What's new
  • Happy Birthday ICMag! Been 20 years since Gypsy Nirvana created the forum! We are celebrating with a 4/20 Giveaway and by launching a new Patreon tier called "420club". You can read more here.
  • Important notice: ICMag's T.O.U. has been updated. Please review it here. For your convenience, it is also available in the main forum menu, under 'Quick Links"!

Dimming groups of Meanwell drivers

f-e

Well-known member
Mentor
Veteran
I have a number of meanwell powered lights, on different power circuits, and wanted a single dimmer. One I could adjust to the exact same position, with a high degree of certainty.

Enter the drawing


And that was it. It turns my lights up and down, and the displays resolution gives me repeat adjustment to within 1%. So the space is exactly how I found it, when I turn them back up to leave.

Each light still retains the individual adjustment, allowing some balancing.


There was an unexpected error, when they were used in a different manner. If my psu was set to put just 3v across the resistor, the meanwells own 10v output could start to figure into things. I could turn one up from nothing, increasing brightness due to the internal 10v, but then continued twisting of the resistor took it to the external 3v. The effect was that the light could get brighter as you turned the resistor to number 8, but from 8 to 10 it dimmed a little again.

In use, you set the external psu to 10v or more for full brightness, with the Variable resistors also on full. Then run around turning down the ones that are too bright, balancing your light. Then the whole lot goes up and down nicely using the PSU. Sorted.


You may find this isn't for you, and that your ballast can be dimmed by modifying the AC input with a more typical dimmer. But hay.. It's a choice.

Edit: I wouldn't use an AC side dimmer myself. The driver is regulating with rapid switching, and the dimmer is doing the same, turning the driver on/off rapidly. The driver must get this quick pulse of power, stabilise, and then get it's own chopping done, before the dimmer cycles back off. This can be literally impossible, and is very much luck. There is real room for error that causes aging or worse. As an example, I have a ups and a temperature sensing fan speed controller. I can't use them together. The extractor spins slow, and makes a high pitched squeal. My fan motor becomes a speaker..
 
Last edited:

f-e

Well-known member
Mentor
Veteran
Looking at the math regarding a single variable resistor for a bank of lights, I realised that leaving the supplied one's connected has the same result. In this pic, I have a number of ballasts with the supplied resistor turned to full, and they are all connected together. Then any one of them can dim the lot up and down. Here, all the lights are at 50%


That really is very simple, as all you need to work them all together, is some wire to connect them together.
 

f-e

Well-known member
Mentor
Veteran


Using a thermostat to dim your leds at high temperatures. Maybe because it's hot outside, or maybe because of failed extraction. It's fixed amount of dimming at a fixed temperature though. You could use a deadband stat to have two levels of attenuation.
 

f-e

Well-known member
Mentor
Veteran
Pics pulled from the last posts as I cleaned them up


Just connect all the controls in parallel, set to full. Then any of the controls can dim all the lights. You can even add another that comes in at a preset temperature
 

Mars Hydro Led

Grow on Earth Grow with Mars
Vendor
That looks cool. The meanwell driver on our SP3000 use the same theory. Any light can be used as master controller.
 

f-e

Well-known member
Mentor
Veteran
I used a commercial stat that has to be able to do cooling, not just heating. As we are bringing on the resistor when things get hot.


I needed terminal 1 and 2 for switching purposes. Then I wanted a resistor in one wire. I decided to put in in the red wire. So I pulled the red wire out of 2 to stick the resistor in it's place, and use a connector to put the red wire on the other end of resistor. Terminal 4 became that spare connector, after snipping a wire link where arrowed. Now the red and black wire can go off the join the dimmer circuit.

The stat used has a couple of degree's between on&off points when used around 30c. It won't really regulate well at 20c though as the difference is more like 5c between on&off (as the stats not powered)
 

Phagoos

Member
A bimetallic thermostat with an anticipator / accelerating resistance needs enough current to heat up the temperature sensor. I'm pretty sure that the 1 mA from your drivers is not enough. The typical current seems to be 100-200 mA.

I remember that when we got a new modern heating system 25-30 years ago, our old bimetallic thermostat wouldn't work properly anymore and we needed a 'modern' thermostat too.
 

f-e

Well-known member
Mentor
Veteran
A bimetallic thermostat with an anticipator / accelerating resistance needs enough current to heat up the temperature sensor. I'm pretty sure that the 1 mA from your drivers is not enough. The typical current seems to be 100-200 mA.

I remember that when we got a new modern heating system 25-30 years ago, our old bimetallic thermostat wouldn't work properly anymore and we needed a 'modern' thermostat too.

Yes that stat I used is alright around 30-35c but at 15-20c the temperature control isn't great. I actually ignore the numbers on it, it's the distance between clicks I'm interested in as I turn it up&down at around that temperature. I have 2c which wouldn't work at home where 23 is the design temp for lounging about but it needs to be 21 if your active. 2c is enough to have you altering your clothes. It's not steady enough for a single activity.

This project isn't to use your lights to regulate the heat like a heater. Though that is possible. I'm switching them off until there is a real difference in temperature. I don't expect a lot of on&off cycles per day and feel my equipment will be happier. Though can't prove it.
 
Top