Help with green LEDs being skipped

[Windell Oskay]

I asked because overvoltage would be one way to explain what you’re seeing.

That is the original power supply, which is good, but we recommend against using voltage adapters. The power supply accepts EU voltage (as you can see on it), and usually functions much better on “clean” mains power. You might be able to find a cheap “plug shape” adapter, which I would is a better solution.

You *may* have a partial connection between LED10 and something else. Can you use your multimeter (with power off) to check the resistance between LED10 to each of the other “LEDX” lines?

[DC]

Thanks, will do this on Thursday and let you know how I get on. Will leave you in peace for a couple of days :-)

Cheers!

[DC]

Hi Windell,

I got another European adapter, just like the one you described but the problem remains.

I checked and I can’t measure any resistance between LED10 and any of the other LEDX connections. I double-checked continuity between LED10 and the other LEDX connections and also got nothing, so no short circuit.

I also checked for shorts betweeen pin 15 on the ATMega chip and the other pins and also no problems.

I used my arduino to light up D45 and when I do this D410 and D105 also light up (although only slightly, probably as I’m only using the 3.3V output).

When D45 is ON, 4 is HIGH and 5 is LOW. So I don’t understand how D410 and D105 can both be on simultaneously. For D410 to be on, 4 is HIGH and 10 is LOW. But for D105 to be on, 10 needs to be HIGH and 5 is LOW. So how can 10 be both high and low simulataneously to create this situation? The same goes of course for the other leds which come on along side D16 and D42.

I’d be very grateful for your input again.
Cheers,
Donna

[Windell Oskay]

I would not have expected the power supply change to affect the situation.

> When D45 is ON, 4 is HIGH and 5 is LOW. So I don’t understand how D410 and D105 can both be on simultaneously. For D410 to be on, 4 is HIGH and 10 is LOW. But for D105 to be on, 10 needs to be HIGH and 5 is LOW.

That’s a slight oversimplification. D45 is a blue LED, which should have about 3.6 V across it. If line LED10 is not connected to anything, the situation is also described as having D410 and D105 connected in series from LED4 to LED5. Because the red LEDs light up with about 1.7 V, that can be enough voltage to light up those two LEDs.

Now, there is some subtlety to the design, and things usually work as intended. However, if even a single LED is damaged or a resistor isn’t connected properly, that can screw everything up quite well.

The next thing to check is (with power off) the resistance from each output pin of the microcontroller to the LED1-LED10 lines. Refer to the circuit diagram, and touch your probes to the “shoulder” of the microcontroller pin and to the end of vertical jumper atop the red ring.

[DC]

Yes, I get approx. 24.5 Ohms between each of the relevant chip output pins and the LED jumper connections on the red ring apart from LED10 which I (correctly) measure 68 Ohms. So that all looks good…

[Windell Oskay]

Well, that seems correct too.

If it’s not the resistors, then it must be the LEDs, or the connections between them.

You might further check all permutations, and make sure that there are no other odd connections between the LEDx lines.

However, I would say that one possible case is that there is an issue with those three blue LEDs. Have you done any resoldering work or relalignment on those since it was working?

[DC]

Hi again Windell (it’s that time of week!),

I double checked all the connections, checking that there were no shorts between any of the LEDs or between any other lines where there shouldn’t be connections using the circuit diagram. I couldn’t find anything wrong so in a trial and error approach I replaced one of the problematic blue LEDs and that solved the problem with those red LEDs, so I went onto replace the other two as well. I notice that when a couple of other blue LEDs light up, two red LEDs also light up simultaneously, but just very, very faintly… you can’t see this when it’s functing in time-mode only in alignment-mode. So I’ve reassembled the whole clock as I think it’s working as best I can get it without an entire overhaul of all blue LEDs which I’m really reluctant to do. I just hope that the slight lighting up of some other red LEDs in conjunction with the blue LEDs isn’t a sign of a future fault creeping in. I’ll keep an eye on it anyway.

Thanks for all of your correspondance over the last few weeks. I really appreciate the help as well as your speedy and friendly responses.

All the best,
Donna

[Windell Oskay]

That sounds quite good, actually. There can always be some little bit of ghosting that could drive you nuts if you try to fully eliminate it. It sounds like you’ve made great progress and if it were mine, I’d call it good there.

[Author]

Posts