Safe to test

[Mickpat]

I finishing building the Aether board, but am still working on the case and waiting for my PIC programmer to arrive.  I have 3 more to build, but want to make sure the first work is built correctly before starting the other 3.  Is there a safe way to test the board and does it do any good to test without the pic being programmed?

[batdive]

I would wait...   I have 5 under my belt and through experience the 120 V isnt that fun to mess with if something goes wrong (see my final build safety note post  ).

The first test to see if it works would be the Ohm test on the LEDs, to make sure the "tab" isnt making an electrical connection with the copper backing.   

The second test I do is the actual PIC programming, as that will light up the red Pwr LED and verify the PIC is seen and SMD'd correctly.

The last test I do is the manual switch test at the end of the install as this verifies all LEDs are operational to the PIC since it uses all 3 to make the cool white.

Trust me I know your excited to do it ASAP, but give it the couple days for your other parts to get their and do it right.

Just my 2 cents..


-JS

[RJ]

I would agree to wait.

But on a side note looks like a very well done job and so I am sure you will have no issues.

Biggest thing we have found is users not use to soldering without soldermask should do a good once over of the back of the pcb to look for any bridges. We have had a couple that had one. they did not hurt anything and once it was cleared it work fine.

RJ

[Gary]

This may sound like a dumb question, but what is soldermask?

[tbone321]

A solder mask is the shiny coating that they put over the board to protect the traces.   It both insulates them and prevents solder from sticking where it is not wanted. 

[RJ]

It also as a side note acts like a insulation layer to some level so on the back of the pcb which acts like a big heatsink we do not have it. so the couple solder joints there require a little more attention than when you have soldermask on the board.

RJ

[Mickpat]

I used my meter and touched the tabs of the LED to the copper back board.  5 of the 6 barely register, but the 6th one looks like it making a full connection.  I haven't powered up the board yet, but what could happen?  How do I fix an LED that is epoxied to the board ? 

[Mickpat]

I got a little impatient and was able to pop the LED off the board.  Its still soldered to the board, but fully raised.  What’s odd is if I perform the test again, top of the tab to the bare copped board, the meter still registers.  I am fairly sure none of the leads are bridged.  Does it sounds like something is wrong?

[RJ]

I used my meter and touched the tabs of the LED to the copper back board.  5 of the 6 barely register, but the 6th one looks like it making a full connection.  I haven't powered up the board yet, but what could happen?  How do I fix an LED that is epoxied to the board ? 

This is why you are suppost to test them when you put epoxy them on not after they are attached permanantly. This is a problem but I see you already removed it in the next post.

RJ

[RJ]

I got a little impatient and was able to pop the LED off the board.  Its still soldered to the board, but fully raised.  What’s odd is if I perform the test again, top of the tab to the bare copped board, the meter still registers.  I am fairly sure none of the leads are bridged.  Does it sounds like something is wrong?

Yes if the meter reading is not  1000's of ohms. you say it registers but what does it register?

It is much easier to bridge the leds leads than the epoxyed body unless you go crazy with very little epoxy or crazy moving it around while it it not hard, ect. So what I am saying is it would not suprise me that the back is not where the problem was.

RJ

[Mickpat]

Thanks RJ.  The LED that has the issue is next to the temp sensor.  I don't see any bridges between the leads of the LED.  When I meter, it registers 100%.  In fact, it registers 100% between Ground and LED pins 1,4, 6.  I have looked for bridges, but don't see any.  What areas should be looking at?  The other 5 LEDs don't have this issue. 

[n1ist]

This is just a guess, since I don't have an Aether and am only going by the pictures...

The LED chain has all 6 LEDs of a color wired in series, with a sinking driver.  This LED is the one at the supply end of the chain.  According to the pinout of the LEDs, the tab is directly connected to the anode of the red LED.  In circuits using the STCS05ADR driver, the anode of the LED strings is connected to the positive rail.  The heatsink region on the board is grounded (hence the use of non-electrically-conductive epoxy for mounting the LED).

It sounds like there is a short somewhere on the board between the positive supply rail and ground.  I would check both sides of the board; pay attention around the bridge rectifier, electrolytic caps, diodes  by the constant-current sink chips, and regulators as that's the most likely place to find such a short.

/mike

[Mickpat]

Just a quick update.  This was an issue with the meter and not the board.   

[RJ]

So are you functional?

RJ

[Mickpat]

Kind of.  See my post DMX Data.  The board powers on and I was able to flash the PIC, but it doesn't respond to DMX.