Reading spec sheets (meanwell LDD-700LW)

Go ask your wife what you should do next.

Did you actually read the datasheet? It's pretty clear once you realize that the "levels" are just digital levels, high and low. ...Jim Thompson

I don't get what you are saying. The data sheet is clear. A valid low is below 0.5 volts and a valid high is between 3.5 and 8 volts. The fact that they don't indicate the low should not be appreciably below ground doesn't mean the limit on the upper voltage is not valid. I'm sure they pointed this out to warn a user to not use a pullup or drive from the LED power source which can be much higher than 8 volts.

What is hard to understand about this...

...Jim Thompson

Datasheets d If you want a logic 0 it has to be = 3.5V If you don't want to burn it up, it has to be < 8V

So if it's between 0.5 and 3.5 volts it is not predictable with certainty. Datasheets are about what you need to know to design something that you can be certain about. They don't say maybe anything.

Now the other issue is that you might not understand PWM. The brightness is proportional to the duty cycle, which is the percentage of time the input is a logic 1. We don't know the frequency because you didn't quote that spec but I would guess a few KHz. It would probably work over a wide range but the frequency has to be high enough that flickering isn't visible.

that's clear enough, it's the bit the OP quoted that was ambiguous

NT

So you didn't actually read the datasheet ?>:-} ...Jim Thompson

So you didn't look at the datasheet either ?>:-} (Though you clearly understood.)

Remarkable how many posters DID NOT view the datasheet... yet took it upon themselves to barf snarkily... engineers ?:-( ...Jim Thompson

looks to me like the part has 2 modes of dimming using the same input pin:

You can apply a logic level PWM signal 100 Hz to 1kHz or you can apply an analog dim voltage 0 to 2.5 volts.

Looks like only a few of the higher-current parts do that. The 700 mA part seems to not.

It's not a very clear data sheet.

I think we can all agree on that.

GH

What? Can't pin your own diapers ?>:-} ...Jim Thompson

I admit I didn't. Such things are legal documents as well as design guidance, so they need to state the basic specs plainly & clearly, otherwise my suspiciators perk right up.

NT

The analog/pwm input spec on some of the parts somewhat vague. (If I input 1.5 V what fraction of current do I get?) I certainly would want to play with one before I designed it into a circuit.

But sure I like to have a spec sheet that pins the diapers on for me. I'm less likely to embarrass myself that way. :^)

George H.

There's a graph towards the end of the data sheet.

The data sheet is a bit vague, especially for beginners, but the MeanWell stuff seems to be cheap and good. I've used hundreds of their open-frame off-line switchers. As with most things like this, the best thing to do is get one and test it.

JT thinks that gross insults are neutralized by complex smilies. Very strange behavior.

Ohh.. I missed it. (Looks like one of my pins came undone, probably won't be the last time. :^)

George H.

No - got that. I have PWM coming out of a raspberry pi, at around 580Hz, si nce the LDD demands between 100Hz and 1Khz.

I agree using a ULN2803 as a logic device isn't a great plan, but that's wh at is on a board that's already developed, built and tested. (It was origin ally going to drive the LED directly.)

So I built a voltage divider/pullup that applies a safe voltage to the base of the 2n2222 (and the 3 paralleled ULN outputs pulls it down close enough to 0 to turn it off). The output of the collector pulls down on another vo ltage divider, so I get either 7.7v or 0.01v out to the LDD, toggling accor ding to the PWM. No design prizes, but it's working.

I'm not getting anything like linear dimming control, and I had to apply a huge curve to the software setting the PWM duty cycle to even get close, bu t it's good enough.

Not too worried about fires. Everything is within spec, and nothing but the LED gets warm - and that's mounted on a heat sink about the size of my fis t. The light is on for brief intervals with long off times; the longest on time is about 2 minutes, and that period has an average duty cycle of 50%. I'm not one for getting close to absolute ratings, which is why I wanted to understand that LDD spec in the first place.

Thanks to all.

No - got that. I have PWM coming out of a raspberry pi, at around 580Hz, si nce the LDD demands between 100Hz and 1Khz.

I agree using a ULN2803 as a logic device isn't a great plan, but that's wh at is on a board that's already developed, built and tested. (It was origin ally going to drive the LED directly.)

So I built a voltage divider/pullup that applies a safe voltage to the base of the 2n2222 (and the 3 paralleled ULN outputs pulls it down close enough to 0 to turn it off). The output of the collector pulls down on another vo ltage divider, so I get either 7.7v or 0.01v out to the LDD, toggling accor ding to the PWM. No design prizes, but it's working.

I'm not getting anything like linear dimming control, and I had to apply a huge curve to the software setting the PWM duty cycle to even get close, bu t it's good enough.

Not too worried about fires. Everything is within spec, and nothing but the LED gets warm - and that's mounted on a heat sink about the size of my fis t. The light is on for brief intervals with long off times; the longest on time is about 2 minutes, and that period has an average duty cycle of 50%. I'm not one for getting close to absolute ratings, which is why I wanted to understand that LDD spec in the first place.

Thanks to all.

no... that's for models "1000" to "1500", and the analog control is between 0.5 and 2.5

his model "700' only does PWM.