MAX7219 + LED matrix module !=

[granddad]

Looks like i need to re-code my 7x5 font..

 

[Nigel Goodwin]

Different modules are wired differently, so its best to make your code configurable if you can, there should be no need to alter the font (assuming it's correct in the first place).

 

[Pommie]

It looks fine, what do you think is wrong with it?

Mike.

 

[granddad]

mmmm

wrong with it?

my bit stream is coded on the red leds one direction .. the same bit stream (green) shows the other orientation ( vertical and horizontal dont know which ) not a huge problem, as i want digits as a mirror image anyway , this was just a test of my coms / connection, timing etc.

 

[Ian Rogers]

When you have played with this "max7219" for a bit and it keeps shutting down... Please tell me how you fixed it!!

 

[Nigel Goodwin]

When you have played with this "max7219" for a bit and it keeps shutting down... Please tell me how you fixed it!!

I've never seen that problem?, what are you powering them from?

 

[Ian Rogers]

I have a pic18f26k80 running at 3v... I have a 74hc244 converting voltage to 5v..

I'm beginning to think there is a glitch in my software... 10 mins and it shuts down, I reboot and its good for another 10 mins..
I was helping MM on another project and his was doing same... BUT!!! I have a similar clock here it has run for months with no issue.

Its got to be noise or a software glitch...

 

[granddad]

Not experienced any similar issues 'yet' I have just one 8x8 of the 4 running at lowest intensity (0x0A00 )at the moment , drive is from 3.3v dsPIC33EP512GM304, I have a TXS0108E level shifter for the signals. The RED led modules have a 6k2 resistor on the ISET pin (18), The max IC is not visible on the GREEN module.
edit miss read resistor is 62k, seems very high ? (623)

 

[Pommie]

Ian, I have a clock using a 32x8 display that has been powered on for over two years without problems.

Mike.

 

[Ian Rogers]

Ian, I have a clock using a 32x8 display that has been powered on for over two years without problems.

Mike.

LIke I said.... So have I... It has run non stop for months... The weighing system is the bit that foxes me.. Same circuit.. It was messing up when I was bit banging 3v.. So I moved to SPI and still messed up... I was convinced that the 3v wasn't properly driving the 5v MAX7219.. So I put a 74hc244 in circuit to give a perfect signal....

The difference is, the clock is updated every second... My system however isn't.. Sometimes it is a while before you weigh.. Then 25 seconds lapse between axles ( obviously to determine the last axle) then another 30 seconds before printing... It acts just like its going to sleep.

 

[Pommie]

I've just modified my clock code so it stops updating, it's now been 12 minutes and it's not shut down. About to go to the shops so will report when I get back.

Mike.

 

[Pommie]

It's now been ~40 minutes and still fine.

Mike.

 

[Ian Rogers]

Must be a software glitch..However!! We have taken enough of Grampa's thread...

I'll post when I find it.

 

[granddad]

As a closing stitch ... on my first attempt with the display (ages ago) I remember adding a 1k pull down resistor to the LOAD pin at first max chip...

 

[Pommie]

Just for information, I'm using a 32 x 8 display with a WeMos mini (so 3.3V) with hardware SPI without any problems. My displays are horizontally aligned meaning writing a byte updates a row of pixels.

Mike.
P.S. sorry for cluttering up your thread.

 

[granddad]

P.S. sorry for cluttering up your thread.

More the merrier ..

 

[simonbramble]

I played with this device extensively when I worked at Maxim. I designed an Ishihara test using RGB LEDs which was quite effective. I put 10k pull downs on the bus, since the bus can start up in an unknown state. From memory, the MAX7221 fixed this issue

 

[Ian Rogers]

I played with this device extensively when I worked at Maxim. I designed an Ishihara test using RGB LEDs which was quite effective. I put 10k pull downs on the bus, since the bus can start up in an unknown state. From memory, the MAX7221 fixed this issue

Cool Thanks for that... I never thought about SPI bus state... I will try that...

 

[granddad]

Plan B , could be C or D , the previous 4x(8x8) Green matrix looked good but was soldered into the MAX chip module. I managed to get some green and blue cc LED matrixii , and after a lot of bit juggling have my prototype working (ish) . I did add 10k pull down to the bit banged data load and clock pins. Wonder if I can scroll the green characters

 

[Tony Stewart]

This thread could go on forever lacking so many design specs