likely to be causing this problem than others?

[RJ]

Check the chips one is likely backwards. What version of the pcb?   

RJ

[deplanche]

Both v4.0.

[castortiu]

RJ can confirm this,

But I had the same issues on 3 LEs V4, if I remember properly the problem was the 3.3V regulator cut off voltage was too high and the output of the 5V regulator was not enough to feed the input of the 3.3V regulator.

I think RJ changed the 3.3V regulator for a regulator with a less cut-off voltage and also works on wider temperatures.

In my case I fixed cutting the input track of the 3.3V regulator and feeding from the same input track as the 5V regulator, the 3 LEs ran fine without problem for 3 weeks that the show lasted, the regulators were a little warmer than usual but not much, barely warm to the tact, still the 3 LEs are working great.

Did you get the component from the coop? and when was it? May be it is a totally different issue but I had the same symptom, without chip worked, with the PIC no 3.3V.

Cas.

[RJ]

Yes this is the reason I asked which version. We had an issues with some V4 units and changed the 3.3v voltage regulator to fix it. If you pm me how many units you have and your address I will send you a couple of regulators to change the 3.3 out with and fix it.

RJ

[deplanche]

I actually stopped building these because of that problem last fall/winter, and I didn't NEED them for last years display.  I went back thru the posts recently, and since I was ordering some parts from Mouser anyway, got a "511-L7805ABV" to replace the one that was included with the coop back then.  Was that the right part to get, and was it a simple part substitution or did I need to do something else as well?

[RJ]

I actually stopped building these because of that problem last fall/winter, and I didn't NEED them for last years display.  I went back thru the posts recently, and since I was ordering some parts from Mouser anyway, got a "511-L7805ABV" to replace the one that was included with the coop back then.  Was that the right part to get, and was it a simple part substitution or did I need to do something else as well?

Ok that is the issue then.

Shoot me a pm with that info and I will get you a different part that will if it for you.

RJ

[deplanche]

Replace the parts with those that you send.  Both boards now have both power LEDs lit when plugged in.  Bad news is that when I run the test, i get the LED display to light up all the way to the "888" then "tst", but none of the channel light LEDs come on.  This is true for both boards.  So I think I still have something not quite right.  Any thoughts?


Edit - I am pretty sure, but not 100% positive the one that was intermittently working before was able to run the test fine when working.  So I don't think I have the LEDs in backwards or anything like that.

[RJ]

Replace the parts with those that you send.  Both boards now have both power LEDs lit when plugged in.  Bad news is that when I run the test, i get the LED display to light up all the way to the "888" then "tst", but none of the channel light LEDs come on.  This is true for both boards.  So I think I still have something not quite right.  Any thoughts?


Edit - I am pretty sure, but not 100% positive the one that was intermittently working before was able to run the test fine when working.  So I don't think I have the LEDs in backwards or anything like that.

Push the select button.  Does one of the leds light?

RJ

[deplanche]

Once it get past the 888 and goes to tst, there is a period after then, when it should be lighting each channel.  But nothing is lit.  And nothing happens when i push either button, I think because it is still trying to light them in the test mode.  If I wait for a minute, and push select button, the whole test sequence seems to start over with the numeric LEDs cycling thru again.

So I guess, my followup question is when should I push the select button for what you are saying in your last post?

[RJ]

You gave me enough info!

I have bad news!     


I am pretty sure you put the leds in backwards   

Take a digital volt meter and put the leads on the two solder point of the channel number 1 led with the negative probe on the top side of the led and the positive on the bottom side. Then run the test again and watch your meter. I think your going to see the voltage jump each time the led should light.

Sorry if thats the case. I even put a little + on them and a extra note to try to keep anyone from doing that. Let me know and I will give you some advice on how to proceed without damaging the pcb trying to get them out.

RJ

[deplanche]

I'll give that a later this week (out of town right now). 

I have been testing most LEDs with a watch battery before installing them too (had 1 blown for another board once and since have been testing before installing).  Would there be any harm in touching the battery to the led on the LE if it was NOT connected to the AC power supply.  That would also let me know if I have the LEDs in correctly I would think.  I always like double checking everything if you counldn't tell.

[rm357]

It's also possible that the opto-isolators are in backwards. Can you post a high res picture of the front and back of your boards?

Some of the folks here are very good at catching things like solder bridges and even bent pins...

RM

[RJ]

It's also possible that the opto-isolators are in backwards. Can you post a high res picture of the front and back of your boards?

Some of the folks here are very good at catching things like solder bridges and even bent pins...

RM

Yes I thought of that last night but had already him told him to check with the meter. He is absolutly correct in that the optos backwards would do the same thing.

RJ

[deplanche]

After some other projects this summer and fall, I am finally getting back to this now that it is decorating time (and I remembered it wasn't working).

I checked the two legs of the LEDs with the voltmeter, and when they should be coming on with the test function, I measure a voltage of 1.62 V.  I did check all of the LEDs with a button battery, and they all seem to be in the correct orientation, so I am pretty sure that is not the problem.

I also put a voltmeter on the legs of the two regulators.  The one has 8V on one leg and 5 V on the other.  The other has 5V and 2.42 V.

Any thoughts on what could be wrong?  I have two boards with the same problem and are measuring the same voltage within 0.02 V.

Jon