MobiFlight Community Support

Hello together,

I have an electric problem with my 7-segment displays. I have build a pcb with the displays soldered onto it. When I now try to show up the numbers some of them aren't shown correctly. You can see this for example at the number 1. The upper part of the number which is the B-segment of the display isn't shown. This dowsn't seem to be a general problem with that segment because it lights up when I try to show a different number on it. For example the number 4 in the second picture. Can anybody give me a hint what could be possible error that I might look after?

Bad solder work !

Best case... Your lines are not tidy and touch each other..... Or your solder Points are not good.
Whorst Case.... You solder the Max7219 chips to hot and damage the Chips itself.

Does your 5V power supply deliver enough current? The 7 segment displays are quite 'hungry'. For instance when I connect a module with 8x7segment I get similar behavior, some work, some don't, the cause being that power from USB2 port was not strong enough. Need a separate power supply.

Thats not correct.
A 7 Segment if you set it correctly need pretty less power. Never problematic.
Here in forum are users who run 12 x 8 digits on a Board with USB power and have absolutly no Problems.

Sure... if you still use a lot of LED or a Motor on same board then its gonna be critical..... whatever just in case of the other devices NOT in case of the 7Seg.

If you realy think you got a power problem ( whatever i not can belive this, expect you did a fatal error in power planing on the mega) then simply check the supported Power on 5V Input Pin of the both Max7219 Chips.

Well, I had troubles and once I used a separate supply for the display the troubles were over.

I'll do some current measurements tomorrow to learn more on power consumption.

Not missunderstand.... I trust in you !

But i think your problem was NOT the Display..... Sure the Display is the device who showed at first a mailfunction if power is problematic.... A LED would simply a bit darker but the 7Seg will result in issues and wrong Data.
BUT
In that case the correct workflow will be NOT to power the Display external..... here you mut check the Overall Power on the Mega and normaly you should external power finaly the device who need to much power..... And thats not the 7Seg i be sure.

Simply rule.... Not use more then 800mA on a Mega at any case.... Better not use more then 500mA on a Mega to get sure USB power is enough !
And sure.... If you use a High number of Megas or other Powerneed Devices on your USB..... then get sure the PC PSU support enough power OR simply used powered USB Hubs !

I did a current measurement today. A display that shows 88888888 has a current draw of 191 mA.

USB2 is specified for max 500mA. I'd advise to use max 2 displays when on USB power. If more displays are needed, external 5V power is the way to go.

Indeed this does not explain why I have troubles with a display not working correct, while there's only one Arduino connected to the USB port and just one display is connected to the Arduino 5V pin ... nothing else. Maybe the USB port of my laptop is not according to spec, I don't know. Anyhow, when I use an external power supply the problem is gone.

Here is something wrong i think.....

https://arduinoplusplus.wordpress.com/2015/09/12/max7219-and-led-matrix-power-requirements/

I never check those tubes myself..... But as i know we normaly talk about less then 40mA per Board..... And in regulary usage you not show all 88888. So the "average" power is less then this..

EDIT I also wondering about the Multiplex Factor..... Technical the Max only light up ONE Digit at a time..... So there should be no big difference whatever you use 1,2 or 8 Diggits..... The only difference occure if your Displays show a "1" with 2 segements or maybe a "8" with 7 Segments on....
Interesting factor we should research maybe in the future !

Hi,

I changed the standard 8-digit-displays with 5-digit displays and I use 14 of them in two chains (8 + 6) without external power. I also modified the built in resistor of 10 kohms to 47 kohm to reduce the current.
With the built-in resistor (10kohms) sets the typical current per segment to 40mA. This is the upper limit (look Datasheet for MAX7219 page 8). Otherwise you have to use special segment drivers.
Max7219 Datasheet is available here: https://datasheets.maximintegrated.com/en/ds/MAX7219-MAX7221.pdf

The digits of every display are multiplexed with a frequency of nearly 800Hz to 1kHz.So of every display only one digit is switched on at the same time. The others are off. So in my case thera are only 14 digits of 70 digits are on at the same time. I do not have any problems with the power

I agree something strange is going on, I just haven't figured out yet what.

First there is this USB port of my laptop. Even if the display draws 200 mA it should be able to deliver that, but it seems it doesn't. Only the Arduino and one display were attached, no additional loads.

Second, the 200 mA is questionable? I measured this with only 5V and GND attached and without input signals all the segments light up. Not sure if it properly does the multiplexing in that case. I'll do a new measurement next week with the display connected to the Arduino and the display showing 12345678.

I might change the resistor to 47k too, the LEDs are quite bright as they are now. I know there's a brightness control in Mobiflight, but the display chip probably uses PWM which means the peak current still is high.

Today I connected the Arduino and a display to my desktop PC. Same issue ... the display doesn't work.

The current meter is at the hobby club, but I did a voltage measurement: as soon as I connect the 5V wire of the display VCC drops to 3.5V. When I use an external supply to power the displays all is well.

Could it be that I have a batch of 'strange, current hungry' displays? Or does the 5V Arduino pin behave strange? I don't know.

Luckily all works well with an external power supply and nothing gets hot, not the display chips and not the power supply, so I bet I just live with this.

You say the magic words..... It work and so it´s fine !

Just take care about one fact..... If something goes wrong in the future.... For example you set another Displays in a row and they stop working or you use a motor and it stottering, a button that send wrong commands.... a encoder that jumps or so.... Please AT FIRST disconnect your "special" Displaysystem and test again before you search for the issue or before you report a problem here in forum.
In easy words.... You got a strange factor that is solved just temporary.... So always get in mind this can occure in other issues aslong you not solve it completly.

Summary: I would also use the power suply and life with it.

Last note about situation..... If you like to do more research.... Test your hardware on a other computer ( PC instead laptop e.g.) ..... Test another Displays in the future and also test another Megas. I would say one of these 3 elements is the problem.

Hi Rudy,

if you measure Vcc against GND you should have +5V. If you measure the Signal wires pay attension: this is not a DC-voltage it's AC because the wires were pulsed. What you measure is the proportion of DC in the voltage. Same with the current. It makes only sense to measure the Vcc-wire.

Regarding the USB of your laptop look into the manual what the USB port can deliver. Standard for an USB port is 500mA. If you need more then you need an external power supply.

Thanks for all the support and for thinking with me guys, it's highly appreciated.

I'll do a couple more swaps and tests and measurements. Maybe it's even as simple as a flimsy Dupont wire. I've had that before ... searched for an hour in my software until I went crazy ... it could not be the software. It turned out to be a flimsy wire that sometimes made contact and sometimes didn't. Since then I always test every Dupont wire before use, but such test does not guarantee the wire is ok, maybe it made contact at that moment but not later on.

I just had to get to the bottom of this, so after dinner I went upstairs again to do measurements and swaps.

First I measured the current from the external 5V supply. I needed a third hand to hold the multimeter probes against the wires and control the mouse at the same time ... but somehow I managed.

8 mA for two display modules with all segments off.
20 mA for two modules, 3 + 3 and 5 digits with random numbers, MF brightness slider on lowest position, 10k resistors on the print.

This sounds like how it should be. This could for sure not cause any problem. But even the earlier measured 200mA with all segments on max brightness USB should be able to handle that. The cause of the troubles must be something else.

I connected the displays to USB power again via the 5V pin on the Arduino. It worked, but after a while, when rotating an encoder, one display stopped working. Why?

I started to exchange Dupont wires. I had 5 red wires in use for 5V connections. I exchanged them all and ... suddenly I had a stable situation.

I had measured the Dupont wires before use because of an earlier bad experience with them. But apparently Dupont had me again. I measured all 5 wires and one of them was flimsy, but in a way that was hard to detect. A quick 'beep' measurement would say it's OK, but a resistance measurement showed it had a higher resistance every now and then, when moved. So, when mounted, it might work, and once I touched the encoders to change values the wires would move and resistance could change. It can explain why earlier the Vcc voltage dropped when I connected a display ... connecting it made the wires move.

Pfew, glad I found it. I learned a few things on the way:
- Display current is less than 20 mA each (for 5 digits) so I can safely use the intended 4 modules on USB power.
- Dupont wires can always fool you.

Thanks for your continued input. Videos are in the making on Mobiflight and the positive athmosphere here on this forum will be mentioned.