1 of my drivers is hot to touch.

diggs99

diggs99

Well-Known Member
whytewidow said:
Those drivers should hit 49.8vdc max not 45.
Click to expand...
ya thats what i thought too, i remember you telling me the A drivers will push more current and wattage than the advertised 480

why would mine only be running at 10.3a @ 45.v?

im not smart enough to figure it out, chief has tried to help me and walked me through all the tests earlier. came up with those numbers.

What do you think i should do next?
 
diggs99

diggs99

Well-Known Member
i checked both drivers

at 10.3a they were both showing 45v

both showed 11a at max pot, ill double check voltage when lights come back on at max pot. i know it never got to 48 or 49.

not saying i did it right tho lol
 
ChiefRunningPhist

ChiefRunningPhist

Well-Known Member
It's because he had the load connected, which I advised. The LEDs flowed 10.3A current, but only needed 45V to do it (it seems), if there were more LEDs in series then I'm sure you'd get the top end V out of the driver, but with the way the load is it only needs 45V to flow the 10.3A.

Did you test 11.01A at one point?

According to the test report, 10.78A was the max flowed at 115VAC input. That's only a little over the 10.3A you measured. I think the DC output measurements are within spec, at least that's what I'm comprehending, and that your Vout is only being regulated as your load is acting like 1 giant diode, which is also to be expected.
 
Last edited:
diggs99

diggs99

Well-Known Member
No i didnt test the 11.01 for voltage

i did just test the " good driver" again tho, it maxed out at 10.47a and still read 45v

I was under the impression that by wiring in parallel i would be splitting the amps between however many strips i used and the voltage would be 48v ...is that incorrect?
 
ChiefRunningPhist

ChiefRunningPhist

Well-Known Member
45V × 11.01A = 495.5W, and what's expected from a 480W driver.

At 11.01A your voltage should be a bit higher than the 10.3A voltage, but should not need much more V change to flow more current. With LEDs, a little change in voltage means a big change in current.

Take an LM301b for example, at 2.7V, it only draws 50mA. Add 6% more voltage (0.18V), and it will pull 400% more current (200mA) and draw ~425% more power! A little change in Vf (forward voltage) effects the power consumption of the LEDs immensely.

If you have 48 of these LEDs in series then you'd have an easier time dimming via voltage. This is because the effective dimming range for a LM301b in terms of Voltage is from ~2.7V - 2.88V, or a delta (change) of 0.18V.
Screenshot_2019-03-05-10-24-14~2.png
The voltage in series ccts is divided amoung the components in the series, so changing the supply voltage by say 10V in a 48 LED series cct would equate to 10V/48 or 0.23V change per component. The LM301b we said had a voltage operating range width of 0.18V. So if you were trying to determine the total PS voltage swing needed to dim the 48 series LEDs from low to high (2.7V to 2.88V) you'd multiply 0.18V × 48LEDs = 8.64V. 8.64V is only the operating range delta, the total operating range of 48 LEDs would be 2.7V × 48 = 129.6V, and 2.88V × 48 = 138.24V, so at 129.6V it'd be barely on, and at 138.24V it'd be super bright (138.24V - 129.6V = 8.64V). So you can see it would be easier to dim a long string or a row of LEDs using the Vo POT than just a single LED using the Vo POT because you'd barely have to touch the knob to go from 2.7V to 2.88V (or to increase 0.18V) for a single LED, but for 48 LEDs to go from dim to bright is a total V change of 8.64V which is a bit easier to do with a manual POT than 0.18V.

Limiting via current or the Io POT is more linear, it doesn't have eratic rates of power increase like you see with Voltage dimming and LEDs as the load. It doesn't increase power by 400% with only a tiny turn on the POT. The thing we care about is current, voltage only allows for the possibility of current to flow. You have to have voltage for current flow but current doesn't have to flow when a potential (voltage) is provided. So limiting by current is more advised imo, you're not trying to be precise about tweaking the voltage just right so that it will flow the right amount of current, with current limiting you're only giving the cct the exact current it needs and letting the driver do all the voltage calculations for you automatically.

So when the load is hooked up and you are flowing the max A, or 10.4A, your driver only needs 45V to do that, the LEDs only require 45V to flow the current, if your driver could provide more current then your V would also go up, or if you added another LED onto each strip light the V needed to flow the 10.4A would have to increase.
 
Last edited:
ChiefRunningPhist

ChiefRunningPhist

Well-Known Member
10.4A ÷ 16 strips = 650mA per strip

Because they are all in parallel they all get the same V, but it looks like they only need 45V to flow 650mA per strip. What strips did you buy? Any reds on the strips? Reds will make your total V drop lower than expected compared to all white.
 
diggs99

diggs99

Well-Known Member
ChiefRunningPhist said:
10.4A ÷ 16 strips = 650mA per strip

Because they are all in parallel they all get the same V, but it looks like they only need 45V to flow 650mA per strip. What strips did you buy? Any reds on the strips? Reds will make your total V drop lower than expected compared to all white.
Click to expand...
I bought the 90 Cri 3000k cutter j series strips and the 80cri j series 3000k

Each fixture has a mix of both
 
ChiefRunningPhist

ChiefRunningPhist

Well-Known Member
diggs99 said:
I bought the 90 Cri 3000k cutter j series strips and the 80cri j series 3000k

Each fixture has a mix of both
Click to expand...
According to the site, the strips have 112 chips each.
Screenshot_2019-06-28-18-26-08~2.png

Your driver is outputting 10.4A, at 45V. When we look at the data sheet of the chips that the strips are made from we can start to put a picture together.

Screenshot_2019-06-28-18-28-42~2.png




Now for some deduction...


What we know...
10.4A ÷ 16 strips = 650mA each strip.
45V wired in parallel means each strip gets 45V.

650mA is too much for each chip, the graph only goes to 240mA. So the strips have to further divide the 650mA into multiple smaller chunks. How many smaller chunks? Idk.. Let's find out...

650mA ÷ 3 chunks = 216mA per chunk.

112 chips divided into 3 chunks means 37.3 chips each chunk. You can't have uneven chips per string, so 3 chunks is not the right amount of chunks that the 650mA is divided into. So lets try another number.


650mA ÷ 4 chunks = 162.5mA per chunk

112 chips divided into 4 chunks means 28 chips per chunk. Ok so how many volts will 28 chips and 162.5mA need? Looking at the chart you can see that 162.5mA roughly corresponds to a Vf of ~2.99V for 1 chip, so 28 in series would be 83.72V and this is far over the 45V that you were measuring so 4 chunks isn't correct, let's try another guess.


650mA ÷ 6 chunks = 108.3mA per chunk

112chips divided into 6 chunks is 18.66 chips per chunk and we can't have uneven chips per string (or chunk) so the strips aren't dividing the current into 6 different chunks, time to try anther guess.


650mA ÷ 7 chunks = 92.8mA per chunk.

112 chips divided into 7 chunks means 16 chips per chunk. How many volts will 16 chips and 92.8mA need? Looking at the chart you can see that 92.8mA roughly corresponds to a Vf of ~2.83V for 1 chip, so 16 in series would be 45.28V and this is pretty spot on to what you're measuring. So it looks like 7 chunks probably.

^^^ This is why I'm saying that the driver is only pushing 45V. Its due to the design of the strips and the particular wiring of the setup. It means that while the load is connected your driver will only push ~45V to flow 10.4A, and may only bump up to ~45.5V to flow the max 11.01A. In order to see 48V while the load is connected, and assuming the strips are 16S7P, then it would require ~18.2A (and draw ~873W) and your driver can only do ~11A, so your max driver V is being limited by how much V the load actually needs in order to push the max driver current.


48V ÷ 16 chips per chunk = 3.0V per chip

3.0V corresponds to ~162.5mA, so 7 of these 162.5mA parallel chunks equates to one big chunk of 1,137.5mA. There are 16 strips in the light so if each strip were at 48V then each strip would flow 1,137.5mA or 1.137A and when that's added up it equates to 18.2A. DC Watts = V·A, so 18.2A × 48VDC = 873.6W, this means even if you used a 600h-48a, you still wouldn't have a big enough driver to push 48V in your setups.
 
Last edited:
diggs99

diggs99

Well-Known Member
ChiefRunningPhist said:
According to the site, the strips have 112 chips each.
View attachment 4357199

Your driver is outputting 10.4A, at 45V. When we look at the data sheet of the chips that the strips are made from we can start to put a picture together.

View attachment 4357198




Now for some deduction...


What we know...
10.4A ÷ 16 strips = 650mA each strip.
45V wired in parallel means each strip gets 45V.

650mA is too much for each chip, the graph only goes to 240mA. So the strips have to further divide the 650mA into multiple smaller chunks. How many smaller chunks? Idk.. Let's find out...

650mA ÷ 3 chunks = 216mA per chunk.

112 chips divided into 3 chunks means 37.3 chips each chunk. You can't have uneven chips per string, so 3 chunks is not the right amount of chunks that the 650mA is divided into. So lets try another number.


650mA ÷ 4 chunks = 162.5mA per chunk

112 chips divided into 4 chunks means 28 chips per chunk. Ok so how many volts will 28 chips and 162.5mA need? Looking at the chart you can see that 162.5mA roughly corresponds to a Vf of ~2.99V for 1 chip, so 28 in series would be 83.72V and this is far over the 45V that you were measuring so 4 chunks isn't correct, let's try another guess.


650mA ÷ 6 chunks = 108.3mA per chunk

112chips divided into 6 chunks is 18.66 chips per chunk and we can't have uneven chips per string (or chunk) so the strips aren't dividing the current into 6 different chunks, time to try anther guess.


650mA ÷ 7 chunks = 92.8mA per chunk.

112 chips divided into 7 chunks means 16 chips per chunk. How many volts will 16 chips and 92.8mA need? Looking at the chart you can see that 92.8mA roughly corresponds to a Vf of ~2.83V for 1 chip, so 16 in series would be 45.28V and this is pretty spot on to what you're measuring. So it looks like 7 chunks probably.

^^^ This is why I'm saying that the driver is only pushing 45V. Its due to the design of the strips and the particular wiring of the setup. It means that while the load is connected your driver will only push ~45V to flow 10.4A, and may only bump up to ~45.5V to flow the max 11.01A. In order to see 48V while the load is connected, and assuming the strips are 16S7P, then it would require ~18.2A (and draw ~873W) and your driver can only do ~11A, so your max driver V is being limited by how much V the load actually needs in order to push the max driver current.


48V ÷ 16 chips per chunk = 3.0V per chip

3.0V corresponds to ~162.5mA, so 7 of these 162.5mA parallel chunks equates to one big chunk of 1,137.5mA. There are 16 strips in the light so if each strip were at 48V then each strip would flow 1,137.5mA or 1.137A and when that's added up it equates to 18.2A. DC Watts = V·A, so 18.2A × 48VDC = 873.6W, even using a 600h-48 wouldn't be big enough driver to push 48V in your setups.
Click to expand...

Holy shit man, you just helped me understand so much in that post lol. Little things that werent very clear to me before reading it.

I have 18 strips, so that drops my ma a little...should be around 577ma per strip according the math above .

So after reading that post, im left asking myself, should i have went with different drivers for these strips? or is this just how they are meant to run? Is this going to cause my lights to run inefficient or improper?

@welight it would be cool if he could chime in, afterall , its his strips im using lol
 
ChiefRunningPhist

ChiefRunningPhist

Well-Known Member
diggs99 said:
Holy shit man, you just helped me understand so much in that post lol. Little things that werent very clear to me before reading it.

I have 18 strips, so that drops my ma a little...should be around 577ma per strip according the math above .

So after reading that post, im left asking myself, should i have went with different drivers for these strips? or is this just how they are meant to run?

@welight it would be cool if he could chime in, afterall , its his strips im using lol
Click to expand...
Dammit lol so then same type of trial and error but with 18 not 16 lol, so yep I calculate 577mA each, so 577 ÷ 7 = 82.5mA each chunk, and then looking at graph, maybe only ~2.815V at 82.5mA, and the multiplied by 16 = 45.04V which is even closer to what you're measuring. The math doesn't lie! Lol

That's the driver you want. The strips are actually pretty well designed imo. Using even a 600h still keeps individual chip current under 100mA, so they'll be pretty dam effecienct at full blast. Pretty nice little setup you've got going on. :bigjoint:
 
Last edited:
ChiefRunningPhist

ChiefRunningPhist

Well-Known Member
diggs99 said:
Ya thats def pretty close on the numbers.

I wish i understood if that was a good thing or not lol
Click to expand...
Knowing what the setup should operate at, and then measuring for what it is operating at will tell you if you theres anything to suspect. But because we're seeing a close correlation with what's supposed to happen with what's actually measured, it means that at least on the DC side things are going like you want.

If both drivers are operating on the DC side like expected but 1 driver is heating up or dissipating power, then it's probably an AC side probem. Checking your DC sides told you that the drivers were feeding your setups in the way that you expected and want, but if 1 driver gets hotter than the other it means it's accepting more power than it's giving. It seems at this point its not effecting the DC output but it may only be a matter of time before it fails worse, and then starts to effect DC side operation, but thats the worst case scenario. Idk what is failing, if there even is anything that's failing, if you just have a noisy coil, or anything, idk any of it, but checking your AC side would tell you if you're consuming more power than the other which could be grounds for a refund. Growing with LED is more effecient and why lots use LED, but if your driver is only 80% effecient then you negate the purpose of using LED, or spending high$$ on a MW driver. An AC side wattage measurement (at 10.4A @ 45 VDC output) would tell you if 1 driver were operating at a different effeciency, if this were the case I'd want a new driver. Not too concerned with a noisy coil (but should probably look more into it to have a decent opinion on the matter) but the effeciency difference would be the deal breaker for me.

EDIT:
Effeciency = (DC watts out) / (AC watts in)

You can't bump or move POT, you have to set the DC side for whatever amount of output and then measure the "AC watts in" that correlate for that particular "DC watts out". Just being thorough, I know you probably understood that, but just in case.
 
Last edited:
You must log in or register to reply here.
Top