1. Home
  2. Electronics Design
  3. Reading spec sheets (meanwell LDD-700LW)

Reading spec sheets (meanwell LDD-700LW)

I'm trying to design an interface to a meanwell LDD-700LW, a constant curre nt regulator for LEDs. I'm having trouble understanding the spec sheet. (I' m no guru and spec sheets often confuse me a little).

It has a PWM input to provide dimming. If left floating, some internal pull up causes the device to go full on. If pulled to ground, it turns the light off. Cycled at some rate, it provides dimming. Simple enough.

The device gets fed (in my application) 20v or 24v, depending on where I ho ok it up. Well within range. But the PWM input is marked differently:

Leave open if not use Power ON with dimming: DIM ~ -Vin >3.5 ~ 8VDC or open circuit Power OFF : DIM ~ -Vin < 0.5VDC or short

I can't figure out if that's trying to say the safe range for turning the L ED on is 3.5-8v (and bad things happen above 8v)... or if it's saying that the voltage needed to turn it on is, eh, maybe 3.5v will do it, you might n eed as much as 8v...

Leaving me confused. If I want reliable operation, do I feed it a little ov er 8v, or strictly less? Clarity appreciated.

For the curious, here's why I care. On/Of/Dim is to be controlled by a micr oprocessor, via a ULN2003A (originally, the ULN2003A was going to drive the LED directly). So I was planning to use the ULN2003A to pull the LDD's inp ut to ground when I want the light off, and do PWM through it as needed. Th e pullup voltage would be the LDD's problem and all I had to do was pull do wn. Simple, right? Except...

If the microprocessor fails or reboots, the ULN will stop conducting and th e PWM input will go high, turning on the LED. NOT what I want. The uncontro lled state has to keep the LED off. Maybe I just bought the wrong part, but since I have it I'd like to make it work. (If someone knows of an equivale nt that is active low, mention it!)

So my thought was to use a 2N222A as a logic inverter. Pull up the base so it conducts by default. Use the ULN to pull the base back to ground as need ed. Use the transistor's collector to pull down on the LDD input. If the pr ocessor doesn't boot, fine, the ULN won't conduct, the NPN will, and the LD D won't. All good.

It didn't work - the transistor never conducts - and by fussing I found I ( apparently) needed to tie the transistor's collector to +24v via a 68k resi stor to load it. Great, now the transistor will turn on, but when it's off that means the +24v is being applied to the dimmer input, and see above for why that might be a bad idea.

I'd like this circuit to work reliably for years, so I'd like to understand what the part needs. It's easy enough to make voltage dividers to make the high signal be whatever voltage it likes, and I suppose I could just aim f or 8v and hope. (I'd rather not because the system voltage does vary a bit, so the "8v" would drift up and down a little).

If someone can definitively interpret the spec - or maybe better, suggest a n active low equivalent that will handle, say, 16v-25v and provide 700ma to a 12v Vf LED, that would be a help. TIA.

(Yes, I can just stick a fat resistor between the ULN and the LED and have done with it, but ugh. I want the LED to last for years; the darn thing cos ts $16 and will be annoying to replace.)

Reply to
Scott M
Loading thread data ...

rent regulator for LEDs. I'm having trouble understanding the spec sheet. ( I'm no guru and spec sheets often confuse me a little).

llup causes the device to go full on. If pulled to ground, it turns the lig ht off. Cycled at some rate, it provides dimming. Simple enough.

hook it up. Well within range. But the PWM input is marked differently:

LED on is 3.5-8v (and bad things happen above 8v)... or if it's saying tha t the voltage needed to turn it on is, eh, maybe 3.5v will do it, you might need as much as 8v...

over 8v, or strictly less? Clarity appreciated.

croprocessor, via a ULN2003A (originally, the ULN2003A was going to drive t he LED directly). So I was planning to use the ULN2003A to pull the LDD's i nput to ground when I want the light off, and do PWM through it as needed. The pullup voltage would be the LDD's problem and all I had to do was pull down. Simple, right? Except...

the PWM input will go high, turning on the LED. NOT what I want. The uncont rolled state has to keep the LED off. Maybe I just bought the wrong part, b ut since I have it I'd like to make it work. (If someone knows of an equiva lent that is active low, mention it!)

Post a link to the spec sheet maybe. (I'm lazy) Can you hang a resistor to ground on the pwm input and then drive the input high with your uC? George H.

o it conducts by default. Use the ULN to pull the base back to ground as ne eded. Use the transistor's collector to pull down on the LDD input. If the processor doesn't boot, fine, the ULN won't conduct, the NPN will, and the LDD won't. All good.

(apparently) needed to tie the transistor's collector to +24v via a 68k re sistor to load it. Great, now the transistor will turn on, but when it's of f that means the +24v is being applied to the dimmer input, and see above f or why that might be a bad idea.

nd what the part needs. It's easy enough to make voltage dividers to make t he high signal be whatever voltage it likes, and I suppose I could just aim for 8v and hope. (I'd rather not because the system voltage does vary a bi t, so the "8v" would drift up and down a little).

an active low equivalent that will handle, say, 16v-25v and provide 700ma to a 12v Vf LED, that would be a help. TIA.

e done with it, but ugh. I want the LED to last for years; the darn thing c osts $16 and will be annoying to replace.)

Reply to
George Herold

to

formatting link

I don't think it says more than I quoted above.

I'd rather not redesign the (bought, built and working) board that ties the processor to ULN outputs. I'd like to consider everything up to and includ ing the ULN to be carved in stone. But everything after that (the NPN, the meanwell device itself) is negotiable.

Reply to
Scott M

I believe the ULN2003 Vsat is like 1.25v or something, thats > 0.5 for a full off. A 2n7002 would be better.

Cheers

Reply to
Martin Riddle

[snip]

The way I read it: You apply a PWM signal whose high level is >= 3.5V and whose low level is < 0.5V... that is, a 0->5V pulse would do it... wider would be brighter, narrower would be dimmer, no pulse (sitting at 0V) would be OFF. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
     It's what you learn, after you know it all, that counts.
Reply to
Jim Thompson

Oh, I thought it meant whatever you do we can say you shouldn't have done that.

Seriously it's just meaningless as a spec.

NT

Reply to
tabbypurr

I didn't find it that bad. For a person with no electronics experience I agree it would be daunting. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
     It's what you learn, after you know it all, that counts.
Reply to
Jim Thompson

5V

That doesn't explain what happens at 8v, or why they listed the 8v at all.

I've fiddled with the circuit some and it seems to work ok if I keep the vo ltage near 8v. If it becomes unreliable, the backup plan is to control the LED direct1y from 3 ULN outputs and a 12 ohm resistor.That's at least simpl e; controlling an LED wasn't supported to turn into a weekend project.

Reply to
Scott M

Max you should apply.

My week-end projects always turn into 3-trips... 3-trips to the hardware store ;-) ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
     It's what you learn, after you know it all, that counts.
Reply to
Jim Thompson

Are you PWM-ing the input AS SPEC'd or just applying a DC voltage?

I'd check with my home-owner's insurance to make sure that electrical hacking is covered >:-} ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
     It's what you learn, after you know it all, that counts.
Reply to
Jim Thompson

Scott M wrote on 12/3/2017 1:48 PM:

This is a module data sheet, not a semiconductor. They assume you have some familiarity with the LED lighting world and are using off the shelf components and not rolling your own light dimmer. I think it is pretty clear that 8 volts is an absolute maximum and that it is intended to be driven by TTL or similar levels.

Which voltage did you keep near 8 volts? The power source or the control voltage?

The circuit I would recommend for MCU control is a pair of 2N2222 transistors (in common emitter configuration) with the collector of the first transistor pulled up to the LED power source (not more than the 30 volt rating of the 2N2222) through a 33 kohm resistor. If the MCU goes offline and the control output drops to ground, the first transistor will release the pullup turning on the second transistor grounding the control input to the LED power supply.

As someone has pointed out the ULN2003A is not good for this design because it has too high a collector saturation voltage to meet the spec of the LDD-700LW control input.

I'm not sure if it was easier to write all this or to draw the circuit in ascii art.

--

Rick C 

Viewed the eclipse at Wintercrest Farms, 
on the centerline of totality since 1998
Reply to
rickman

Yes.

No.

Sounds like 5 volt level is good. I'm surprised they didn't allow for

3.3 volt logic. 5 volts is so last century. 
--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

The OP seems totally lost to the fact that the required control signal is NOT DC, but a PWM signal. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
     It's what you learn, after you know it all, that counts.
Reply to
Jim Thompson

Three trips to the hardware store is the "minimum load" for weekend projects. :(- Don't ask me how I know....

Ed

Reply to
ehsjr

I was reordering for this old board, and had a hard time finding a 5V xtal. ~lifetime buy.

George H.

Reply to
George Herold

PWM is DC while it's high or low.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

Few modern FPGAs can handle 5V bank supplies. On the Zynq, things get restricted if i/o banks are powered from 3.3. We are tending towards

2.5 or 1.2 volt logic. Core voltage is 1.0.

That all becomes a problem when driving 5V or even 3.3V CMOS glue logic or SPI things.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

So sayeth the banty rooster...

The doofus (aka OP) is feeding it ~8V, which is ~100%, but the doofus is too dumb to know that's equivalent to "float". ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
     It's what you learn, after you know it all, that counts.
Reply to
Jim Thompson

He asked a reasonable question in an electronics forum. Does that make him a doofus?

What about people who ask about hinges and kitchen cabinets and other trivial job-jar issues in an electronic forum?

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

He asked. I answered. He chose not to believe my answer... including an additional number here who can't understand the datasheet.

The banty rooster bloviates on... and on... and on... ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
     It's what you learn, after you know it all, that counts.
Reply to
Jim Thompson

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.