My Low budget build - Chanzon 100w Full Spectrum / Meanwell SE-600 36v power supply

Sun May 20, 2018 4:25 pm

sdfoster22 wrote: ↑
Sat May 12, 2018 6:47 pm
Does the display say it's in cc mode? Maybe you have to change it to CV mode to adjust the voltage. This is just a guess though.

I've found that changing the voltage or current (setting one or the other) puts it into CC or CV.

Sun May 20, 2018 4:34 pm

electrok-d wrote: ↑
Sat May 19, 2018 5:59 pm

sdfoster22 wrote: ↑
Sat May 12, 2018 6:47 pm
Does the display say it's in cc mode? Maybe you have to change it to CV mode to adjust the voltage. This is just a guess though.
There is no information in the instructions on how to do that or whether or not it is even possible. It is likely that it switches between cv and cc modes internally, based on currently onknown factors.

*edited to add newly discovered info*

I think I found the reason behind the voltage regulator switching to cc mode. The power supply is cv by default and rated at 36v/15amps. The cob led units will apparently take as much amperage as given until they blow up. The regulator is limiting current by switching to cc mode in order to protect them. That's great.

What I don't quite understand is why I can't increase the voltage in this case though. I can max the output voltage on the power supply and set it really high in the regulator, but no matter what I do, it remains slightly under 30 volts.

Is the true output of the PSU variable (can you control it - like a Meanwell "A"-type driver)? The buck converter. by design, will have a 2V drop - could that what you're seeing?

sdfoster22 · Sun May 20, 2018 4:39 pm

I was reading reviews and Q & A on Amazon lol

electrok-d · Wed May 23, 2018 6:50 pm

bvolt wrote: ↑
Sun May 20, 2018 4:34 pm
Is the true output of the PSU variable (can you control it - like a Meanwell "A"-type driver)? The buck converter. by design, will have a 2V drop - could that what you're seeing?

The power supply has a pot for the voltage. I can crank it up to about 38 volts. The voltage drop that I'm getting is close to 4 volts at the regulator though. No matter how high I set the power supply voltage, the OVP, or the target voltage on the regulator, it will never go above about 29v.

electrok-d · Wed May 23, 2018 6:53 pm

sdfoster22 wrote: ↑
Sun May 20, 2018 4:39 pm
I was reading reviews and Q & A on Amazon lol

They are funny until the part where the guy reveals that this particular regulator failed and blew a huge string of his leds. This is why I chose cheap leds to experiment with. I'm not sure why he didn't at least have inline fuses. It is something I'm planning on doing in the future by heading to a junkyard and pulling a fuseblock from an old beater.

sdfoster22 · Wed May 23, 2018 8:55 pm

electrok-d wrote: ↑
Wed May 23, 2018 6:50 pm

The power supply has a pot for the voltage. I can crank it up to about 38 volts. The voltage drop that I'm getting is close to 4 volts at the regulator though. No matter how high I set the power supply voltage, the OVP, or the target voltage on the regulator, it will never go above about 29v.

Maybe that's what the cheap cob is drawing, and the specs are just that far off? Do you have another driver to test the cob on?

Wed May 30, 2018 5:26 am

sdfoster22 wrote: ↑
Wed May 23, 2018 8:55 pm

Maybe that's what the cheap cob is drawing, and the specs are just that far off? Do you have another driver to test the cob on?

A common example of inferior binning/manufacturing-period. That gets my vote.

electrok-d · Fri Jun 01, 2018 6:14 pm

Aye. This could indeed be an example of getting what you pay for. I'm ok with that, in this case. The system is working and it has my foot in the door for indoor gardening at a really low cost. I'm getting over 88 watts per cob and was able to jump start my outdoor garden from seed. I was also gifted a huge 6 ft plant, and after repotting and about a week under the 4 cobs, it is seeing new growth and appears to be really happy.

Ultimately, what this tells me is that this type of system can indeed circumvent the typical convention of using a standard led driver. Before buying more expensive leds, I would definitely factor in a few fail safes though. There are more pieces to the puzzle, which in turn opens up the door for more things to go wrong. If I ever add more expensive leds to this system they will be fused and probably have some level of resistance wired in to ensure protection against component failure.

Thank you all for participating in this thread. I'll add to it if I discover anything new. I love electronics and I am discovering a new love for gardening. This site is the best!

Wed Jun 06, 2018 7:40 pm

A great experiment/demonstration of how those w/o access to premium raw materials can still get the benefits of growing with LEDs.

Thanks for the info and discussion.

DC-10 · Sun Jul 15, 2018 5:20 am

Really interesting experiment here. It's cool how much power you can get out of an inexpensive supply. I will be interested to see how much light you can get out of it once you switch to some high-efficiency brand-name COBs, such as Bridgelux or CREE units. My experience with the cheap COBs from Amazon is that the luminous efficacy tends to be very low, once calculated at less than half my Bridgelux ones of the same power. You may find that, when you switch to higher efficiency COBs, you will need less fan speed to do the same cooling, as more energy is emitted as light rather than sent to the heatsink as heat.

You mentioned additional protections needed for more expensive COBs. The primary one I usually use is a temperature sensor connected directly to each heatsink and controlling a relay on the AC power input directly. That way, if for any reason the COBs get too hot (overcurrent, broken fan, blocked airflow, etc), they will be protected. Remember to add some hysteresis in so that they won't just flicker on and off if it trips. I see that the DROK converter already has overcurrent protection built in, so that's a nice feature as well.

If you were interested in testing your Chanzon LED's luminous efficacy, even without a quantum meter, it could be done. Attach the LED and a calibrated temperature sensor to a block of aluminum of known mass. Turn on the LED at known power and run for a set time, record the temp before and after of the block. Using the specific heat of aluminum and time, you can calculate watt-hours of heat absorbed, and then divide by watt-hours of electrical power consumed and you have a crude measure of efficiency! Some of the highest-end blue LEDs can be up to ~70% efficient. Even the best COBs are quite a bit lower due to the phosphor conversion, but I think I read that they are still in the range of 20~30%.