did anyone have issue with making 8x8 led matrix work with MAX7219?
In my case, the seller just shipped wrong led type inserted into the board with 7219.
Because of incompatible pinout, all leds were always on.
After re-wiring, it started functioning, but the PCB seems to become useless:
The correct LED: HL-M1388AR. And incompatible one: HL-M1388BR. The LEDs are reversed in it.
OK, you were supplied the wrong LED matrix for that PCB. You are correct, it is the "reverse" variant, otherwise perfectly functional.
You may care to tell us who the seller was.
You may wish to see if the seller will supply an appropriate replacement. He should.
The whole assembly costs about $2 each from China, albeit takes a few weeks to be shipped. Just buy a couple more.
What do you think we can do?
Thanks for your reply, I started discussing the issue with the seller so maybe he will send replacement LEDs, will see. Concerning the reason why I posted this to the forum, it's because I was searching for the information on similar issues on the Internet but failed to find anyone's similar experience reported. On the other hand, I saw several reports of "all LEDs on" with no visible software reason for that, and it was considered the 7219 ICs were faulty. So, at the moment, I just shared my experience on this forum and was hoping to find out if sellers ship incompatible LEDs time to time or it's something unique.
Well, we shall see if any others have had similar experience.
It is a characteristic of the MAX7219 that if the "reverse" LED matrix is fitted, all LEDs will light - and this is probably not good for either the 7219 or the matrix. The MAX7221 may not do this, certainly not if not initialised.
I just received my 8x8 LED matrix with MAX7219CNG +1508, and... an HL-M1388BR on it. I bought it on amazon. I got totally mad this evening trying to understand why this was happening. As I'm not very experimented in electronics (I'm more of an IT guy), I thought I did something wrong and was about to post a new topic on the forum after 3 hours on searching the Internet when I stumbled upon this post.
Here is my simple wiring,
could you please check that I'm doing right, I'm still not very sure... Here is the code I used to test: LCDemoMatrix.ino from Arduino tutorial: LED Matrix red 8x8 64 Led driven by MAX7219 (or MAX7221) and Arduino Uno - YouTube and I also tried the version of LedControl from the 13 march 2015, in case it was the new version that had a bug.
Could you guys please confirm me that I did well and should contact the vendor for a replacement?
As you can see on the photo, all leds are fully bright, but sometimes some rows are dimmer than others, reacting somehow to the commands sent by the program, but it is actually not the expected result.
@nick_f: could you please share your workaround wiring, maybe I can also try to check and confirm that it is the wrong component.
Thanks a lot,
Could you guys please confirm me that I did well and should contact the vendor for a replacement?
So he sent you another "B" instead of an "A"? This is an absolute scream because it means he is supplying not just one or two, but all of them incorrectly. Anyone who does not complain will either have not actually assembled the device or believed that they themselves have made some terrible mistake.
Did you get it assembled, or as a kit? If a kit, then the vendor is not in a position to test each (or any?) kit, though they could set up a jig to do so with ZIF socket(s). But when they supply such a curious range of "products" it becomes pretty clear that they are not in a position to test anything - nor have the skills to do so.
As you can see on the photo, all LEDs are fully bright, but sometimes some rows are dimmer than others, reacting somehow to the commands sent by the program, but it is actually not the expected result.
Precisely what you would expect from the "reverse" display - LEDs are only not illuminated during the time they programmatically should be as the voltage is in each case, reversed.
Your "patch" wiring is correct - as demonstrated in the video. There is no other "workaround wiring".
Hi Paul, thanks for the reply.
I bought it assembled, and indeed if they tested it they should have noticed it wasn't working. Lost for lost, I rotated the display, and I get the LEDs properly switching on and off, but of course, as the PINs are not symmetric, what is displayed is totally crap, it's useless. But this hints that the matrix is of wrong type.
As your video shows, one display is always the same as the other, rotated 90˚ and with various pins swapped over.
Either use the odd wiring, or wait the wait and get one on eBay (from Alice).
To make the reversed matrix work, I swapped the pins using the following approach:
If you rotate the schematic diagram of the "reversed" matrix clockwise, what was columns becomes rows and the rows become columns. The leds are in "correct" direction if you consider the matrix at such "angle".
So, you only have to calculate the pins numbers. I recommend opening the HL-M1388BR pinout and rotate it in any editor.
Unfortunately, I don't know how to upload the images to this forum, so I'll just describe what I see in rotated schematic diagram (I don't have the experimental hardware setup nearby at the moment) and explain it.
The ex-row 8 connected to the pin 5 of matrix is now column 1. So, it should be connected to the socket at the position where the pin 13 was hooked originally.
Ex row 7 is now column 2, so the LED pin 2 goes to the 3rd pin of the PCB socket. Pin 7 goes to 4, 1 to 10, 12 to 6, 8 to 11, 14 to 15, 9 to 16.
Now the ex-columns. Ex-column 1 is now row 1. So, the pin 13 of matrix should be reconnected to the 9th contact in the PCB socket. 3 goes to 14, 4 to 8, 10 to 12, 6 to 1, 11 to 7, 15 to 2 and 16 to 5.
Concerning the slightly blinking LEDs in the fully lit matrix, I observed the same behavior too.
Please let us know if you were able to use the described approach and make you LED work.
A comment on the LED matrix rotation in the socket.
Sure, it was the first thing I considered. If all the columns were on the one side of the LED package and all the rows were on the other, the issue would be fixable by the software means. Unfortunately, the rows and columns are scrambled and it's impossible to alter the software to make matrix display properly.
Ah yes - not the same OP.
OK, nick_f clearly knows how to re-wire the matrix to make it work - it was his video to which I was referring. As you will see, he used jumper wires to make the swap over. This is not very convenient or neat, but it does work.
Presumably you both obtained these from the same dodgy seller - or is this more common than we thought?
There must be an awful lot of people thinking they have crook MAX7219s when they are in fact, perfectly functional.
Unfortunately, the rows and columns are scrambled and it's impossible to alter the software to make matrix display properly.
Unfortunate indeed. The first thing I looked at was the pinout. No help at all.
Note that there are more than one design of the module. For starters, two versions of the "old" design with exchanged location of the CS and CLK pins on the edge. But that design is rubbish anyway as it is not stackable. In the later designs, the board is the same size as the display and can even be stacked in both dimensions using pin jumpers. This seems - sensibly - to be progressively taking over, but I wonder if some variants of those just happen to use the alternate display polarity. I have swapped displays over between the two variant of the old boards, so I know they are the same.
Saw your question on Youtube, glad you fixed the issue too.
My answer was in post #8 but you provided an illustrated guide which is good.
Also, we see the sellers seems to send incorrect LEDs to many people.
Cilyan got his module from eBay, mine and yours were from Aliexpress Advanced Tech.
I discussed the issue with the seller, sent my explanations and videos but it didn't look like they really understand the problem. Still, they promised to send me a compatible LED matrices, but it didn't sound like they consider it as their fault. If you're going to ask for replacement too, you may point them to the photo of the correct module shown on the Aliexpress lot photo. It clearly shows 1388AR.
I discussed the issue with the seller, sent my explanations and videos but it didn't look like they really understand the problem.
My solution for this issue with pictures -
And as I said there, if your principal business is selling pens, iPhone cases and sexy lingerie, you may not be in much of a position to comprehend testing and Quality Control on electronic things.
If you're going to ask for replacement too, you may point them to the photo of the correct module shown on the Aliexpress lot photo. It clearly shows 1388AR.
I ask them for refund (dispute in progress) - no answer yet...
UPDATE: Seller "Advanced Tech" accepted full refund (I will keep this incompatible LED Matrix) without any comments... I hope they will not sell these items to anybody...
Update: I've asked the seller for LEDs replacement and at last I got it. So, in the end I got 10 LEDs and 5 PCBs with MAX7219 at the price of 5 kits and after more than 2 months of mailing with seller and waiting. While I was resolving the issue with this seller, I got even better kits from other seller, with smd-type components.
While I was resolving the issue with this seller, I got even better kits from other seller, with smd-type components.
Well, the better kits could just as easily be supplied with the wrong LEDs if the seller was careless.
What is good about the better kits is that the PCB is the exact size (square) of the display and the coupling pins are located so that the boards can be interconnected with jumpers and used to make proper strips and larger matrices. (It is easier if you solder the coupling pins on the back.)
I have continued to accumulate a few of the "old" boards not to make matrix displays, but simply so that I can use the MAX7219 to drive 7-segment displays and construct 4 by 4 by 4 LED cubes. Some day.