Hello,
I have a magnetic revolving "levitating platform" like the one pictured.

One of the four LEDs went dim and then died, altogether. The LEDs not only illuminate whatever is on the platform, but are also guides for proper placement of the platform in the magnetic field.

I don't know anything about diodes. I checked the voltage on the "legs" of the 3 good LEDs and they are all around 2 VDC. The "legs" of the bad diode show 6 VDC. What can I do to determine whether the diode is bad or if something else went wrong in the circuit?

The platform does work with just the 3 LEDs.

Thanks for any assistance...

 

Measuring the voltages as you have would suggest that either the LED is bad or there is a bad connection to the LED.

 

Sounds like the bad LED is open circuit. Replace the LED.

 

The legs show 2V or 6V where? It's important to know where you've placed the leads. If you are just measuring on the legs, you should reverse the leads as well.

What does the base use for power! Wouldn't happen to be 9V PP3 battery with only 8V remaining. An open LED should show the (supply voltage minus 2V). Or 8V - 2V = 6V

Based on my assumptions A shorted LED should show 0V. My bet is on the open LED.

 

The legs show 2V or 6V where? It's important to know where you've placed the leads. If you are just measuring on the legs, you should reverse the leads as well.

What does the base use for power! Wouldn't happen to be 9V PP3 battery with only 8V remaining. An open LED should show the (supply voltage minus 2V). Or 8V - 2V = 6V

Based on my assumptions A shorted LED should show 0V. My bet is on the open LED.

The 3 LEDs that work (measured on the legs):
2.99 VDC and -2.99 VDC
2.35 VDC and -2.35 VDC
2.62 VDC and -2.62 VDC
(probes one way and then reversed).

The bad LED:
6.38 VDC and - 6.38 VDC on the legs.

The base uses 12 VDC (1.2 A) via power adaptor.

If my measurements indicate that the LED is bad, what specs do I look for in ordering a new one?

 

Measure the diameter of the LED and buy the same size. If you're concerned about current, trace the circuit for us.

 

Measure the diameter of the LED and buy the same size. If you're concerned about current, trace the circuit for us.

Personally; I'd buy 4 identical LEDs and replace them all in one go.

There's no component markings to identify them and check the specification of the old ones.

 

The 3 LEDs that work (measured on the legs):

2.99 VDC and -2.99 VDC
2.35 VDC and -2.35 VDC
2.62 VDC and -2.62 VDC
(probes one way and then reversed).

The bad LED:
6.38 VDC and - 6.38 VDC on the legs.

The base uses 12 VDC (1.2 A) via power adaptor.

If my measurements indicate that the LED is bad, what specs do I look for in ordering a new one?

In this picture, which is the bad LED? I'm assuming the bottom center one because it's not lit. Is there a resistor in series with the bad diode. One appears to be on the other three.

As far as a replacement you need to match the size and color. In your pics, they appear to be white LEDs. Then, to match their color generally, their current must match. This is one of the purposes of the resistor in the picture below. This discussion will assume the LEDs require 20 mA, a typical value.


Your base has four of those simple circuits in parallel. In one of them, the LED or resistor has blown.

Here by strictly calculating, it gets a bit mucky. Your supply is purported to be 12V. The LED will use up 3V of that, leaving 9V across the resistor. Using Ohms law,

R = V / I
R = 9V/0.02A
R = 450 Ω
​

The only way I see anything near 6V across the legs of one LED is if one of the series resistors went bad. Let's say it dropped to only 300Ω. It would only drop 6V at 20mA leaving 6V for the LEF and frying it.

 

​

In this picture, which is the bad LED? I'm assuming the bottom center one because it's not lit. Is there a resistor in series with the bad diode. One appears to be on the other three.

As far as a replacement you need to match the size and color. In your pics, they appear to be white LEDs. Then, to match their color generally, their current must match. This is one of the purposes of the resistor in the picture below. This discussion will assume the LEDs require 20 mA, a typical value.
View attachment 122984

Your base has four of those simple circuits in parallel. In one of them, the LED or resistor has blown.

Here by strictly calculating, it gets a bit mucky. Your supply is purported to be 12V. The LED will use up 3V of that, leaving 9V across the resistor. Using Ohms law,

R = V / I
R = 9V/0.02A
R = 450 Ω
​

The only way I see anything near 6V across the legs of one LED is if one of the series resistors went bad. Let's say it dropped to only 300Ω. It would only drop 6V at 20mA leaving 6V for the LEF and frying it.

LEDs generally go *VERY* leaky in their death-throes - but that isn't the only failure mode.............

There are exceptions - but resistors hardly ever fail low, they usually go high or completely open.

 

"Measure the diameter of the LED and buy the same size. If you're concerned about current, trace the circuit for us."

Sound's easy enough; but I don't even know what you mean by "trace the circuit." Should I be concerned about current, if I just buy one LED (or 4) on eBay or somewhere?

"Is there a resistor in series with the bad diode. One appears to be on the other three."

Yes, bad LED is the unlit one. Unfortunately, there are no (at least conventional-looking) resistors visible for me to test (see underside of the board - the bad LED is where the white-out mark is.

 

Can you get a closeup of the bad LED, front and back as well as two of a good LED for comparison. When I zoom in to see the LED connections, it is too blurry. We need a closer picture.

 

Two shots of good one first, then two of the bad one (with whiteout marks).

 

One more of "bad."
Try re-melting the solder?

 

One more of "bad."
Try re-melting the solder?

The right leg of the LED in that picture looks like it is not soldered. The back side looks like it is. Melt and add a touch (small amount).

 

I remelted the right leg along with a dab of electronic solder, but the LED still doesn't light. I tried running leads from both good LED legs and from bad LED legs to a flashlight bulb (since both are DC I thought it might work), but no light in the bulb. Could that be because the LED is still in the circuit? Or voltage too low? If I'm getting 6 vdc on the bad LED legs, I assume I must have a good connection to the board.
I took an LED out of a nightlight that wasn't working, and applied it to both the good and bad legs of the LEDs. No light. So I hooked it up directly to the 12 vdc power adaptor for the "levitator." It lit up momentarily, twice, with a little smoke, and now that LED doesn't work anymore. So that tells me that LEDs are sensitive to voltage (or amps?). The power adaptor still works. Learning on the fly, I guess I need a new LED that works at 2-3 volts.............?

 

It's not quite that easy. Can you unsolder the bad LED from the board?

I mention the bad LED because we're likely going to have to eventually successfully remove a good LED to make some measurements.

 

For heaven's sake! The guy just needs a 5mm white LED, available on eBay for about 10p! I doubt that any difference in colour, forward current, forward voltage drop, temperature coefficient or anything else will be the slightest bit noticeable!

Simple problems demand simple answers!

 

(Laughing) Yeah, I figured I'd be pulling the bad LED and replacing it. It is a simple matter, and am sorry for beating the subject to death. I'm a biochemist, not a computer techie, and am sort-of learning about LEDs as I go, with the help of the good people on this board.

I'll buy an LED somewhere, and stick it in. Then will bother everybody here again if it still doesn't work.

 

Wait - one more thing.
This table for 5mm LEDs on eBay lists voltage a little higher than I was observing (for white). Think they'll work?
And can the LED be installed in either direction? (Am confused by the "forward voltage/reverse current).

 

Wait - one more thing.
This table for 5mm LEDs on eBay lists voltage a little higher than I was observing (for white). Think they'll work?
And can the LED be installed in either direction? (Am confused by the "forward voltage/reverse current).

Green should be 2.0V - Thorn consumer Electronics used one as a voltage reference on the CRT base board on some colour TVs - apparently the green LED Vf is fairly stable.

Most LEDs are just like most diodes - they only work one way round. Polarity is indicated by a flat spot on the rim at the bottom of the package or one lead longer than the other. The cathode is usually mounted on the stumpier lead inside the package - but some early high brightness types were the other way round. The most common bidirectional LEDs have 2 different colour chips in inverse parallel - a different colour for each direction of current flow, AC produces a secondary colour.