7 Segment Displays keep burning out on me

[mistermoses]

I'm a teacher and I have 70 Elegoo UNO R3 arduino kits from amazon. The 7 segment displays kept burning out on me last year, so I instructed my students to use 330 ohm resistors instead of the 220 ohm resistors suggested in the elegoo instructions, and I ordered a new class set of 7 segment displays from amazon. Now these new ones are burning out at an even more startling rate. Frustratingly, it is even seeming like some LEDs have burned out between last school year when I tested them all and the start of this unit. I'll have students using a display in one class period, and then in the next class period it will be burned out for the next student, both with the same circuit. My students are not observing the lights being on for a second and then turning off when burned, except in the rarer cases where my students do mistakenly have the wrong circuit.

My students are using one 330 ohm resistor on each of the LED pins, for a total of 8. I am confirming that their wiring is correct. Some of my units that have burned out are labeled with HG5611AH, 5611AH, CL5611BH, 5161AS. I am aware that some are common cathode and some are common anode, but there doesn't seem to be any correlation between units burning out and their serial number.

I am at my wit's end with these things. Should I start using 1000 ohm resistors? Should I suspect that somehow the arduinos are sending spikes of voltage above 5V? I haven't found any suggestions online except to be conservative with resistors.

[UKHeliBob]

Please post the schematic of one of your project that has damaged your displays. The schematic should show all components, how they are powered and connected to the Uno.

Please also post a sample sketch that uses the hardware, using code tags when you do

[mistermoses]

This schematic is electrically identical to what all of my students have been using, except that the resistors that we have been using all year are 330 ohms, and when students use common cathode displays the wire connected to 5V would be connected to ground.

Here is a sample snippet of some code that my students would be using to turn a Serial input into a number on the display.

void setup() {
  pinMode(2, OUTPUT);
  pinMode(3, OUTPUT);
  pinMode(4, OUTPUT);
  pinMode(5, OUTPUT);
  pinMode(6, OUTPUT);
  pinMode(7, OUTPUT);
  pinMode(8, OUTPUT);
  Serial.begin(115200);
}

void loop() {
  if(Serial.available() > 0) {
    int a = Serial.parseInt();
    if (a == 8) {
      digitalWrite(2, HIGH);
      digitalWrite(3, HIGH);
      digitalWrite(4, HIGH);
      digitalWrite(5, HIGH);
      digitalWrite(6, HIGH);
      digitalWrite(7, HIGH);
      digitalWrite(8, HIGH);
    }
  }
}

[mistermoses]

I've just had a thought. I'm aware that supplying 5V of power to the display without a resistor will be too much for the LEDs and can burn them out. I have been warning my students to avoid this, and checking their circuit for this possibility each time they inform me that a display is not working so that I can be aware of the mode of failure if it happens. However another mistake that they often make while testing their displays is connecting power to pin 3 and ground to pin 8 [as shown below], almost always with a 330 ohm resistor on the 5V line as they were instructed to have. Could this do damage as well? My own intuition tells me no, that the current should just pass through the display and that there would be no voltage drop over the LEDs, but I am struggling to think of a scenario that could cause this widespread failure.

[camsysca]

@mistermoses

HG5611AH

HG5611AH datashee:

[6v6gt]

I don't imagine that "normal" usage would account for the high degree of failures of this display type. Much more plausible would be an (accidental) over-current, that is a direct power supply connection between the anode and cathode without any series resistor.

I'd suggest changing the circuit diagram to include a say 100 ohm resistor in the common path instead of a direct connection. This will have a (slight) adverse effect that the brightness of the segments will diminish as more segments are lit. That is a "1" will be brighter than an "8". But that is maybe not so important for a class setup.

Of course, if malicious students take pleasure in destroying the equipment then you have to seek other solutions.

[van_der_decken]

5611AH's use 10mA under test conditions, so with a Vf of 1.8V, that works out to 320R series resistance needed @ 5V. So 330R should be perfectly safe.

IF the displays are meeting the datasheet specs.

Off the top of my head, I think there's three possibilities.

1. The displays you're getting from Amazon aren't capable of handling 10mA. In which case a higher value current limiting resistor might do the trick.
2. The always present user error, and someone hasn't used a resistor.
3. What @6v6gt just brought up: not user error, but user maliciousness.

Edit: looking further down the datasheet, I noticed a spec of "Power Dissipation Per Dice". Assuming that was supposed to be "die", that could be the total power dissipation that the display can handle. It's rated at 100mW. If we assume Vf of 1.8V, If of 9.7mA per segment (330R, 5V), that's 17.5mW per segment, or 140mW per display.

If that's the problem (and it's a big if at the moment), using larger value resistors would do the trick. 470R or higher would get the displays under 100mW total power dissipation.

[gilshultz]

Contact the vendor and see if you can exchange them or get a refund. In the near future get a display from another vendor and see if it is a display problem.

[mistermoses]

@mistermoses

HG5611AH

If it's the case that the units cannot handle a total power dissipation that matches their rated current and voltage under normal operation, that is very frustrating. I hope that this is not the case. I tried to find this information on the datasheet of the other models that have failed, and I didn't see this metric.

I am very confident that my students are not doing this maliciously. It is happening to a wide variety of highly motivated students, and if they did want to be destructive they already know that they can make the plain LEDs explode - and they don't.

It's entirely plausible that the ones sold on amazon are underperforming their specs. Unfortunately I'm limited in my options for what retailers I can get my school to make a purchase from. Do you have any recommendations for where I should source more reliable 7 segment displays in a quantity of 100?

[xfpd]

@mistermoses

reliable 7 segment

I have had ONE bad, electronic, part in hundreds of orders from Amazon. I trust Amazon.

[camsysca]

To test - replace the 330 ohm resistors with 1k resistors. View the display in dim light. If the displays survive, yeah, it's a device that isn't up to snuff, but there's nothing else wrong. Seriously, I doubt a broad range of devices fail similarly.

One thought. Check your 220 and 330 ohm resistors. Make sure that's what they are. Not an insult - I've seen "brown" lines that were nearly indistinguishable from black lines on other resistors. It's an incredibly remote possibility you've got mismarked components.

[van_der_decken]

Following a hunch, I looked up some red 7 segment LED displays on DigiKey. Using their filters I selected Vf of 1.8V and If of 10mA. All of the displays I saw had similar power dissipation values, with 100mW being towards the high end. So, upon investigation, I can't imagine that value is for all segments combined. It's got to be for one.

Edit: and after looking at umpteen different data sheets on DigiKey and finding many that spelled out that the power dissipation rating was for a single segment I'm confident that yours is the same. I think you can safely take exceeding power dissipation off the table as a cause.

For simple stuff like LED displays and what not, I've personally had no reason to distrust Amazon. Although... you could order a couple of displays with similar specs from a distributor like DigiKey and try them under the same conditions as a test. If they fail too, it's not the displays. And if they don't fail... DigiKey's not that much more expensive than what I'm seeing on Amazon.

My purely irrational guess is that it's not the displays. There's something else we're not seeing here.

[6v6gt]

CL5611BH CL5611BH - 0.56-inch Red 1-Digit CA LED 7-Segment Display - XLITX Technology

The absolute maximum reverse voltage is 5v according to the data sheet but with a Uno you will not exceed this even assuming an accidental polarity reversal. However, it does seem close to the maximum.
A silicon diode (say 1n4007) in series with the common could lower the risk.

[ShermanP]

When a display fails, do all segments go dark, or just some of them?

+1 on checking the resistors.

[van_der_decken]

I wondered about that reverse voltage rating too. But when I looked up comparable displays on DigiKey, everything I looked at had the same value. It's a possibility though.