Tried this circuit and am getting a constatnt 10.8V with the 10V diode and a 12V suply. With just a 12V supply and no zener diode on the output, I cannot get the output any lower than 6V. So tried a 9V supply (which is really feedng 10.8V) and I get just below 6V on the low side and just above 8V on the high side. Not exactly what I wanted, but it is ramping up and down between 6-8.5V.

As suspected, the meanwell did not recognize the triangle wave produced by the op amp design. So still looking for a good solution.

What if you series an appropriate cap to round off the triangle wave?

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

Click to view attachment

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....

Just got my ELN-60-48D today, and also am seeing about 250 mA through 6 LEDs at the minimum current setting of SVR2. I checked it with my multimeter set to amps, and also checked volatage across a 1 ohm resistor connected in series with the LED string. Got the same results both ways.
Also the LM317 dimmer circuit worked perfectly. It didn't give me good results when it didn't have a load, but with it hooked up to the dimmer circuit of the driver, I can get even dimming from 10V to about 1.5V. The current at 1.5V is about 150mA and at 10V is exactly 1 amp. The LEDs are actually quite brite even at 150mA.

Good deal. Glad to see it worked for you.

Well I about gave up up making the pwm circuit. Actually I did give up making it. After reassembling it on the breadboard and ensuring that everything was connected just perfect I was still only pulling 290 mA regardless of pot position. I could not get it to dim or brighten no matter what I did with the circuit. I'm sure the circuit will work for everyone else on here with the Meanwell ENL60-48P driver, just not me.

Reading some other things in this thread I decided to go for broke and start screwing with the internals of the meanwell and my 12v power supply. I put the 12V across the dimmer and tried adjusting SVR2 just too see what would happen. Sure enough I could run from dropout current (where the LEDs cut out so I'm calling it that right or wrong) to >1 Amp, along with noticeable dimming and brightening.

I haven't run this setup but for a minute or two so I can't confirm that it won't fry everything I have for this project and don't condone anyone else try this unless it is blessed by those with more knowledge. I'm mostly posting to see what others have to say about this setup.

You did nothing wrong. As I posted (and also Gomer), the SVR2 can be adjusted down to about 250mA. So when you adjusted it, all you did was increase the max current. I set mine by putting the dimmer at 100% on, then adjusted SVR2 to get 1000mA. That will make it so that you can then use the dimmer to get anywhere between zero and 1 amp, but will not exceed one amp.

I have been dimming my ENL60-48P with an analog circuit for about 5 months without any adverse affects.

Here's the circuit I've designed to feed PWM 0-10V to the Meanwell "P" type driver from an Arduino. I've tested yesterday with a fan and the dimming worked fine. I'll try tonight with the driver and the leds to see what happens but before that I'd like to know what you guys think about it ?

Thanks in advance !

Looks like my circuit although I used slightly different resistor v alues and fed my fans a pwm signal as well. I am now past the prototype stage and am woking on putting everything in boxes. The arduino sure makes it easy with these LED running the P driver.

I am about to order a pair of 48D's from nanotuners (and a pair of 34 D's for the wife next year ) and despite the wealth of info in this thread I am still a little confused on the dimming aspect. I have a bunch of wallwarts kicking around, can one of those be brought to 10v with resistors, then wired with a pot in-line or is that a big no-no? Could I use 1 wallwart for dimming 2 separate drivers independently?

I think I read that someone was playing around with an auto dimming of like 30 or 40 secs with caps. So when the LED's came on via a timer they would slowly get to full power instead of an instant on. I would LOVE to have that 30 second delay for both my whites and blues for both on and off . . . any idea of how that would work?

Thanks!

~Will.

I've got 4 "D" drivers for my 2 bc29 builds and am wondering if I can dim them with the LM317 circuit. It seems like everybody's using the RKL for the "D"s and am looking for something more economical.

The LM317 circuit was designed for the "D" models.

Sorry, must have misread something somewhere.

Found it a couple pages back, any idea what size C1 would be needed for like a 2 minute delay? I checked digikey for caps, but couldn't find anything suitable at .01F, any idea where to get bigger caps that will work in this application? also, could I just use 1 powersupply for 2 of these (VIN) or would I need 2 separate ones to dim 2 meanwells?

thanks!

~Will.

When I set up my 8-XRE rb array on the 60-48D, I was also getting around 220mA with the SRV2 on minimum, and dimmer input at 9.14v. I expected much more current.

I was confused at first, but I just kept the dimmer input at 9.14 (the max that my 9v power supply will put out)---- and then while watching the multimeter amperage reading, I "eeked" up onto 960mA, which is good enough for me. I still had a lot more room to turn the SRV2 up even higher but saw no need to.






Seems like you had the same thing I had---- same reading of around 220-250mA. Was this at full 10v input on the dimmer switch?


You seem to have gotten perfect results with the dimmer switch. Did you turn up the SRV2 from the minimum setting?


I have to admit that these LED's aren't cheap, but MAN are they AWESOME and great ways to light a tank.

Yes, I turned the dimmer all the way up (10V), then adjusted SRV2 to get 1000mA. I also cranked up the voltage SRV1 all the way. I'm running 14 LEDs off it right now at 1000mA and it's putting out 50.2V.

Out of curiosity, why did you crank up the voltage? I thought the driver automatically sensed the voltage need and adjusted accordingly.

Were you getting lower-than-desired voltage readings without adjusting the SRV1?

Yeah, the driver will only put out the required voltage. The SRV1 adjusts the max voltage it will put out. Not really sure why it's adjustable, as I don't see any usefullness to the feature...
When I hooked up the 14th LED, the voltage was maxed out at like 48V, so I wasn't getting 1000mA. I just turned SRV1 all the way up and problem fixed. I have heard others caution against this, but the meanwell spec sheet says voltage is adjustable up to 52.8V, so I saw no problem with adjusting within the specified limits.

If you get this 1F capacitor, and use around 500 ohm resistor for R3, I believe it would take about 15-20 minutes to charge. Can anyone verify this?

Thats a 1uF, not a 1F . . . pretty big difference . I remember years ago when the 1 farad caps came out for car audio (to keep your lights from dimming when the bass hit) and they were the size of a couple soup cans stacked up . . . are newer ones any smaller?

~Will.

I also thought it must be 1uF, but the product info says 1 Farad, and the data sheet says 1,000,000 uF. Where do you see that it is 1 uF?

*Edit*
Ok, I see it in the product title. But in two other places it says 1 Farad, so which is correct?

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

Click to view attachment

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....
[/quote]

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


Click to view attachment

The board:

Click to view attachment

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.

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

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.

Click to view attachment

I'd like to try that circuit, but am confused about the stuff on the left. Is that just the 12VDC power supply?

And is that a 2.2 Megohm pot?

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.

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?

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?

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



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

Those work fine.

thanks! I'll try radioshack again

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.

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

Any updates on ALC module and Meanwell????

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.

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

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.

They released a beta firmware, and some people have claimed it works...
Forum support thread for ALC Module

Nick says that the Beta Firmware works.

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.

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?

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.

Click to view attachment
Click to view attachment
Click to view attachment

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...

Can I run 3 Cree XR-E LEDs off of a meanwell eln-60-48??

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.

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)

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.

Correct. 7-13 LEDs on an ELN-60-48 (down to 6 with a little work).

How do you think the ELN-30-27D will do with 6 cree LEDs?