Hmmmm, that’s very interesting indeed.
I have quite a number of those style board/led combinations and they exhibit strange flickering under certain mid level master brightness settings but NOT at full brightness!
I’d verified the on board cap to be 100nf and the res, to be 75R0 but never physically measured their connection to the LED chip. After all the cap had to be for bypass and the res had to be an inline noise value right?
Goes to show what you get when you assume…
This adds a whole new light (ahem) to these devices, as their power draw has always been on the low side. Not surprising as an inline 75R0 value means at 20ma the LED is getting 5V-1.5V=3.5V!
I wonder if these were limited to prevent overheat issues (speculation on my part)?
Or is it a simple matter of someone screwing up the PCB board design and getting the res on the wrong pins?
I’ll have to try and “fix” a set of these I have lying about to check for power draw, heating and other issues…
I have no idea why they have put the resistor there as you cannot current limit these types of LED’s like that (as we have found out!). Good design dictates that the cap should be there though so I suspect it is just poor design…
I’ve had ten of these at 100% brightness (RGB+W) for around five hours now and they do get hot (as you’d expect), but the little PCB does help to dissipate the heat.
I can confirm that jumpering the resistor eliminates the instability in my strips.
So far I’ve retrofit 25 out of a 50 LED string and the flicker issue very obviously starts up once you reach the 2nd LED after the last jumpered unit.
I even went so far as to jumper a few while “live” with a pretty bad flickering pattern (0x80808080
with 50% Master Brightness). I could see the next LED after the one I just fixed become stable suddenly, while the rest would flicker.
Did some current measurements and I’m getting about .3W/LED at full RGBW which is what I would expect. Voltage drop across 25 LED’s at full brightness is about .3V which is what I would expect and matches well with my other strings.
Looks like I’ve got some work to do to fix some strings .
I’m glad my findings have helped, it’s always good to have the fix confirmed.
Maybe this should be put in the wiki so other people can find the info easily?
Any other solution for this issue? I am using a Dig Uno and having the same issue, my led strips are BTF 144 led 1 meter IP65 I cannot solder to bypass the resistor.
Also, I have an SP108E controller which works perfectly fine with the same led strips, same wires same everything.
EDIT:
Found my solution, I was powering my Dig Uno with a USB brick 5v 2.4A … turned out to be the issue.
I connected the Dig Uno to the same power supply of the Led’s and now it is working fine, I noticed the same behavior for the SP108E, so lesson learned just use a standard power supply for the controllers too!
Glad you got yours fixed, but definitely a different issue.
Your setup sounds like a standard power issue, described many times in various messages.
One of the sure ways of diagnosing your type of issues is to set the brightness to a low (25%) value and noting that everything seems to work fine. As you increase brightness (and thereby power consumption) you’ll see the strip “break” as you increase consumption past what the PS can provide.
The issue in this thread was very different, even with proper power the strip would give strange flickering as too many data channels (RGBW) were changed often getting worse as brightness decreased. The 75R0 resistor in question was measured to be inline with the power pin on each LED. In your case, if you were to measure properly you would find the same resistor (if present) is inline with the data line. That’s normal and where it should be.
Just registered to say THANK YOU, I have the same boards with a 75R and I couldn’t find out why they wouldn’t just work right. Never considered the boards would be just weird.
FWIW, I ended up being pretty anal about a 450+ LED install I had already put up. Over a span of a few weeks I beat at it and removed/jumpered every 75R0 resistor in the setup.
It was pretty obvious after I did about 75, that this was going to solve my problems and that there was a repetitive (and tedious) process I could use to fix these.