Odd LED function

Dr.Jeff

Dr.Jeff

True Classic
I received a shipment of LED light bulbs; they are a very specific design for a particular custom application. I wanted to test them to verify they functioned as desired. So I connected a pair of test leads to a motorcycle battery that is sitting on one of my work benches. All of the LEDs displayed a "flashing" behavior, alternating in a "high and low" brightness. But these were supposed to be a constant, single-brightness type....not flashing. Therefore I emailed the seller to ask if possibly they sent the wrong ones. Their reply stated they should be correct and asked me to video the effect for them to see. When I reconnected the test leads to make a video I found they no longer flashed, but remained constant as they should. This took me a while to figure out. Apparently the first time I tested them the battery was connected to a "maintainer" charger, while the second time I tested them it was not. I assume the first time the "float" feature of the maintainer was 'pulsing' the voltage (to maintain the charged battery) and causing the LEDs to 'flash' as the voltage pulsed. Does this seem correct? Otherwise I can't think why else they flashed one time and not the second.
 
Dr.Jeff said:
I received a shipment of LED light bulbs; they are a very specific design for a particular custom application. I wanted to test them to verify they functioned as desired. So I connected a pair of test leads to a motorcycle battery that is sitting on one of my work benches. All of the LEDs displayed a "flashing" behavior, alternating in a "high and low" brightness. But these were supposed to be a constant, single-brightness type....not flashing. Therefore I emailed the seller to ask if possibly they sent the wrong ones. Their reply stated they should be correct and asked me to video the effect for them to see. When I reconnected the test leads to make a video I found they no longer flashed, but remained constant as they should. This took me a while to figure out. Apparently the first time I tested them the battery was connected to a "maintainer" charger, while the second time I tested them it was not. I assume the first time the "float" feature of the maintainer was 'pulsing' the voltage (to maintain the charged battery) and causing the LEDs to 'flash' as the voltage pulsed. Does this seem correct? Otherwise I can't think why else they flashed one time and not the second.
Click to expand...
Sounds like a possible explanation. As long as they work when you connect them to a battery, you should be good.
 
I should test them again, both with and without the maintainer connected to see if the results are the same. Although the maintainer would need to be in the "floating" state again and it does not have any indicators for the charging status.

I've never seen this happen before. But this maintainer is a new one so perhaps it functions differently than my others.
 
TomasL said:
Depends on the the type of onboard driver.
A proper driver should not flicker when the input voltage changes, a cheap driver does.
Click to expand...
That is very likely the cause, as these were not expensive LEDs.

It was very odd, they pulsed bright and dim at a very steady rate, roughly full brightness and half brightness. As if they were made to do that. But the second time I tested them (without the maintainer attached) everything was steady. And this happened to two completely different designs of LED, from two different makers, so I might not have expected both designs to behave exactly the same if it wasn't due to the power source (i.e. maintainer).
 
Dr.Jeff said:
That is very likely the cause, as these were not expensive LEDs.

It was very odd, they pulsed bright and dim at a very steady rate, roughly full brightness and half brightness. As if they were made to do that. But the second time I tested them (without the maintainer attached) everything was steady. And this happened to two completely different designs of LED, from two different makers, so I might not have expected both designs to behave exactly the same if it wasn't due to the power source (i.e. maintainer).
Click to expand...
It depends on what LEDs you have. They can come in a number of configurations:

1. Bare LED
2. LED with voltage dropping resistor to control current
3. LED with voltage dropping resistor and bridge rectifier (non polarized)
4. LED with controller chip (this would have a driver and often a bridge rectifier)

If the pulse rate is not too high, you might try putting a voltmeter across the LED and see if the light output fluctuates with the voltage. The types 1,2,3 above will definitely be modulated by the change in applied voltage. An LED with a controller chip is likely to be less sensitive to the problem, especially if it is using a constant current source driver, and the applied voltage is within the compliance range of the controller.
 
dllubin said:
It depends on what LEDs you have. They can come in a number of configurations:

1. Bare LED
2. LED with voltage dropping resistor to control current
3. LED with voltage dropping resistor and bridge rectifier (non polarized)
4. LED with controller chip (this would have a driver and often a bridge rectifier)

If the pulse rate is not too high, you might try putting a voltmeter across the LED and see if the light output fluctuates with the voltage. The types 1,2,3 above will definitely be modulated by the change in applied voltage. An LED with a controller chip is likely to be less sensitive to the problem, especially if it is using a constant current source driver, and the applied voltage is within the compliance range of the controller.
Click to expand...
Honestly I'm not certain what these are. They were premade items that I found online, but the descriptions are not very detailed. Only that they were built for a 12V automotive application and have multiple LEDs on a board (referred to as a "COB" design). Of the two light designs I tested, one has a tiny module inline of the power/ground wires. Initially when I saw them "flashing" I thought that was a flasher circuit so I bypassed it, connecting the power and ground after the module. But no change. So I even removed it completely and still no change. I think the module is to prevent the LEDs from effecting the BUS communications on modern cars (which I obviously don't need). I'll try to remember to take a picture of the little module to show.

I agree I should be able to read the voltage change (if there is one) as they flash. The rate was not fast. When I get some free time I'll play with it more. But for now I think they are working as they should so I'm not overly concerned about it.
 
Last edited:
Today I retested the same LED lights with the same battery and maintainer. They did not 'flash' or 'pulse' this time, with or without the maintainer connected. However the maintainer may not be in the same functional mode it was before. There is no way of telling when it is "floating" or otherwise. I also connected a multimeter to the battery terminals with the maintainer attached. There was no variation in the voltage. So none of this told me anything.

I tried to take some close up pics of the little module that is inline on one of the LEDs. Overall it is smaller than a dime and spliced into the power and ground leads going to the light. Ignore the sloppy solders on one side (left, output side) where the leads attach. That was where I'd previously removed the module completely (for testing), then did my usual poor resolder job (like I keep saying, I really suck at soldering).

025.JPG

026.JPG

023.JPG

027.JPG


Any idea what this is? The light works the same with or without it in place. I'm guessing it does something to prevent the LED from screwing with the CAN BUS system on newer cars, but don't know.
 
Dr.Jeff said:
Today I retested the same LED lights with the same battery and maintainer. They did not 'flash' or 'pulse' this time, with or without the maintainer connected. However the maintainer may not be in the same functional mode it was before. There is no way of telling when it is "floating" or otherwise. I also connected a multimeter to the battery terminals with the maintainer attached. There was no variation in the voltage. So none of this told me anything.

I tried to take some close up pics of the little module that is inline on one of the LEDs. Overall it is smaller than a dime and spliced into the power and ground leads going to the light. Ignore the sloppy solders on one side (left, output side) where the leads attach. That was where I'd previously removed the module completely (for testing), then did my usual poor resolder job (like I keep saying, I really suck at soldering).

View attachment 57145
View attachment 57146
View attachment 57147
View attachment 57148

Any idea what this is? The light works the same with or without it in place. I'm guessing it does something to prevent the LED from screwing with the CAN BUS system on newer cars, but don't know.
Click to expand...
The unit in the first photo uses a PT4119 LED driver. It is basically a switch with some options like dimming. The output device is a power FET. I found a data sheet with a schematic and specs but all the other information looked to be in Chinese. The two lead device is a K14 Schottky barrier diode, probably wired to protect from reverse polarity. Schottky diodes are often used because their forward voltage is lower than a silicon diode. Also used for high speed switching applications but the K14 looks to be designed for power applications.

I could not make out what part numbers are on the remaining photos. The inductor could be in series with the LED and driver as this is shown on the PT4119 application circuit in the data sheet.
 
dllubin said:
The unit in the first photo uses a PT4119 LED driver. It is basically a switch with some options like dimming. The output device is a power FET. I found a data sheet with a schematic and specs but all the other information looked to be in Chinese. The two lead device is a K14 Schottky barrier diode, probably wired to protect from reverse polarity. Schottky diodes are often used because their forward voltage is lower than a silicon diode. Also used for high speed switching applications but the K14 looks to be designed for power applications.

I could not make out what part numbers are on the remaining photos. The inductor could be in series with the LED and driver as this is shown on the PT4119 application circuit in the data sheet.
Click to expand...
I'm curious, if this module is the LED driver, then why does the LED connected to this work the same with or without it?
 
You must log in or register to reply here.
Back
Top