Jump to content
Innovative Marine Aquariums

LED MeanWell power supply?


zingtaw

Recommended Posts

  • Replies 832
  • Created
  • Last Reply
Ok, here is something I made to aid in clarification for those trying to make the pwm circuit for the "P" models.

 

555PWM.jpg

 

Black lines are component leads

Red lines are bottom side solder traces

Blue lines are top side wires

 

Use a 12v or greater (up to 32v) power supply for this.

 

 

I spent 30mins looking through diodes at radioshack and there were a lot of them. Could one of you guys clarify on which diodes I'm supposed to get?

Link to comment
Yes it's the power supply. You can discard it if you already have one.

Again yes, it's a 2.2 megahom potentiometer. Sorry for the confusion.

Thanks, just bought all the parts. I have another question, though. If the meanwell requires about 3mA to the dimmer circuit, will it get that current with the zener in there?

Link to comment
Hi,

 

I am not an expert on electronics but regarding the dimming of the D model: Would'nt it be easier to take a fix voltage regulator instead of the lm317?.

I mean: Hook up a cheap 12V powersupply to an 7810 with a pot behind it and voila.

 

Cheers,

Nikolas

Either way works. Costs about the same. Gets the same results.

 

I spent 30mins looking through diodes at radioshack and there were a lot of them. Could one of you guys clarify on which diodes I'm supposed to get?

 

http://www.radioshack.com/product/index.js...oductId=2062576

 

Those work fine.

Link to comment
Thanks, just bought all the parts. I have another question, though. If the meanwell requires about 3mA to the dimmer circuit, will it get that current with the zener in there?

 

The zener is to prevent that signal to meanwell do not excede 10v. It will block voltage if it exceds that value, but not the current. Current is limited by r3, but it will be enough to work with the meanwell.

You can forget capacitor c4. I've placed it to filter the noise of the relayit's but it's not necessary.

Link to comment
Would this not be the same as an active low pass filter?

 

post-14691-1262621544_thumb.png

 

This is basically a non-inverting op amp with a cap between the feedback voltage from the output.

 

Not sure if the power supply will recognize a rounded signal on the "P" models.

 

I am going to go ahead try this and see if it will work as I still think an op amp will give the widest range of (0-10V) signal and is the best in terms of real linearity. Would be very simple and I would be done if I had the "D" meanwells....

 

 

Well this is a bit of a cop out, but I bought some "D" analog Meanwell Eln-60-48 and did an op amp circuit on a breadboard that worked perfectly. I can run my program from the arduino and it dims perfectly. So I went ahead and tested the circuit on TINA, the Texas Instruments version of Spice and smoothed the output with the correct capacitors. I ran it for a few hours and everything was great. So I designed the pcb and send it out for production of 10 prototypes. Should have them in a week. This is the circuit. Since it is RTR Op Amp, it is truly linear so you can get to both rails and actually hit 0V and step up to the Vdd (op amp regulated power supply). So If I put on 9V PS, you get 0-9V. If I use a 12V supply, I get 0-12V. The op amp can take up to 16V. But for our uses, a 9V or a 10V PS works with the meanwells, In fact, can get the leds to run at less than .5V. So I full control over the dimming.

 

 

# Quantity Label Value

1. 1 C3 .01uf

2. 1 C4 .01uf

3. 1 R1 10k

4. 1 R5 10k

5. 1 C1 .1uf

6. 1 R4 4.02k

7. 1 VG1 0-5V

8. 1 C2 6.8uf

9. 1 V1 9V

10. 1 RTR Op Amp TLV272

 

 

post-14691-1263655113_thumb.jpg

 

The board:

 

Spark_Dual_Op_Amp.pdf

 

I am going to test with the "P" models as well, but I am back to drawing board for this circuit. It is going to be less straight forward. I was going to use a charge pump, but you cannot find one that goes below 2.5V. It does not seem to see the op amp triangle wave, and the transistor circuit tried gave a constant 10V (with a diode). Without the diode, I could only get down just below 5V and up to 10V. So once I finish this and have it above my tank, I will come up a good "pwm voltage doubler" that give a true 0-10V.

 

My boards arrived yesterday. I put components on one and it works great. I will post some pictures later...

Link to comment

Not yet. Still don't have my ALC yet. Last I checked, DA was still having a few issues getting the fade in/out working correctly with LED drivers. It's supposed to be fixed in the next firmware update.

Link to comment

is it possible to run 12 leds and a fan in series on on d model or do i need to get a separate power supply. I want the fan to come on at the same time as the leds.

Pat

Link to comment

You need to get a seperate power supply. LED drivers are for LEDs. Just connect the fan power supply to the same timer as the LEDs.

Link to comment
I'm not an expert in electronics neither, but I think that this circuit will do the job. It will give auto dimming feature to "D" version of "Mean Well". I haven't the leds yet, so I cant test it, but I think it will work.

 

You will need a 12 to 15V transformer to supply the circuit.

It will auto dimm from 0 to 10V and from 10 to 0V. You must have a timer with a relay where you can connect 0V, 12V and send it to the circuit.

After an action from the timer, the circuit will start to increase or decrease the signal to the Mean Well, at a rate that you can change with pot R2, from some seconds to more that an hour.

I've used half of a lm358 to the dimming circuit, and used the other to switch off or on the Mean Well. If you only need the dimming circuit you can use a lm741 instead of the 358.

http://www.nano-reef.com/forums/uploads/monthly_01_2010/post-49796-1264262710_thumb.jpg[/img

Got my parts today and made the circuit. It works great! The only thing is that the Meanwells don't dim all the way to zero (spec sheet says 5%), so the LEDs are actually kinda bright when it first turns on.

I also tried it on buckpucks and it works much better with them. Will dim all the way down, and turns on very slowly too.

Link to comment

Good to know that works fine. :)

I think that the fact that it is too shine when it's powered on has to do with the meanwell and not the circuit, and can't be modified. It's not a good thing, and acording to the specs, it seems that the pwm version will have the same behaviour. Anybody can confirm this please?

Link to comment
Can't wait to see how it looks like spark. Keep it up !cheers!

 

Well, here are some photos of the boards. I have been playing around with some differenet components. Tested and all work well. I was a little dissappointed in that the PCB manufacturer did not put the last layer on correctly which had the text on it that should have been on the boards. Those were the labels for inputs, outputs and ground. Does not affect functioning of the board at all. I used double the copper thickness (2 oz instead of 1 oz) on the traces to ensure good performance.

 

post-14691-1265256525_thumb.jpg

post-14691-1265256563_thumb.jpg

post-14691-1265256537_thumb.jpg

 

First time I have ever actually made a PCB board, so all in all I am happy. It was a fun learning experience.

 

You can actually use this with any dual op amp and choose the resistors and capacitors. Rail to Rail is best on the Op Amp. The loopback resistors just have to be the same to produce a voltage doubler. The inputs resistors do not necessarily have to be 4.02K and .6 W is sufficient for this particular use. I use a PSpice simulator to find the best resistors and capacitors to produce the most rounded waves on the output. I don have a scope to look at the waves, but the simulation give a good idea. I think I can see an oscilloscope purchase in the near future! The .1uf capacitors are better than the .0068 (which was a little more square than smoothed).

 

The 6.8-μF tantalum capacitor in parallel with a 0.1-μF ceramic capacitor on each supply terminal are specific Texas instrument recommendations on the Vdd input for Op Amps. This is for power supply decoupling.

 

I chose to use a 9V regulated power supply, but that can be whatever you have a need for up to 16V. But whatever you choose will be the max that the Op Amp will take you up to. 9V is fine with me even though you could take it up to 10V for the Meanwell PS. Perfect for the Arduino - Meanwell D's and I will test the more square wave with the P's to see if it is a good solution.

 

and yes I spotted the missing resistor in the photo and added it onto the pcb board...

Link to comment

Yes, but why would you want to since BuckPucks will run that number of LEDs and cost less. For what it's worth, you can run between 1 to 13 LEDs (in a series) off of each Meanwell.

Link to comment
Yes, but why would you want to since BuckPucks will run that number of LEDs and cost less. For what it's worth, you can run between 1 to 13 LEDs (in a series) off of each Meanwell.

Meanwell LPC 35-700 costs $25 vs $20 a buckpuck + a power supply (lets say around $15 more plus shipping)

And you're still limited to a max of 6 LEDs

 

Go with meanwell, better quality, more reliable, cheaper at the end and will hold double the quantity of LEDs in case you upgrade.

Link to comment

The reason I asked is because I have a spare ELN-60-48 lying around and I would like to use it on my 1,7gal tank.

 

Can I use this to run 3 Cree XR-E's? (2 royal blue, 2 white)

Link to comment

responded to your pm,

 

short answer no. it does not meet the minimum voltage requirements that evil discovered. it will over volt and potentially kill leds.

Link to comment

Archived

This topic is now archived and is closed to further replies.

  • Recommended Discussions


×
×
  • Create New...