LED matrix question

[Matrix12x]

I just finished my MB6582, all works fine except LEDS in the Matrix for rows 1-8 - filter 1 2 3 and pulse width 1 and 2.

i.e. the bottom 3 rows of the matrix seem to work fine.

any ideas.

I am going to check my wiring again and my solder joints.

Edited November 20, 2009 by Matrix12x

[Matrix12x]

I ran the LED test program and I had no LEDs light, including the working ones.

Is this because Wilba's LED Matrix is wired differently than the standard control surface?

[Wilba]

I ran the LED test program and I had no LEDs light, including the working ones.

Is this because Wilba's LED Matrix is wired differently than the standard control surface?

I suspect yes, because if you were running the one specific to MB-6582 then you wouldn't be asking this question:

BTW the MB-6582 wiki page had a link to this post/test app but it was broken since the forum change.

I really should host that app somewhere else, or at least add another asm/hex to the default one :whistle:

[Matrix12x]

Thanks. All LEDs except the top 5 rows of the Matrix lit up.

Is there anything in common with the first 5 rows that is not common with the last 3 rows?

[Matrix12x]

I'm an ass. I just started measuring the resistors for the LED Matrix and the ones that worked had +5V approx. the ones that did not had 0V.

I flip over my PCB and lo and behold missing solder joints on one side off all 5 resistors for the rows that are out.

Gheez. I must of been distracted by my 18 month old whilst I was soldering.

Thanks for the help and for the other LED test program.

[Wilba]

No, the LED matrix there is very repetitious...

http://www.mb6582.org/plans/MB-6582_CS_PCB_R2_Color.pdf

I would suggest taking out all the 74HC595 and testing each LED individually by supplying +5V to R48-R55 (the end closes to IC, so power goes through the resistor) and supplying ground to JD8 D0-D7 (i.e. the LED matrix columns). If you can get LEDs in those rows to light up then the problem is with the shift registers (74HC595 ICs). If the problem remains, then you'll need to do continuity tests and work out if those tracks are broken somewhere, or shorting with something else.

I did notice your status update says "well, the heatsink I used was too tall and shorted out my CS... lol" so maybe it's related?

Explain what happened there, and it might explain why five rows of LEDs don't work.

BTW you probably should not assume the LEDs should work until you test every one individually with +5V->220R->LED->Gnd

[Matrix12x]

It was my missing solder joints on the backs of the resistors.

All now works great on the LED test and in normal operation.

also of note..... When I hit any two buttons at the same time on the left side of the matrix, the matrix changes from matrix to meter mode and vice versa. anyone else see this behavior?

[Wilba]

Yes this is the default way of switching between matrix modes if you don't use a matrix mode button (I added that for MB-6582).