MeanWell LED Drivers: 3 in 1 Dimming Function.

[stardustsailor]

Quite a few led driver series from MeanWell, have the " 3 in 1 Dimming Function" :

- LPF
-NPF
-HLN
-HLP
-HLG
-HLG-C
&
-HBG .

Reference / examle ,the HLP series:

 

[stardustsailor]

Way #1 : Using a "resistor" . Dimming by resistance .

- Rule #1 : If more than one driver ,have to be simultaneously dimmed ( 'controlled' ),
then value of resistance has to be R / N ,where N is the number of drivers and R the rsistance of a single driver .

So,according to spec sheet ,if Dim- and Dim + wires remain open or on a single driver a
resistance of 100 KΩ ( 100 000 ohm ) is applied between wires ,then Output current of
led driver ( Io ) would be 100% of the rated .
Say it is 2100mA for the example.

Now if the resistance is 10 KΩ ,then the Io would be 10% of the rated .
Thus ,on the example ,210mA .

Shorting the Dim- and Dim + wires will drop the Io close to zero,
but that's not a very good idea.It might cause failure of the led driver.

That's the basic function of resistive Dimming .
The simplest to build and apply.
Mainly in two ways :
-By using a potentiometer
-By using a rotary 12-position Switch.

 

[stardustsailor]

Before going into further details some basics electric characteristics of "3 in 1 DF " ,
regarding Dim- & Dim+ wires :

Vd: 10VDC +/- 0.5 V
Id_max : 500uA (per driver )
Vdext _max : 11 VDC

Where Vd is the voltage between Dim+ & Dim - ,
Id_max is the max output current (signal current )
and Vdext_max is the max External voltage that can be applied .

Total Power dissipation of Dim circuit : 0.005 W = 5mW .

 

[stardustsailor]

There are two main types of pots :
Linear and Logarithmic.

http://www.beavisaudio.com/techpages/Pots/
http://sound.westhost.com/pots.htm

Linear type pots are the ones of interest .
Dimming with them , is smooth and ,of course, LINEAR.

Low cost Pots usually have an action of a (almost) single turn
( 0 to max resistance value )
and can dissipate about 1/4 of W ,
meaning 250mW .

( A single low -cost pot can control up to ...~ 50 drivers ,
before it's resistive material heats up and fries ...)




As we said ,shorting Dim- and Dim+ ,ain't such a great idea....
So a "Low Limiter' resistance must be set in series with the Pot .

In case of a single driver ,that Low Limiter should be 10K (Io= 10% of Io_max ) .

So If wiper goes to A ( 0 Ω ) , minimum resistance will be 10K .

*Low limiter can be placed in series after wiper ,also.

**If A and Wiper are used then action should be ,leds go brighter when pot's knob is
turned clock-wise ,and dim down with anti-clockwise motion.

If B and wiper are used then turning Pot's knob clockwise will dim the leds down ,
whilethe anti-clockwise rotation will increase driving Io up to max.

In case of multiple drivers :

Number of drivers ____Pot value___Dis.Power.______LowLimiter Value10%____
2 50KΩ 10mW 5K
3 47KΩ 15mW 3.6 K
4 25ΚΩ 20mW 2.5K
5 20KΩ 25mW 2K
6 20KΩ 30mW 1.6K

 

[Observe & Report]

As we said ,shorting Dim- and Dim+ ,ain't such a great idea....
So a "Low Limiter' resistance must be set in series with the Pot .

Why do you think so?

 

[stardustsailor]

Why do you think so?

because the Output Over Load Protection circuit ,through dimming is done ,
will force PWM and PFC control to 0% Duty Cycle operation.
Personally ,I 'm not that fond of that happening in a SPS ...
Ain't such a thing of what is considered as a 'normal operational state' ...

 

[Positivity]

One thing I noticed sds..

When dimming my lights and checking power factor correction some anomalies occurred.

Off the top of my head from what I remember..

Power factor correction on my drivers dropped from .99 to half that at about 50w out of 400w while dimming. That wasn't very comforting..

My only gripe with the 1/4w resistor was it cut it off a little too early I thought. Instead of say 350ma, 250ma or so would have been nice.

One thing to consider also sds..I've been dimming for short periods, 10min to 30min, for almost a year now. I haven't run into any problems yet.

My main reason for dimming to the lowest level every time is that it is the most comfortable level for enjoying the garden.

 

[stardustsailor]

Well ,I trust otherwise...
Neither the leds/ COBS should be dimmed lower than a certain limit ,
neither the driver(s).
'Things ' tend to be unstable then ,though the light ,might be 'reflecting' a dim ,comforting feeling ...
(I agree on that one !!! )
Shit might not happen for long-long time.
But if it happens ,just once ..
Is enough ...
Why increase the 'risk-factor ' ?

 

[Positivity]

Well that would be my concern!!

I guess when things stop working in electronics, especially in this case, maybe a potential fire.

I'm personally not worried about a driver just not working one day. Chalk it up as a lesson learned and buy another.

But if its possible fire hazard that's a whole nother' story.

But if fire hazard was a possibility wouldn't it be in the best interest of these companies selling dimmable drivers to point that out. They seem to illustrate in the datasheets the usable range of dimming.

Not trying to be a pain in the arse sds. I respect and trust your knowledge. Just trying to get a better understanding

 

[stardustsailor]

Dimming using a rotary switch ....


Another simple and low-cost dimming method is the 12-pole rotary switch ...


Along with that ,12x resistors ( 1/4 Watt ) are needed ....


Underneath ...


..the first pole should be marked as "1" .( the square hole on the pic .)


R1 is the Low Limiter ....

R1+R2+R3+...+R11+R12 = 100K Ω / Ν


For example .
Io_max of driver = 2100 mA

If_min of COB is 350mA and If_max is 2800 mA.

4 Drivers should be dimmed simultaneously..


Thus R1+R2+R3+...+R11+R12 = 25KΩ

Ιο_min should be 350mA ...

350mA is 350/2100*100= 16.66% of Io_max ...
For a single driver LowLimiter should have been 16.7 K Ω .
For 4x drivers is 4.2 KΩ .
So
R1 = 4.2K ..closest resistor is 4.7 K ..
18.8% of Io_max = ~395 mA

Posistion 1 is set at ~395mA ...

(For exact limit set R1 can be a 10K small trim pot ,Wiper with either A or B connected
)

25K - 4.7 K = 20.3 K

20.3 K / 11 = 1.845 K



1.8 *10 = 18 K
18 +4.7= 22.7 K

2.3 KΩ are missing ...

Resistors R2 to R11 should be 1.8 K
( 2100 *0.018 *4 = 151.2 mA steps )
R12 should be 2.4 K
(2100 *0.024 *4 = 201.6 mA final step )

Set currents :

1_ 395 mA
2_ 546 mA
3_ 697 mA
4_ 845 mA
5_ 1000 mA
6_ 1151 mA
7_ 1302 mA
8_ 1453 mA
9_ 1605 mA
10_ 1756 mA
11_ 1907 mA
12 _ 2109mA


........
Note of course ,you can set an 'upper limit' also ,or
of course calculate the resistances needed to get another
"pre-set " scheme like :

1_ 350 mA
2_ 550 mA
3_ 700 mA
4_ 850mA
5_ 1000 mA
6_ 1200 mA
7_ 1300 mA
8_ 1400 mA
9_ 1600 mA
10_ 1750 mA
11_ 1900 mA
12 _ 2100mA

( Do the math ,by yourzelvez .....).

 

[rufus01]

Well, why not just add a simple 2A DC analog amperemeter to the linear pot!

 

[stardustsailor]

Well, why not just add a simple 2A DC analog amperemeter to the linear pot!

If case you do that ,you do not need 2A but rather 500uA amperemeter ...

In order to measure Io of driver ,
the amperemeter should be connected in series with the driver's led output
(in series with CoB/ leds ) .

Not on the Dim+ And Dim- wires ...
(You 'll need a "scaling circuit" (op-amp / transistor ..) to do so ,
only to achieve an .....estimation of Io ... )

Cheers.

 

[stardustsailor]

Resistive dimming with either utilising a pot (smooth linear dimming) or a rotary switch (pre-set adjustments )
is mainly applied for MANUAL dimming-controlling of LED driver(s).

For AUTOMATED dimming-control ,
then the Pulse Width Modulated dimming option should be used .

Automated control can be achieved ,using the resistive way ( via digital potentiometer IC ),
but extra circuitry and components are needed.Cost is increased also.
So it stays -far - aside ...

 

[stardustsailor]

In the spec sheet MeanWell states that frequency of PWM signal should fall in the
range 100Hz to 3000 Hz .500 Hz- 1Khz will do fine .

Dimming is achieved by altering the square signal's Duty Cycle .
(The percentage of a full period, that signal stays " High " / " 1 " / "On " / +5 VDC <= TTL Level *)

For automation dimming-control 3 'sub-units' are needed :

-A Control Signal Source ( A microcontroller unit like Arduino Uno or Raspberry Pi )
- A 10 VDC power Supply
&
-A " PWM switch " .


*Transistor -Transistor Logic Level :
Low / 0 = 0 to +0.8 VDC
High / 1 = +2 to +5 VDC
Noise / Jitter / Float = +0.8 to +2 VDC

 

[rufus01]

If case you do that ,you do not need 2A but rather 500uA amperemeter ...

In order to measure Io of driver ,
the amperemeter should be connected in series with the driver's led output
(in series with CoB/ leds ) .

Not on the Dim+ And Dim- wires ...
(You 'll need a "scaling circuit" (op-amp / transistor ..) to do so ,
only to achieve an .....estimation of Io ... )

Cheers.

But Ampermeters with internal shunt should be already calibrated, or not?

 

[bicit]

Automated pwm dimming is a topic of great interest. this is great.

 

[stardustsailor]

But Ampermeters with internal shunt should be already calibrated, or not?

That has nothing to do ,with the fact that if the Amperemeter is connected after the pot ,
in series ,with Dim - & Dim+ ,it will measure the 'dimming signal's' current,
which has a max value of 500 microAmperes ...

In order to measure the driving current,amperemeter should be connected at drivers Led output circuit
(Led out + & Led out - ) .

 

[stardustsailor]

Control Signal Source

Usually a microcontroller ,has this role .
( A simple device (a small computer) ,that operates like this:
When x value on Input A ,then y value on Output B )
Depending on the programming of the microcontroller and it's 'output circuitry'
a square PWM (or even a sinusoidal or triangle or sawtooth ) waveform,
will be it's output signal.

This signal is of " Transistor-Transistor Logic " level ,
or in some cases of "CMOS Logic Level ....
.
Meaning that signal 'High' is -usually- about + 5 Volts DC .
(Like on Arduino Uno,for example.)

 

[stardustsailor]

10 VDC power Supply

If an Arduino is directly controlling a Meanwell led driver ,
via TTL PWM signal ,then at 100% duty cycle (maximum ),
the driver will have an Io = 50% of Io_max.
So...
In order for the driver to output the 100% of Io_max ,
PWM signal going to Dim+ & Dim- ,has to be +10 VDC .
Double the TTL level .

The easiest way to provide a +10 VDC power supply for the signal ,
is to make a simple circuit ,that will be powered from the same +12 VDC,
that powers Arduino and/or the cooling fans ,if any.

3 to 4 components have to be used :

-One Integrated Circuit ,called "voltage regulator" ,the type " 7810 " (+10VDC output)


-2x capacitors to 'stabilise' the voltage regulator

-and optionally one diode ( types used: 1N4001 or 1N4148 ) ,to protect the regulator.
--------------------------------------------------------------------------------------------------------------------------



Vi= +12 VDC
Vo= +10 VDC
C1 = Electrolytic Cap 85°C , 47 uF ,16/25 V
C2= Ceramic or MTK/MTP 100 nF
-------------------------------------------------------------------------------------------------------------------------
*opt. D1= 1N4001 rectifier diode or 1N4148 signal diode .


-------------------------------------------------------------------------------------------------------------------------
Pins of the IC : Input (connects to +12 V ) - Ground / "Com" (connects to Common ground of devices / subcircuits ) -Output (outputs +10 VDC )



Heat pad of the IC is connected internally to "Ground" .

 

[stardustsailor]

" PWM switch "


The cheapest ,stupid simplest and easiest way to make a "pwm Switch "
is to use one transistor and one " Base " current limiting resistor .

We 've already a PWM square signal (' digital signal ' ) source ..(say it is an Arduino Uno ) ,
which signals 'High' at +5 VDC ,and provides our dimming control.

Also we 've s +10 VDC power supply ,that is capable of fully driving up to 100% the dimming circuit ,
of the MeanWell drivers (s ) .

Last thing is the pwm switching interface ,that will "connect " thw two mentioned above sub -circuits .

A transistor used as a switch .