1. We are currently experiencing issues with viewing and uploading images, our team is working on the issue.
    Dismiss Notice

Bridgelux Gen2 BXEB-L0560Z-30E2000-C-B3 First Look

Discussion in 'LED and other Lighting' started by GrowLightResearch, Jan 30, 2018.

  1.  
    GrowLightResearch

    GrowLightResearch Well-Known Member

    This past weekend I threw together a test fixture to test the newish Bridgelux Gen 2 strips.

    I collected data on currents, temperature, and PPFD.

    6 Bridgelux BXEB-L0560Z-30E2000-C-B3 strips. $7.36 ea.
    Driver: HLG-185H-54A, Constant current ≈$50

    3 pair of strips with two strips wired in series
    Vf @ 500mA (Sets Measured Individually)
    37.94v
    37.86v
    37.77v

    The 3 pairs were then wired in parallel

    No heatsink needed even at max current
    Temperature at 1 Amp ≈ 42° C
    Parallel Vf = 39.0v @ ≈1 Amp
    Parallel Vf = 39.6v @ ≈1.4 Amp (max)


    The currents in all three sets of strips were periodically measured.

    Currents in the strips were unbalanced due to the parallel wiring.
    Each pair had a 0.010Ω 1% resistor wired in series to it's current supply
    The resistors voltage drop were measured and converted to current.

    The yellow current measurement were supposed to be around an amp. But instead varied 15-28% becasue of the unbalanced loads.

    Yellow 965 mA to 1245 mA, Vf= 36.06v
    Blue 12238 mA to 1523 mA, Vf= 36.60v

    ebGen2CurrentsSixStrips.jpg


    =============================================================================

    The test floor of a Gorilla LIte tent: 30" x 24".
    Fixture: 22" x 24"

    I will be adding a seventh strip when I get a new driver to run all the strips in series.
    The unbalanced strip currents did not cause major problems with uniformity.
    The unbalance load will put stress on some of the strips.

    The forward voltage of these strips are rather good.
    The Vf of each individual strips @500mA are:18.55v, 18.88v, 18.89v, 18.89v, 18.92v, 18.94v.
    I was surprised to find nearly 30% differences in strip current.
    The thing is they all look good, the PPFD measurements are good. At distances less than 36" the measured irradiance had greater fluctuations.



    measurementSetupPPFD.jpg

    =====================================================================

    PPFD was measured at 36” with a US StellarNet Blue Wave Spectrometer.


    The current was adjusted to get the values around 100µmol/m²/s
    The actual measured values were normalized to 100 in order to compare them to the simulated values which were also normalized to 100 max.

    The yellow strips show the relative location of the strips to the sensor measuring the irradiance. The sensor was moved inch by inch along the line shown above and below (simulator screen shot).




    ebGen2PPFD-6Strips.jpg

    ========================================================================

    PPFD was simulated at 36”

    This is an app I have been working on since August 2016. It simulates the uniformity of grow lighting. Here we can see it works fairly well. The simulated values nearly match the measured.

    heatmapPPFD36inchHeight22x20.jpg


    measurementSetupStrips.jpg



    measurementSetupPPFDabove.jpg
     
    Last edited: Jan 30, 2018
    DankaDank and Randomblame like this.
  2.  
    wietefras

    wietefras Well-Known Member

    Those calculated values look like nothing you would expect from 6 strips. But of course if you only compare the bit where the values are all pretty much the same anyway and then "normalize" to the same value they will look vaguely similar yes. Even then you are 10% off on the edges where the values do actually change a bit (89 and 90 instead of 82) .

    Clearly your measurements were done with the fixture not hanging level either. It's all just so hopeless.
     
    linderstein likes this.
  3.  
    GrowLightResearch

    GrowLightResearch Well-Known Member

    The fixture was perfectly level. See the level in the picture?

    Did you see the current was unbalanced?
    Currents in the three sets: 1105mA, 1047mA, and 899mA shown on the right side of the measured values.
    Could that make the results look like the fixture was not level?

    What would you expect?
    Are you referring to the 111 µMoles?

    Maybe you missed this?
     
    Last edited: Jan 30, 2018
  4.  
    wietefras

    wietefras Well-Known Member

    Well whatever is the reason, it's a mess.

    [/QUOTE]What would you expect? Are you referring to the 111 µMoles?[/QUOTE]You don't get that shitty uniformity with actual strips. But then you don't understand reflection and at 36" you would have lost most of the light anyway. Who knows what more went wrong there.
     
  5.  
    GrowLightResearch

    GrowLightResearch Well-Known Member

    Is it maybe your reading comprehension?

    There is a disparity of 200mA from one end to the other. Wired in parallel. You guys keep telling me it's okay to wire in parallel.
    Now it's a mess?
    I think the uniformity is good considering the unbalanced current.
    Much better than this:

    Untitled2.jpg Untitled.jpg
     
  6.  
    OLD MOTHER SATIVA

    OLD MOTHER SATIVA Well-Known Member

    "There is a disparity of 200mA from one end to the other. Wired in parallel"

    and if you had a different driver and wired series it would be better?

    ...please show..

    you do seem to be working hard..wtg
     
  7.  
    1212ham

    1212ham Well-Known Member

    Constant current with that driver and parallel wiring? :roll:
     
  8.  
    nfhiggs

    nfhiggs Well-Known Member

    Move the pair with the lowest current to the center. The center is where you have the most "overlap" of multiple light sources. That would tend to counteract the bright spot in the center and make the light more uniform overall.
     
  9.  
    GrowLightResearch

    GrowLightResearch Well-Known Member

    I believe so. I have an HLG-240H-C1400 on order. 89 ~ 179V and 1400mA max. That can power up to 8 or 9 of these strips in series.
    The current HLG-185H-54A gets too hot to hold when running these 6 strips at max current.
     
  10.  
    GrowLightResearch

    GrowLightResearch Well-Known Member

    What alternative would you suggest?

    I posted my ideas here: http://www.rollitup.org/t/pwm-dimmer-circuit-and-driver-open-source-development.958069/
    I had that drive on hand and I wanted to see if it would be unbalanced. I have heard many times on this site that running CoBs and strips in parallel worked fine.
    There is no visual difference. By sight all strips appear to be working fine. And even with the imbalance the uniformity is not that bad.
    I will be running them all in series soon.
    I considered LDD-H but they only have PWM dimming I have a circuit board laid out for a PWM generator and USB/Serial port.
    If someone has a link to a fixture with better uniformity, I would like to see.
    For example the QB PAR map I posted the normalized numbers would look like this:

    My numbers are for 2 ft. not 2.5
    Left to Right, Center
    310 => 35
    460 => 52
    800 => 91
    880 =>100
    750 => 85
    410 => 47
    273 => 31

    Untitled2.jpg
     
    Last edited: Jan 31, 2018
  11.  
    GrowLightResearch

    GrowLightResearch Well-Known Member

  12.  
    GrowLightResearch

    GrowLightResearch Well-Known Member

    That's debatable. I think the uniformity is very good. It could be better.
    I have written a uniformity simulator. The simulator screen shot I posted was from that simulator.

    You enter the number of LEDs on the strip, the distance between the LEDs, the spacing between strip, the size of the grow area, and the range of height to simulate. It then makes a normalized PAR map at each height and uses the numbers from each square inch and calculates the range (max-min ÷ average) and standard deviation at each height.
     
  13.  
    GrowLightResearch

    GrowLightResearch Well-Known Member

    I ran the simulator from 36" down to 2" and the problem becomes obvious. This will be a 7 strip when I get the new driver. And I think I need to move the strips so the space between them is the same. Not to spread them in the middle.

    That's why I like the simulator. It took a long time. But looking to be worth it.


    uniformity36-2.jpg
     
    Last edited: Jan 31, 2018
    Randomblame likes this.
  14.  
    wietefras

    wietefras Well-Known Member

    Worth it for what? I programmed a simulator in a few hours, with reflection. It's pointless modelling grow lights without correcting for reflection. I ran a few models at different heights to confirm my measurements and was done. Never needed it again really.

    I simply confirmed that the optimum height for led strip fixture relates to the distance between the strips. Similar to the way you determine optimum height for COBs .Albeit that strips can be slightly closer in relation to the distance between them.

    Your simulated images look very much off too. Clearly not incorporating reflection is messing things up a lot. You really should see full uniformity from 8". In reality, increasing distance from there won't improve uniformity. Which is completely different from what your simulated images suggest. Actually even from 4" it should pretty much uniform. Although your tent is too large for the fixture so it's a poor match there too.
     
    GrowLightResearch likes this.
  15.  
    1212ham

    1212ham Well-Known Member

    You said "Driver: HLG-185H-54A, Constant current ≈$50"
    How do you get constant current with that driver and parallel wiring?
     
    Chip Green likes this.
  16.  
    GrowLightResearch

    GrowLightResearch Well-Known Member

    As long as the Vf is between 27 ~ 54V, there will be constant current.

    Theoretically if I set the HLG for 3 Amp each set of strips would get 1 Amp each. But they will never run balanced.

    As long as the forward voltage is with in the HLG's CC range the current will be constant. The current will be divided by the strips based upon their ESR. I found the ESR to range between 0.025Ω and 0.040Ω. The figures are in my table in the ESR column. The ESR is calculated by dividing Vf by current ESR = Vf÷Amps. I added 0.01Ω resistors for each strip so I can read the current with a volt meter.
     
    Last edited: Jan 31, 2018
  17.  
    JavaCo

    JavaCo Well-Known Member

    Get a HLG-185H-C1400 and wire in series and see what the test results are.
     
    GrowLightResearch likes this.
  18.  
    GrowLightResearch

    GrowLightResearch Well-Known Member

    That is almost the driver I will be trying next. Because I will be adding a seventh strip I have to go to the 240H. The 185H cuts off a 143v. Too close for comfort.

    Although when they are run in series the voltages will drop. In parallel the strip with the highest voltage will rule. Also, the one with the highest voltage will draw more current than the other two, raising it's Vf and consequently raising the others as well.
     
    Last edited: Jan 31, 2018
  19.  
    Randomblame

    Randomblame Well-Known Member


    I also like such tools, the more the better...!
    Even if it does not include the reflections ... it is good for open space calculations. Any plans to make it accessible to all, as a web app maybe?
     
    Moflow likes this.
  20.  
    GrowLightResearch

    GrowLightResearch Well-Known Member

    Absolutely. It's not ready fro that yet. I go into the source code to enter the parameters for now.

    I do have one you may like that is ready for prime time. It's a Height vs. Penetration app.

    It's used to adjust the PPFD and height to an optimal level. For example a 1000W HPS is usually about 40" over the canopy. That's why it gets such good penetration. When the plants are small you don't need penetration. Depending on the depth of the canopy and the desired PPFD there is an optimal height. http://www.growlightresearch.com/ppfd/depth.php
    I also have an Inverse Square Calculator.
    http://www.growlightresearch.com/ppfd/calcISL.html

    Pay no mind to the reflections. At LED heights and the directional nature of LEDs, the reflections are negligible. At high heights and with an isotropic light source then we can talk about reflections. If someone wrote a reflections calculator that worked, they deserve a Nobel Prize. It could not be done in a matter of hours if at all.
     
    Last edited: Jan 31, 2018
    DankaDank, Moflow and Randomblame like this.

Share This Page