DiY LEDs - How to Power Them

AquariusPanta said:

Would it be correct to assume that six drivers wired in series to a single power cord, each pulling around 30V, wouldn't work here in the United States (120V standard) but that by wiring three drivers per power cord (in series, ~90V) could be accomplished?

My understanding is that 120V AC is supplied per inlet/outlet and that the components drain away the available current provided per room; essentially every outlet/inlet is ran in parallel from the main power line.

If I'm asking too many questions, please feel free to stop me .

Click to expand...

Drivers don't pull 30v on the AC side. I think your mixing AC and DC.

If you put a driver requiring 120v, in series with 2 more they would cut the voltage betweent them all and it wouldn't be enough to even operate.

But your understanding of AC in a house is good. All wires in parallels to the same breaker. What ever the breaker is...is how many amps anything(s) plugged in have the potential to draw from.

Greengenes707 said:

Drivers don't pull 30v on the AC side. I think your mixing AC and DC.

If you put a driver requiring 120v, in series with 2 more they would cut the voltage betweent them all and it wouldn't be enough to even operate.

But you understanding of AC in a house is good. All wires in parallels to the same breaker. What ever the breaker is...is how many amps anything(s) plugged in have the potential to draw from.

Click to expand...

O you're right. I didn't think of that . Thank you for clearing that up for me in an efficient and timely manner.

I'm running a fixture that is pulling around 4-5 amps because each COB has a dedicated driver. Another reason to go with the mega drivers (HLG and similar MW makes).

AquariusPanta said:

O you're right. I didn't think of that . Thank you for clearing that up for me in an efficient and timely manner.

I'm running a fixture that is pulling around 4-5 amps because each COB has a dedicated driver. Another reason to go with the mega drivers (HLG and similar MW makes).

Click to expand...

Amps x volts= wattage.

If the efficiency and power factors are the same they should be close in total amp draw for the same wattage worth.
Hlg-c's are just top performing in ever category.

I could be wrong on this one. Confirmation anyone?

Greengenes707 said:

Amps x volts= wattage.

If the efficiency and power factors are the same they should be close in total amp draw for the same wattage worth.
Hlg-c's are just top performing in ever category.

I could be wrong on this one. Confirmation anyone?

Click to expand...

Here's the scoop, dawg:

My fixture = 6x Vero 18's w/ 700mA (&& 1050mA) drivers hooked to each of them, with each driver wired to a single power cord. Using your logic, the output is not always the input. That being stated, the 1050mA drivers take .9 amps in order to operate (AC), while the 700mA drivers need .6 amps in order to run (AC). It equates to around 4.2 amps going into the drivers.

Now the MeanWell HLG 185H-C700 has a range of 143-286 VDC, plenty enough to power my mentioned COBS. The amp input for the HLG 185H-C700 is typically around 2 amps @ 115 VAC.

Correct me if I'm wrong but this was the point I was making with my previous post.

If that were the case wouldn't I be popping breakers left and right running each driver at 2 amps on top of everything else I have running? I see your logic, I wondered the same thing when I read the label on my drivers. But need clarification

Are the power factors and efficiencies the same as the hlg? If not that is where the extra amps are coming from.

AquariusPanta said:

Here's the scoop, dawg:

My fixture = 6x Vero 18's w/ 700mA (&& 1050mA) drivers hooked to each of them, with each driver wired to a single power cord. Using your logic, the output is not always the input. That being stated, the 1050mA drivers take .9 amps in order to operate (AC), while the 700mA drivers need .6 amps in order to run (AC). It equates to around 4.2 amps going into the drivers.

Now the MeanWell HLG 185H-C700 has a range of 143-286 VDC, plenty enough to power my mentioned COBS. The amp input for the HLG 185H-C700 is typically around 2 amps @ 115 VAC.

Correct me if I'm wrong but this was the point I was making with my previous post.

Click to expand...

So the 1.05A drivers are pulling .9A each from the wall? If so, that is 108VA, much higher than I would expect even if not power factor corrected.

29.57V * 1.05A = 31W LED dissipation. Assuming they are 88% efficient drivers (LPC-60-1050?), that is 35.5W total (driver + LED dissipation).

If the drivers were power factor corrected, that would be about .3A on the AC side but if power factor is the .5 that we normally see, it would draw about .6A. If power factor was .3, it would draw about .9A.

The HLG-185H-C1050 is 92-94% efficient and is power factor corrected, so yes you would draw a lot less current on the AC side. For 6 Vero18s at 1050ma, I would expect the HLG to draw 1.65A at full power.

SupraSPL said:

So the 1.05A drivers are pulling .9A each from the wall? If so, that is 108VA, much higher than I would expect even if not power factor corrected.

29.57V * 1.05A = 31W LED dissipation. Assuming they are 88% efficient drivers (LPC-60-1050?), that is 35.5W total (driver + LED dissipation).

If the drivers were power factor corrected, that would be about .3A on the AC side but if power factor is the .5 that we normally see, it would draw about .6A. If power factor was .3, it would draw about .9A.

The HLG-185H-C1050 is 92-94% efficient and is power factor corrected, so yes you would draw a lot less current on the AC side. For 6 Vero18s at 1050ma, I would expect the HLG to draw 1.65A at full power.

Click to expand...

Actually, I have no idea on what the drivers are pulling from the wall; I glanced over the tops of the drivers (mix of APC-1050 and PLD-700) and reported the numbers printed on the tops of em'. I'm not familiar with power factor and correcting mainly because I haven't found a reason to look into it but I have noticed it before. It would seem from your example(s) that if my 1050ma driver was pulling .9A from the wall, that it would have a power factor of .3, which doesn't sound very good.

Feel free to further enlighten me, as well as others, my dear Supra.

Pleaseee

Just got 4 small-ish drivers and was wondering if I each one was going to need its own individual plug.

No hooking up these drivers in parallel huh?

4 Vero 18's at 900 mA $12 Fastech driver
1 Vero 29 at 1800 mA Ebay driver

That seems like it is going to be bumping up against one breakers Amperage limit.

Could you shed some light on this Supra?

One power strip for the Vero 18's and then the Vero 29 gets its own outlet? OR split them 3/2. Idunno.

Gonna be a little wire jungle I know that much

ReeferDance said:

Pleaseee

Just got 4 small-ish drivers and was wondering if I each one was going to need its own individual plug.

No hooking up these drivers in parallel huh?

4 Vero 18's at 900 mA $12 Fastech driver
1 Vero 29 at 1800 mA Ebay driver

That seems like it is going to be bumping up against one breakers Amperage limit.

Could you shed some light on this Supra?

One power strip for the Vero 18's and then the Vero 29 gets its own outlet? OR split them 3/2. Idunno.

Gonna be a little wire jungle I know that much

Click to expand...

Not to jump Supra, but just some quick math.....


You will be using about 5.5 amps DC and pulling around 100-110 watts for the Veros and 65-70 for the 29....

So......5.5 amps DC max 180watt draw....

5.5 amps AC x 110v = 605 watts....

I would guess that 5.5 amps DC is not converted directly from 5.5 amps AC....Do you have datasheets for the drivers? Does these Ebay drivers even include them?

I would also guess that you would be pulling somewhere along the lines of 1.5 amps of AC at the mains + whatever the conversion losses are before DC.....


You can hook your AC mains in Parallel....what isn't recommended is hooking up your DC mains in parallel to individual cobs, at least not without circuitry.....

tested CXA3070AB with HLG-150H-42B....
with dimm: from 0 to 100% perfect and linear: 41.25v and max 2.68Amp= 119.8W
without dimm: 41.25 and max 2.69 Amp= 119.9W
the driver, if you see the data, can run max in tollerance 46v>3.6Amp..the A version can adjust....
but it's not possible to go up....
I think "near 42v" is the up-limit for CXA....it can be different i think from CXA to CXA....
but...it's like a sort of "protection"....or in reality the B version is limit to real max 42v...
what do you think?
tell me if I'm wrong please....

Greengenes707 said:

It automatically puts out the right voltage if it is a constant current driver and within the voltage range of the driver. The driver pushes current through the cob at a fixed/constant amount(700ma,1400ma...), and the cost of going through it is the Vf of the cob.
You could think of it as the cob pulling from a possible 42v...only taking what it needs(~36v). So it's more like the cob deciding what to take, not the driver deciding what to give.

Click to expand...

Thanks for the explanation dude, it's a subtle change which I keep overlooking.
What I also struggle with is how a LED resistance changes non linear to its current draw. I thought I understood dc resistance haha.

If I were to plug my CXA3070 into an accurate dc voltage supply @ 37V it would function no differently, is that correct?

I think what I really don't understand is the concept of a constant current device;

When we talk current, were talking about electrons in wires or devices.
point 1) The only way to make current flow in a circuit is to apply some p.d to either end of that circuit.

2) to make 700mA or 1400mA flow through our COB circuit, don't we just need to apply the appropriate voltage?

3) when the ccd is connected to the circuit, current immediately starts flowing, does the ccd just monitor the current flowing and then adjust its output voltage to reach the desired output current?

Sorry for all these questions, I am reading online aswell

zangtumtum said:

tested CXA3070AB with HLG-150H-42B....
with dimm: from 0 to 100% perfect and linear: 41.25v and max 2.68Amp= 119.8W
without dimm: 41.25 and max 2.69 Amp= 119.9W
the driver, if you see the data, can run max in tollerance 46v>3.6Amp..the A version can adjust....
but it's not possible to go up....
I think "near 42v" is the up-limit for CXA....it can be different i think from CXA to CXA....
but...it's like a sort of "protection"....or in reality the B version is limit to real max 42v...
what do you think?
tell me if I'm wrong please....
View attachment 3364305

Click to expand...

I would guess the driver is limited at 42V / 3.6A because the internal circuits will overheat if it tries to deliver more power?

@zangtumtum Thank you for the test data. Yep at 2.68A that particular CXA3070 must be pulling ~42V, the max voltage range for the driver so the current can go no higher unless you used a lower Vf COB. You would get more current and a lot more light out of that driver with a Vero29 for example.

@ReeferDance Abiqua is right on, no problem to connect those drivers in parallel on the AC side if you prefer it that way.

The 4 "900mA" drivers combined should draw about .85A on the AC side if the are pf corrected or 1.6A if they are non pfc. The sticker says they are, but I would be certain until checking it. I would suspect they are closer to 700mA than 900mA on the DC side, also worth checking.

The ebay driver should draw .65A if PFC or 1.2A if non pfc, assuming it will put out the full 1.8A on the DC side

So in summary, even if none of the drivers are power factor corrected, your total draw should be about 3.4A which is well below the max recommended 12A for constant use on a 120V circuit.

thanks @SupraSPL for your answer....I know learning from your experience and data ...exactly this power supply was initially chosen for Vero...
the next lamp i put inside 9XVero29 at max....i'm waiting the new series....
when it will be available ... I hope soon
There are warnings using this combination of CXA3070 and this driver at max power for long time?
taken off the heat off course....

As long as you can keep it reasonably cooled it should be no problem to run it that hard. About 32-33% efficient at 2.7A. The Vero29 3k would be about 35% efficient at 2.7A and 31% @ 3.6A.

UKpeanuts said:

Thanks for the explanation dude, it's a subtle change which I keep overlooking.
What I also struggle with is how a LED resistance changes non linear to its current draw. I thought I understood dc resistance haha.

If I were to plug my CXA3070 into an accurate dc voltage supply @ 37V it would function no differently, is that correct?

I think what I really don't understand is the concept of a constant current device;

When we talk current, were talking about electrons in wires or devices.
point 1) The only way to make current flow in a circuit is to apply some p.d to either end of that circuit.

2) to make 700mA or 1400mA flow through our COB circuit, don't we just need to apply the appropriate voltage?

3) when the ccd is connected to the circuit, current immediately starts flowing, does the ccd just monitor the current flowing and then adjust its output voltage to reach the desired output current?

Sorry for all these questions, I am reading online aswell

Click to expand...

Supra?
GreenGenes?
SDS?
ChurchHaze?

You have it correct, the constant current driver adjusts voltage (pressure) to keep the current (flow) steady. This is helpful because the Vf of the COB changes with temperature and there is variation from COB to COB.

I'm looking for some assistance finding low mA drivers for a veg light. Likely running vero10s(or cree cxa/b). In an ideal world pwm dimmable would be great but I dont think low wattage drivers exist(for a decent $$).
Looking at these drivers. Might configure two in parallel to give me 150/300ma running mode.
Or these, don't know if there is any difference between a led standard driver and a t8/t10 led driver(You can ask for them in 100-300ma configuration,and their efficiency seems really good).

another result, I bypass the multimeter in series for Amp....
I use only multimeter sensor temp between heatsink and the base of the cob...
tested the same COB CXA3070AB with the same HLG-150H-42B...
result after 120 minuts in air (fan 12": 70 cfm/1.2 W)
129.5 watt @ 32°C
i test the sensor it work fine....
at touch i think are 32°C real...the anodized heatsink it's cold and designed for high powers and is planarized with a surface roughness of 1,2 micron