Test/Review of 19mm CC 3000mA LED Driver 3 Modes, 5.5-12v, XM-L,SSC P7,MC-E

It it had run at 3A the efficiency would have been better and the temperature lower.

:) Yeah but it is not 3A

Aha…
So that sucks…

Surely though, a 7135 also uses PWM, but on a much higher frequency, right?

A 7135 does not use pwm, but most drivers using a 7135 adds pwm. The pwm can be anywhere from 100Hz up to above 10kHz.Using a high pwm frequency with the 7135 do have some problems, because the 7135 is not that fast to turn on, especially not when driving a pack of them from a single microprocessor output pin.

Here is an example from one of my driver tests:

The microprocessor has its output on for nearly 2us, but the led is only on for about 0.1us. The rest of the time is used for the 7135 to turn on.

Then how can a 7135 stay cool, when it’s not on or off, but ‘half way’ or something, like a resistor?

(It does amaze me though how fast a 7135 has to be, to keep up with the PWM)

It does not stay cool, but with pwm it has a long* time to cool off between turning on.

50us to 100us is a long time when the on period is 2us.

Really?
So, just to verify, a 7135 just burns off excess voltage? :frowning:
Like when you run a LED on 3 Volts, from 6 Volts battery, with one 7135, the 7135 uses as much power as the LED??
:^\

Exactly (Usual a led will be a bit above 3 volt and batteries driving a led will be below 6 volt).

If you want to measure voltage, you must do it when the driver is in max., a DMM will not show voltage correct when pwm is active.

Good thing I still have some analog meters too.
Thanks for clarifying this.

Now I have even more questions… :smiley:

I’m thinking Class-D amplifiers now, which is PWM, efficient and no ripple due to output filter…

Analog meters have the same problem, they will both show the average voltage and that is mostly useless with pwm.

I wouldn’t trust a DMM’s view on the average voltage.

For the rest this is all a bit over my head…
I find it hard to believe though that a, say, 100kHz PWM-ed power supply can’t be smoothed to its average with a coil and a capacitor, like a class-D amp loaded with a speaker.
How about this cute thingy to modify into a driver?
Probly not, but it’s tiny enough!

drivers… where do I start…? :weary:

Depends on the DMM.

A inductor improves the smoothing very much. It is often called a buck or boost driver.

For an audio amplifier you do not really need to remove the switcher frequency, it is above the audible range.

It might be possible to modify it, but it is a waste of electronic. It is much more complicated than a driver and does not deliver much power.

You can check my website about drivers, all my reviews are collected there.

Soldering on this thing looks like this guy did it:

Thanks for the review, though!
Yours are the best!

:smiley: :smiley: I cant tell you how many times Ive torn apart budget flashlights and thought the exact same thing about the chimps that must have put them together. Its sad that so many people have absolutely no pride in their work or such justifiable disdain for their sweat-shop employers.

As always, thanks for the excellent review. I almost purchased one of these cursed drivers but didnt when lck-led’s customer service refused to answer my email questions.

True, but you don’t want to broadcast the switcher frequency, speaker wires as antennas… :wink:
Hence the coil (on the example just ‘beads’ though) and caps.

Thanks for your replies.

That is correct, but not all manufacturers stays completely within the regulations. With a loose circuit board you do not even need to, because you can just say that the filter must be added to the finished product.

Any way to disable the low voltage protection? I want it to run on high until I shut it off. Nimh battery here.

Write you own program for the processor.

Dang that was quick. Disregard my pm. Well that sucks. I was hoping to get rid of it. Now my light only runs on high for about 20 minutes. If I got rid of that it would go for over an hour.

There’s probably a voltage divider circuit in that mess somewhere. If you can find the right one, removing a resistor could effectively disable lvp by tricking it.