Nice little linear LED driver
problem controling a nicrew 50W lamp
- Thread starter Sral
- Start date
- Tagged users None
Nice little linear LED driver
Perfect thanks..Here ya go...
U1-3 appear to be the same chip..
Main LED board layout:
1.1kOhm = 555 mA current.. Pretty typical.
To address an older question...a ton of those resistors are "000" .
Basically jumpers.
Last edited:
So the chip lists pwm as a dim control but no specifics.
Honestly that seems pretty common, the lack of info.
So how do you know what works?
Most pwm has a v range and a frequency range that work.
Did I miss it?
Maybe not necessary for non switch mode drivers?
Not as critical for linear drivers. There will be a maximum but not real minimum, based on the switch transistor used. Its pretty classic cheap electronics from cheap suppliers to not list it, but I doubt feeding it 50kHz is going to go well.
Sral
Valuable Member
Yeah, as long the PWM surpasses a certain detection threshold this IC probably won’t care too much. About the frequency: @Tom Bishop mentioned that flickering got worse (slower) when he fed it a lower frequency 10V PWM instead of the PCA9685 maximum of 1500 Hz, so I’m assuming that Nicrew uses them with something around 1kHz.
But is that an issue with the "ADC" board or the driver ic?
That board is expecting a DC signal.
Afaict it creates the frequency.
So instead of chopping a DC signal you chop a chopped signal.
bishoptf
Valuable Member
Yeah before I got my parts I played with it for a good hour trying lots of frequencies and what I found was that around 330 was about the best that I could do, above 25% no pulsing but lower values I could never get them smooth. It's not a flicker but more of a pulsing and it was odd, I tried high and low values but never could totally eliminate it at the lower voltages.
I think we have shown it wants a more analog circuit where the Kessils can handle either, the new layout smooths it out enough to where its not an issue at all.
Sral
Valuable Member
Yeah before I got my parts I played with it for a good hour trying lots of frequencies and what I found was that around 330 was about the best that I could do, above 25% no pulsing but lower values I could never get them smooth. It's not a flicker but more of a pulsing and it was odd, I tried high and low values but never could totally eliminate it at the lower voltages.
I think we have shown it wants a more analog circuit where the Kessils can handle either, the new layout smooths it out enough to where its not an issue at all.
It’s definitely an issue of the ADC/signal input board. If they had soldered just 2 additional capacitors to the analog signal input pins it would have worked great with PWM as well. This way it does seems to chop a PWM signal, which results in a beating signal which is visible as flicker, since it has a much lower frequency.
bishoptf
Valuable Member
@Sral I am about to create another thread but want to make sure I have all the details correct with parts listed. I tried to make it close to @Ranjib version with just the additions, let me know if you see something wrong.
Updated not sure what I did incorrectly but if I try to generate a schematic from the fritz above its looks terrible , not sure how you generate the schematic properly (maybe you have to do that manually), if anyone has any suggestions let me know.
This is what I am going to go with and call out the 10mf cap and diode parts, let me know if I missed anything.
Updated not sure what I did incorrectly but if I try to generate a schematic from the fritz above its looks terrible , not sure how you generate the schematic properly (maybe you have to do that manually), if anyone has any suggestions let me know.
This is what I am going to go with and call out the 10mf cap and diode parts, let me know if I missed anything.
Last edited:
Sral
Valuable Member
I did it the other way around a few posts earlier, e.g. made the schematic and then the fritzing from that.@Sral I am about to create another thread but want to make sure I have all the details correct with parts listed. I tried to make it close to @Ranjib version with just the additions, let me know if you see something wrong.
Updated not sure what I did incorrectly but if I try to generate a schematic from the fritz above its looks terrible , not sure how you generate the schematic properly (maybe you have to do that manually), if anyone has any suggestions let me know.
This is what I am going to go with and call out the 10mf cap and diode parts, let me know if I missed anything.
I would write 10 micro Farad rather as 10uF. Otherwise people might confuse it with milli. Apart from that it looks good. I almost missed the GND connection between top and bottom rail
bishoptf
Valuable Member
Done, made a new post. One thing that bug me was the lm2596 board thats in the fritz parts bin had the negative connector mis labled, so I spent some time figuring out how to correct that, lol. Thanks for all the help, really appreciate it.I did it the other way around a few posts earlier, e.g. made the schematic and then the fritzing from that.
I would write 10 micro Farad rather as 10uF. Otherwise people might confuse it with milli. Apart from that it looks good. I almost missed the GND connection between top and bottom rail
Sral
Valuable Member
Quick question: can you set your PWM to 0% and measure the DC voltage on the modified signal lines ? Might be interesting for people who want to use this. I'm thinking it should be somewhere around 0.6-0.8 V.
bishoptf
Valuable Member
The led's will not come on until I hit I think 7%, but again thats fine by me, they don't dim down enough to be useful for a moonlight etc. I need to find a 12v led strip and will use that to blend with and use as moon lights. Need to find something that is just running normal strip led's, but most that I can find have to many whites, I want a 10 to 1 blue to white led strip....Wow, okay, that's lower than I though. Good to know
Sral
Valuable Member
Another one: was that with or without the Jack attached ? I'm guessing it was with, so another one would be:
Set PWM to 1% and measure again.
Saying this just gave me a slight panic attack though ... that jack is directly connected to a capacitor charged with up to 10V.
DO. NOT. REMOVE. THAT. JACK. WHEN. IN. USE.
Pulling it out will short that charged capacitor to GND, possibly either you, that capacitor or your setup. We absolutely need to change that modification to include something like a 1k resistor between the cap and the signal line.
bishoptf
Valuable Member
Doesn't work that way for me since I am not using barrel jacks, I am using JST SM connectors, I hate barrel jacks. Well tell me how to modify it so it will be safer...
I was measuring at the jack end though... although I will say folks will probably unplug the jack from the lights from time to time, so yeah if we need additional protection let me know where that extra resistor needs to go.
Last edited:
Sral
Valuable Member
Interesting, what do you get for 2% ?
Regarding the safety, I was talking about the blue wire going from capacitor to the jack in the fritzing diagram. Well, to add safety is a bit tricky in fritzing, due to the limitation in placement. Personally, I would simply solder a 1k resistor into the blue wire, but that's not quite good enough for a guide I believe
Here's something one could do on the board although it would eat up even more space:
Regarding the safety, I was talking about the blue wire going from capacitor to the jack in the fritzing diagram. Well, to add safety is a bit tricky in fritzing, due to the limitation in placement. Personally, I would simply solder a 1k resistor into the blue wire, but that's not quite good enough for a guide I believe
Here's something one could do on the board although it would eat up even more space:
Similar threads
