Apogee Par meter and LED's

[Archaic37]

Apparently there is some sort of conversion when using the apogee par meter under LED's but I can not find consistent information. Some say multiply it by 5 others 15%.

 

My current lighting is the DIY Full spectrum 40B version.

[Nano sapiens]

You will find various opinions on this. A lot depends on the violet/hyper violet/blue makeup of your LED setup. Most 'full spectrum' LED systems are relatively heavy in this lower visible light range where the Apogee sensor is less sensitive, so I believe a higher percentage is is the more accurate choice.

 

I multiply my readings by 15% for my 'full spectrum' LED setup and I figure that's close enough.

[Archaic37]

You will find various opinions on this. A lot depends on the violet/hyper violet/blue makeup of your LED setup. Most 'full spectrum' LED systems are relatively heavy in this lower visible light range where the Apogee sensor is less sensitive, so I believe a higher percentage is is the more accurate choice.

 

I multiply my readings by 15% for my 'full spectrum' LED setup and I figure that's close enough.

Awesome thank you.

[Archaic37]

So I just did a quick test. The highest point on my rockwork only gave me a value of 130 on the apogee read out. so with 15% conversion thats still 150. No way that can be right.

 

I did just turn my lights on so I am not sure if LED's have to warm up to output their max potential.

 

I tested it under my par38 bulb from rapidled. At 11" of water it was getting 100. While at 4" of water with my DIY i was getting 130.

[Semisonyx]

That could be right. Not sure of the details of that bulb, but if there are no optics, it very well may be the case. Go to apogee's website and find the graph for the model you are using. There is a rectangle or ideal square over a curve, and you have to read the graph for each spectrum of diode. For HV, it can be 25% off, RB 15-20%. You have to check each color channel separately if you really want a good idea of what is going on. If you can't, then get ready to do some algebra!

[Archaic37]

I am assuming if I do each channel individually and find the % error I would then add all the channels with their %error up to give me a true par value.

[Archaic37]

That could be right. Not sure of the details of that bulb, but if there are no optics, it very well may be the case. Go to apogee's website and find the graph for the model you are using. There is a rectangle or ideal square over a curve, and you have to read the graph for each spectrum of diode. For HV, it can be 25% off, RB 15-20%. You have to check each color channel separately if you really want a good idea of what is going on. If you can't, then get ready to do some algebra!

Youre talking about these graphs? http://www.apogeeinstruments.com/light-intensity-measurements-for-light-emitting-diodes-leds/

[Nano sapiens]

I am assuming if I do each channel individually and find the % error I would then add all the channels with their %error up to give me a true par value.

 

Correct. As much as possible, I did this on my setup and ended up with 14%.

 

Between any two different lighting systems there will be differences, but this will get you in the ballpark. I use my PAR measurements for my system, mostly, which then gives me a baseline if I should make any changes.

 

Your PAR reading does seem low. What is the total distance from the top of the rock work up to the LEDs?

[Archaic37]

Correct. As much as possible, I did this on my setup and ended up with 14%.

 

Between any two different lighting systems there will be differences, but this will get you in the ballpark. I use my PAR measurements for my system, mostly, which then gives me a baseline if I should make any changes.

 

Your PAR reading does seem low. What is the total distance from the top of the rock work up to the LEDs?

13". Even putting the par meter 3" away I yield 1000 and 6" 250. This is out of the water readings.

 

Also I am not sure how to read the graphs or if I am looking at the correct graphs like Semisonyx said

[Semisonyx]

Those are the ones,and you are correct on the math . I think 17% is what I came up with on my last diy (it was HV heavy). Are there any optics on your bulb?

 

Look across the bottom and find the spectrum you want to measure, let's say 420. Follow it up to the curve and then to the left, you get about .75, depending on how big your finger is. That means the sensor only reads about 75% of the light at 420nm. Inversely, if you go over in the 550 range you get something like 1.1, meaning the sensor reads 10% higher than what you are really getting at 550nm. Make sense?

[Archaic37]

Those are the ones,and you are correct on the math . I think 17% is what I came up with on my last diy (it was HV heavy). Are there any optics on your bulb?

No optics on it. Could you give me a quick run down on reading those graphs? There are like 4 of them and I am not sure how to use them or find the % error since the values are 0-1.2 relative output.

[Archaic37]

Those are the ones,and you are correct on the math . I think 17% is what I came up with on my last diy (it was HV heavy). Are there any optics on your bulb?

 

Look across the bottom and find the spectrum you want to measure, let's say 420. Follow it up to the curve and then to the left, you get about .75, depending on how big your finger is. That means the sensor only reads about 75% of the light at 420nm. Inversely, if you go over in the 550 range you get something like 1.1, meaning the sensor reads 10% higher than what you are really getting at 550nm. Make sense?

Awesome thank you. That makes much more sense. It looks like everything past 675nm theres a 100% error.

[Semisonyx]

Right, not even sure why they would have the square go past 660 if it can't be read by the sensor.

 

So, no optics, that makes those #s sound correct. I need to find my data sheet, but I had 16 leds w no optics at 700ma, and only got About 50par at 20 inches. I'll dig around and see if that paper is on my work bench.

[Archaic37]

Right, not even sure why they would have the square go past 660 if it can't be read by the sensor.

 

So, no optics, that makes those #s sound correct. I need to find my data sheet, but I had 16 leds w no optics at 700ma, and only got About 50par at 20 inches. I'll dig around and see if that paper is on my work bench.

So what angle optics would you recommend? It is a 40B (36"L x 18" D x 16" H). The light bar is 11" off the water and 24" long with 2 clusters of 8RB, 4NW, 4True violet and 2 OCW. I can lower the light bar either by lowering the arm its hanging on or make longer y hangers.

[Semisonyx]

I have a 40br, so we are in the same boat. I had a Maxspect Razor, but have sold it to move towards a t5/led combo. Right now I'm running 2 39w T5 coral+ bulbs and 7 Cree RB@1.2A, but I have no idea about the par until I can get my hands on a meter again, as I'm considering adding a string of 6 HV.

 

#1- At what amperage are you running your strings now?

 

#2- I prefer reflectors over optics, and will probably put these on my build:

http://www.mouser.com/ProductDetail/Ledil/C12469_LISA2-R-PIN/?qs=JK0Mn6q9DKTpRqCgmdhRAQ%3d%3d

or these:

http://www.mouser.com/ProductDetail/Carclo/10170/?qs=%2fha2pyFadugCAEgwHho55vjAI%252bfL9aZAg1BdlC95eU7u8ODBrQAYxw%3d%3d

[Archaic37]

Cree RB and NW are run at 1000, true violet at 700 and OCW at 500. These are the max amperage that the leds can handle

[Semisonyx]

The Cree can go 1500mA if sufficiently cooled, but I think the reflector route would be best.

[Archaic37]

The Cree can go 1500mA if sufficiently cooled, but I think the reflector route would be best.

I am using the LDD drivers which only go up to 1000mA. I ordered some lenses from RapidLED so they should get here soon.