LED Matrix 5x7

[Jordy Dominicus]

Hello,

http://www.kingbrightusa.com/images/catalog/SPEC/TC20-11CGKWA.pdf
I want to use this LED matrix in a project (5 of them in total). To control the LEDs I'm going to light up every row separately (Multiplexing). There are 7 rows in total so every LED will only turn on for a small amount of time. I want to push as much current as possible through the LEDS to increase the brightness. The maximum peak current through a single LEDs is 150 mA. This is the case with a duty cycle of 1/10 and a pulse width of 0.1 ms. Now my question: What will be the maximum current through a single LED if I use a duty cycle of 1/7 and a pulse width of (1 ms or 10 ms) This information can't be found in the datasheet and I'm not sure how I have to calculate this.

Thanks!

 

[(*steve*)]

The first calculation would be power (i.e. energy over time).

So 1/7 vs 1/10 - change from 150mA to 105mA.

The change in the pulse width is a more complex calculation involving the thermal capacity of the device. I don't have time to do it at present, but if you look at other LED datasheets, they often have a graph showing max current for various pulse durations and duty cycles. Check one of these out to see how the recommended max current changes (as a multiple of the max continuous forward current) from what it shows for 0.1ms to 1ms or 10ms for an equivalent duty cycle. This may not be exact, but it should be a reasonable indication.

 

[Electrobrains]

It seems generally the manufacturers specify the maximum current at a duty cycle of 10% at 1kHz.

If you look at this Data Sheet (figure 6), Lite-on specifies the maximum current (300mA) down to a minimum of DC=2%:
https://media.digikey.com/pdf/Data Sheets/Lite-On PDFs/LTP-757G.pdf
At DC=10% the max current is specified to 90mA.

Another manufacturer (VCC) has a less favorable diagram. See this one (figure 5):
https://media.digikey.com/pdf/Data Sheets/Chicago Miniature Lamps PDFs/VAOM-(A,C)07573G9-BW32.pdf
They simply limit the max current below DC=10%

My approach: In cases where I do not know the data (no data sheet or not specified area), I have a simple way of testing components. Not so scientific, but it has worked very well:
Make a stress test yourself (warning, can be destructive):
Apply double the intended Voltage or Current for a couple of days and observe if the component withstands the test, gets destroyed or degraded.
If everything is ok, then most probably the component can be safely used with half that stress (in fact a quarter of the stress because P∝I² and P∝V²).