Vero 29 LED

@PICOGRAV

A list of meanwell drivers:
http://www.meanwell.com/product/led/LED.html

The one Supra was referring too.
http://www.meanwell.com/webapp/product/search.aspx?prod=hlg-185h-c

With that model you can also use resistance dimming as well as 0-10v dimming. Or in the case of SDS designs, a bit of both.

Those power supplies you have are pretty cool, but probably overkill for the typical user and huge. Meanwell switching power supplies are simpler to instal and much smaller for those with limited space.

SupraSPL said:

These are the ones many of us are using lately. I tested the HLG-185H-C1050B at 93.5% efficiency but it was powered by 120V and should be slightly higher with 240V.
http://www.meanwell.com/search/hlg-185h-C/HLG-185H-C-spec.pdf

We can run (5) CXA3070s in series at 1050mA or (3) CXA3070s in series at 1400mA, so that works out to as little as $15 per COB. I will be using a 350mA version to run the whole veg room. Very handy that they are dimmable by internal pot or external pot, and they are all power factor corrected.

Click to expand...

That's cool. They are far too noisy to be used for any radio or audio applications. I wouldn't want a room full of these in my house pouring all that noise into my clean AC.

I would take the big guns and go with the Vero 29's and linear PSUs. I could build a PSU at 80 to 85% efficiency no problem if it was connected to a set number of LED's and it would be dead flat output voltage, like a battery but never goes flat man! Also who needs a potentiometer when you could add a network port and assign it an IP address...

Then the whole thing would then be made in the USA and output some clean ass light. Dont get me wrong, the CREE stuff might be alright but when it comes down to Lumens and LUX stick to the BridgeLUX

bicit said:

@PICOGRAV

A list of meanwell drivers:
http://www.meanwell.com/product/led/LED.html

The one Supra was referring too.
http://www.meanwell.com/webapp/product/search.aspx?prod=hlg-185h-c

With that model you can also use resistance dimming as well as 0-10v dimming. Or in the case of SDS designs, a bit of both.

Those power supplies you have are pretty cool, but probably overkill for the typical user and huge. Meanwell switching power supplies are simpler to instal and much smaller for those with limited space.

Click to expand...

You do realize its always PWM dimming, its just the control is either through an external PWM signal or a simple pot.

There is no voltage/current regulator after the final rectifiers.


OLP is just the over load protection.

If it was PWM dimming, we would see no decrease in current droop when dimming. But I have tested that many times and there is a very strong gain in LED output efficiency as I dim the output. That could only be possible with constant current dimming.

For those who do not know, PWM dimming (pulse width modulation) flickers the light on and off at full current to achieve a dimming "effect" but because the LED is driven a high current, the led suffers the consequences of current droop, significantly reducing efficiency. So PWM is not an appropriate method to dim an LED, especially for our application where efficiency is so important. However, not to be confused with using a PWM signal to control the dimming output. That is juts a way of communicating with the driver, not the actual dimming method.

Trust me, they are PWM dimming. If you don't call mean well and talk to one of their engineers.
I am not interested in just going back and forth on such trivial and really just elementary electronic issues.
Look at the diagram and point out in what box this resistance dimming would happen. Remember it would have to come after the planar transformer and connect directly to the output to function.

The switching is only used to charge and discharge the coil where the system is designed to be overdamped. The coil resists changes to current and loses stability at lower currents (becomes underdamped and "rings"). That's why switch mode voltage supplies have a minimum current. With constant current power supplies, you don't really have to think about that, because it's designed by the supply designer around a certain current.

The detection feedback tells the controller how much the duty cycle for switching current to the coil is needed to regulate the output voltage to push a certain current through the load.


(Overdamped = stable)

t=0 is when the supply turns on. The result is a smooth DC current with minimal noise. (unless you're transmitting a signal with your leds).

Linear regulation makes a lot more sense for bench power supplies for doing circuits/electronics labs or testing, or powering precise analog instrumentation. It's definitely cleaner than switch mode regulated output, but less efficient by a lot.

If you're just going to use a constant voltage supply for cobs, you may as well go completely unregulated. (just the transformer, a diode bridge rectifier, and smoothing caps.)

@ Pico This is not a trivial issue I would like to get it sorted out. If we were using PWM dimming by flickering the LED at full current, we would be complete fools, sabotaging efficiency that we pay so much $ to get.

Unfortunately I am not educated as an electronics engineer, your knowledge is above mine so hopefully you can help us sort this out. So if we are using PWM dimming, why am I not seeing current droop when it is dimmed? In other words, what can explain the phenomenal increase in efficiency when dimmed. If it were using PWM dimming there would be no change in efficiency.

Wow CH, great explanation thank you for that ! This is the driver that I notice the very high frequency flickering, do you think it is using a different architecture, poorly designed or a non issue?

Probably hollow capacitors.

SupraSPL said:

I notice the very high frequency flickering, do you think it is using a different architecture, poorly designed or a non issue?

View attachment 3337405

Click to expand...

Here's a transient simulation of a buck converter design (found on google images)

http://www.youspice.com/ys/en/project/tps40195-synchronous-buck-converter.3sp

Its the voltage is switching, not the current.

Also well designed linear low dropout regulators can actually be more efficient than switched*

*http://cds.linear.com/docs/en/application-note/AN140fa.pdf

HM on page 4 it says: "For applications where output voltage is close to the input voltage, LDOs (low drop out linear supply) may be more efficient than an SMPS (switch mode power supply). "

But then on page 8 it says: "the overall efficiency (AC switching) is about 93% at full load, much better than that of an LR (linear regulator) or LDO supply. "

However, the authors should call Mean Well engineers because they are getting 95% with DC conduction and AC switching losses combined. (kidding/sarcasm)

The give this example regarding linear vs switch mode:
"nowadays a 12VIN, 3.3VOUT switching mode synchronous buck step-down supply can usually achieve >90% efficiency vs less than 27.5% from a linear regulator. This means a power loss or size reduction of at least eight times. "

They point out that this is especially important in high current applications.

Thats because you are going from 12 volt down to 3.3 volt, thats a big difference.

There is no AC switching in a linear power supply.

Yes of course, it is just an example they chose to give but it show how bad linear power supply efficiency can be. The big Mean Wells are able to up or down and maintain 92-94% efficiency. The smaller ones are able to do 91% in most cases. That is where we want to be for a grow lamp that is running 12-20 hours/day.