LED Driver Schematic

Good catch, my first draft had it this way, then my clean(er) copy has the error. I misread the value of R2 as well.

Thanks! I will update.

appears to be a linear current sink - there is no duty cycle term in the power dissipation equation.
http://www.micro-bridge.com/data/add/amc7135.pdf

Your absolutely right and it makes way more sense anyway.
Conventions have changed a bit since I labeled that image. At that time led- was often called the output pad and since led+ feeds the IC and is directly connected to B+ on the other side I labeled it input +.

Even if y’all can’t agree what’s right! :slight_smile:

So when people stack the 7135 chips for more power, they’re basically changing the chip to a higher level of output as though a different chip were used? How is it that the chips can be added seemingly without discretion, in whatever number is wanted to step up output?

Wouldn’t it be nice if there were a plug in module for the blinkies labeled “pull” so they could easily be done away with without affecting the other performance of the board?

This really helps me understand the throughput, and why it’s better to have a driver than use direct drive. Thanks again.

[quote=Dimbo The Blinky]

Sorry to hijack this thread, but does C1 control mode memory on this driver? [/QUOTE]
No, It doesn’t.

[quote=Dimbo The Blinky]

IOW, can I short C1 (or replace it with some R) to eliminate mode memory on a NANJG 105C?[/QUOTE]
Nope, that won’t do it.

[quote=Dimbo The Blinky]

(Conversely and PERversely, increasing C1 to create “next-mode memory”…?)[/QUOTE]
Also, no.

PPtk

By soldering the correct star to the ground ring, you can pick your mide group. One group is 3 mode hi mid low. :wink:

How do you know which star is which group? Trial and error?

Argh! What I get for daring to dream. Thanks for waking me! Hijack deleted.

Dim.

The most important function of D1 is to provide reverse polarity protection for the processor chip…

This page shows what the stars do:

http://intl-outdoor.com/amc71353-5mode-circuit-board-nanjg-ak47c20-p-596.html

Woody

You are paralleling current sinks so each chip will adjust its internal resistance so that it sinks its setpoint current - with these chips you can get integer multiples of 0.35A. For in between values you can shunt this parallel network of ICs with a resistor.

There is another issue and that is the ‘compliance’ of the current source.

What he said. Where ever you buy the driver should have info on star connection/mode selection. If they dont, buy elsewhere. The last thing you want to do is swap a driver and STILL be stuck with blinky modes unless you have to be.

Thanks pyro for the rev protection note, forgot that little tidbit :wink:
As for the mode selection, if you buy the two group version, the stars have no effect on functionality. Only the four group version uses the stars.

:slight_smile:
PPtk

:slight_smile:

The face rub, oh Lord have mercy!

Relic, thanks for the schematic and to all who provided inputs! :beer:

Hehe… I try.

I guess maybe they’ve changed the design of the AMC7135 since I last looked at it. Perhaps 0A, 7A, 14A, etc are the only attainable current values?

I’ll take Apple, please!

:expressionless: confused a bit. Dimbulb didn’t say even integers only so isn’t he correct? Since integers are simply whole numbers 1 x 7135, 2 x 7135, 3 x 7135, etc. these produce integer multiples of .350 A. Feel free to give me the face rub if this is not so.