Ceramic Metal Halide (CMH)

[Beta Test Team]

you posted this article
http://cpl.usu.edu/files/publications/poster/pub__3801011.pdf
View ImageView Image
View ImagePic drag and drop seems to not work. http://prntscr.com/5m49sn

unless I misunderstand something, thoes super power efficient red photons suck. No one wants the red vs the blue light fed plant above.

So exploiting the physics of super cheap red photons and arguing it is MORE EFFICIENT to mostly only provide red light is kinda bullshit, right? No one looking at that pic above is not willing to pay 35% more power for the blue plants vs the red plants.

No one (that I'm aware of) is using only red radiation to grow Cannabis, or really any other plants. I think maybe you got some things turned around about what I wrote regarding photosynthetic efficiency (and this is pretty far off topic so I'll be brief). That study you're citing isn't a great resource for this discussion (I only posted about it in regards to intracanopy and intercanopy reflection of green photons):

I don't know what you mean by "super cheap red photons," because when I write about of efficiency it's about turning joule/s into unweighted photons within 400-700 nm.

In the example I mentioned, of Gavita, the spectrum isn't terrible, but it's not great, either. The point about photosynthetic efficiency I made was only to show the range of efficiencies available, to give context to Greenbeams and other CMH luminaire photosynthetic efficiencies.

Photosynthetic efficiency (umol/s within PAR range per joule/s) is very important to running a business, or even just a home grow for the anal retentive people. It's an important metric used to compare and understand luminaires, lamps, and reflectors (like the reflector 'photosynthetic radiation efficiency' metric I came up with yesterday).

I have this question I am not sure I have the language to ask. Given the commonly understood par curve, and that light at the red end is way cheaper to make than a more balanced light SPD, how can we rate or quantify how well a given light source covers the "area under the PAR curve". And apply THAT efficiency of PAR coverage to the PPF efficiency?

That's not really the best way to do things, it's better to use PPF (unweighted photons), than YPF (yield photon flux), which is PPF adjusted (weighted) by how plants use PAR range photons for photosynthesis (what you're asking for). So it's better to use the photosynthetic efficiency I wrote about above, rather than what you're asking for. However, that said, using both is even better, as long as the former (PPF) is used as the primary data set.

This is too off topic for this thread for me to want to post a lot, so just very briefly: there are important limitations to doing what you're asking (using weighted photons), so it's more accurate to use PPF rather than YPF, when defining efficiencies and comparing lamps, when thinking about photosynthesis.

So the answer to your question is to find YPF efficiency (umol/J in PAR range weighted with how plants use photons for photosynthesis) you need to multiply lamp's SPD and the relative quantum efficiency (RQE) of photosynthesis by wavelength, then integrate all solutions and you get YPF, then finally divide by input joule/s (watt). (Our spreadsheet does this in an relative automated fashion.)

Because I think it can/should be argued that the 2.1 umol of the DE 400V bulb (that is mostly only red) should be compared to the far better SPD of the 1.9 umol of the greenpower. And i suggest that IF that metric is well understood, then a thing we have noted; that perhaps the 942 bulb is perhaps as good as the agro. Not because it has the umol output, but because its lower umol better fits the PAR curve.

Does any sort of metric like that exist?

Yes, though like I wrote above, it's not preferred vs. unweighted photons for reasons I'd rather not explain here (they're off topic). All the graphs in the following thread represent the exact thing you're asking for, the data used to make those graphs are various RQEs, that when multiplied by relative or real spectral energy provide the weighted irradiance:
https://www.icmag.com/ic/showthread.php?p=6718314

And even if we used the RQE of photosynthesis to convert PPF to YPF (either by area, or not), the Gavita would still be greater due to the greater effect on photosynthetic rate from red range photons than blue and green.

To get a real understanding of spectrum quality, we need to adjust the PPF by many RQEs, such as that of photosynthesis, blue-light response, stomatal opening, stomatal closing, etc. (See the thread link for those curves.)

 

[LargePrime]

As i have a great deal more questions, do you have a thread in which i can bother you more about this?

 

[Beta Test Team]

Here you go, for that exact subject:

"Action spectra and RQEs weighted with Cannabis absorptance"
https://www.icmag.com/ic/showthread.php?t=295933

 

[Fishywater]

OK, I am considering getting the ATL 220$ kit.
I emailed them to see how much they would sell me a kit for without the step up trans as I have 240V in my room already.
Then I got to thinking, HOW will I put a timer on a ballast wired to 240 V ???

 

[rives]

Either use a 240v timer like the Intermatic T-104 or use a double-pole power relay (see link) triggered via a conventional digital timer.

Timer -
http://www.amazon.com/Intermatic-T1...d=1419902209&sr=1-1&keywords=intermatic+t-104

Power Relay -
http://www.automationdirect.com/adc/Shopping/Catalog/Relays_-z-_Timers/Electro-Mechanical_Relays/Power_Relays,_Open-Style,_40A_%28AD-PR40_Series%29/AD-PR40-2C-120A

 

[Fishywater]

Either use a 240v timer like the Intermatic T-104 or use a double-pole power relay (see link) triggered via a conventional digital timer.

Timer -
http://www.amazon.com/Intermatic-T1...d=1419902209&sr=1-1&keywords=intermatic+t-104

Power Relay -
http://www.automationdirect.com/adc/Shopping/Catalog/Relays_-z-_Timers/Electro-Mechanical_Relays/Power_Relays,_Open-Style,_40A_%28AD-PR40_Series%29/AD-PR40-2C-120A

Thanks man !
As soon as I read those I was like DUH !
Thanks for jogging the memory... sucks getting old...

 

[LargePrime]

hardware store has electric water heater timers that work too.

 

[Hottish]

Or a Smart-relay

Or a Smart-relay

Or a Smart-relay, add a temp sensor & turn on fan or lights off to control heat run pumps, CO2 ..........

http://www.ebay.com/itm/251773155665

http://www.ebay.com/itm/271718602780

 

[rives]

Or a Smart-relay, add a temp sensor & turn on fan or lights off to control heat run pumps, CO2 ..........

http://www.ebay.com/itm/251773155665

http://www.ebay.com/itm/271718602780

A smart relay is still going to need a power relay to switch the high current required for the ballast.

 

[Hottish]

A smart relay is still going to need a power relay to switch the high current required for the ballast.

Rives they both have 4 x 10A contact relays

The 1st link the older FL1D Operating Voltage 100-240V AC, 100-240V DC, no need for 24V power supply or extra relays for Fishy`s needs & it might get him on the road to futher automation

 

[rives]

All amps are not created equal.

The Idec specs don't break the allowable load down by capacitive, resistive, or inductive, but they do state a surge current of 30a is allowable. A capacitive load like an electronic ballast is the most difficult load for contacts to switch. They may work fine, but bear in mind that they are non-replaceable so that if you weld the contacts, the controller is bricked, and they are also not double-pole relays, which are required for a 240v load like the Philips ballast.

Smart relays and micro PC's are a great way to automate a grow, or damn near anything that can be handled within the scope of their I/O. There are some other expenses involved though, like the interface cable, the software for programming the device (although many companies are doing free downloads for their proprietary software), the enclosure, terminal blocks, etc.

This is the Allen-Bradley unit that I use. Each output drives one of the plug-in relays in the lower middle area, and then the ballasts are driven by 40a power relays in a separate enclosure -

 

[Jhhnn]

Rives they both have 4 x 10A contact relays

The 1st link the older FL1D Operating Voltage 100-240V AC, 100-240V DC, no need for 24V power supply or extra relays for Fishy`s needs & it might get him on the road to futher automation

Not to speak for rives, but it's important to look beyond resistive load rating to the inductive load rating which is expressed as HP. Both magnetic (inductive) & digital (capacitive) ballasts have huge inrush currents much like electric motors.

1 HP is 746W, 3.1A @ 240v. Lots of 10A 240v devices are de-rated to 1/4 or 1/2 HP for inductive loads, which is only 187W or 373W respectively.

Lots of growers don't understand that, end up overloading their timers w/ resultant problems over time. If that smart relay is rated for 1/2 HP at 240v, it'll safely run 1 (uno) 315W ballast from 2 of the outputs. Which is not to disparage using it- IDEC makes very good hardware IME, but it's important to RTFM.

Safety first, guys. Every time I read about a grower fire, it makes me wince. Hell, even grow shop guys don't really understand it- they'll send you out the door w/ a 1000w light & a 15A timer (rated at 1/4 HP) never knowing they aren't getting it right.

 

[Hottish]

All amps are not created equal.

All amps are not created equal.

Rives & Jhnn, thank you both for putting me straight.

[FONT=Arial, Helvetica, sans-serif] I had assumed that all amps were equal & I have an FL1D 100-240V on the way, intended to simply wire in 4 x 315w directly to its outputs 240V and use as a timer . Time to research & order [/FONT][FONT=Arial, Helvetica, sans-serif]double-pole relays. I have an enclosure, circuit breakers & 35mm DIN rail

T[/FONT][FONT=Arial, Helvetica, sans-serif][FONT=Arial, Helvetica, sans-serif]here was a local grow fire recently got me thinking about the cheap timers I am plugged into atm ............

This is becoming a project in itself now that might be wordy of its own thread and to keep this one one topic ,

H

[/FONT]
[/FONT]

 

[Jhhnn]

Rives & Jhnn, thank you both for putting me straight.

[FONT=Arial, Helvetica, sans-serif] I had assumed that all amps were equal & I have an FL1D 100-240V on the way, intended to simply wire in 4 x 315w directly to its outputs 240V and use as a timer . Time to research & order [/FONT][FONT=Arial, Helvetica, sans-serif]double-pole relays. I have an enclosure, circuit breakers & 35mm DIN rail

T[/FONT][FONT=Arial, Helvetica, sans-serif][FONT=Arial, Helvetica, sans-serif]here was a local grow fire recently got me thinking about the cheap timers I am plugged into atm ............

This is becoming a project in itself now that might be wordy of its own thread and to keep this one one topic ,

H

[/FONT]
[/FONT]

If your 4 x315w lamps are running on the same cycle, the relay rives linked above will def handle the load. It's very conservatively rated at 2 HP 240v, 1492W inductive. That's for 50,000 cycles, once a day for 137 years. Old school bulletproof. I use one myself. Check my albums for pics.

If you're going to run different cycles from the same controller, you just need more relays.

 

[Lightweight]

Howzit? Icmag CMH experts. I'm looking into getting a CMH light solely for vegging. Is there a bulb that would be better than others for this purpose? Thanks very much for any info in the right direction.

 

[dudex]

Sorry had to weigh in on Satyr and Betas argument. I think Satyr just needs an example and Beta, thank you for all your knowledge, but sometimes you gotta dumb it down a little. Too many big words for the stoners.

Based on what Beta stated(If I remember correctly, maybe not but heres the idea), the GreenBeams reflector is 95% efficient at reflecting light out of the reflector and reflects 90% of the light from the lamp. Therefore, 10% direct hit, 95% of 90% is 85.5%, for a total of 95.5% of emmitted light leaving the reflector. Thats a loss of only 4.5%, pretty damn good.

Satyr- you were arguing for a bare bulb setup, and yes that would be 100% vs. 95.5% but many times a vertical bare bulb setup is not possible and I will make the argument that in many cases still less efficient than perfect uniformity at the optimal light density.

Lets say the most efficient amount of light is 700umol/m^2. You now have a room with greenbeam reflectors that has 700umols/m^2. To achieve this you actually have to create 731.5umols/m^2 because of the reflector loss. But this room will be growing at the highest plant growth rate per umol. Thats the important part.

Now you have a vertical bare bulb with no light loss. All of it hits your plants. But because of the configuration of your bulb and where the plants are around it 25% of the canopy is lit at 400umols/m^2, 25% is lit at 1000umols/m^2, and the rest is perfectly lit with 700umols/m^2.

For the sake of argument I will make up efficiency numbers, and who knows where the exact sweet spot is but it will illustrate the point.

At 400umols/m^2 the plant is 75% efficient at using the light so 300umols/m^2 is used.

At 1000umols/m^2 the plant is also 75% efficient, so 750umols/m^2 is used.

So the grand total of umols/m^2 is then

.25*300 = 75 umols
.5*700 = 350 umols
.25*750 = 187 umols
total = 612 umols
Note-I did all the math in my head so might not be exact but very close

So with the reflector we have 700/730 = 96% overall efficiency
Vertical bulb 612/700 = 88% overall efficiency

Obviously this is highly dependent on how efficient the plant is at converting the light and the actual light density but it illustrates the point

As the light density becomes less uniform efficiency decreases.

If you want to take it all the way to the extreme, then one square inch gets 100,000 umols and the rest get nothing and your plants are dead.

Of course the best scenario is bare bulb with perfect uniformity at the optimal light density. I have always been a believer in keeping your light further away from the canopy as long as you have reflective walls and the entire footprint is covered in plants. This method attempts to increase uniformity and eliminate hot spots regardless of reflector used. I have always used daystar ac hoods because of the uniform footprint and after eight years of use have never once had to clean the reflective aluminum. Nothing sticks to it, but I do clean the glass once a month.

Also heres a sweet chart I came across. Each line represents temperature in celsius.

 

[Beta Test Team]

Interesting way to look at the data.

What I and Satyr were discussing, and I think is kind of off topic to this thread, was reflector radiation efficiency, specifically I was focusing on the radiation within PAR range (400-700 nm) because that's the data I have. Satyr was claiming Greenbeams are inefficient when it comes to reflector radiation efficiency (due to its vertical lamp and reflector geometry), as compared to something like Gavita DE or Epapillion.

Here are the calculations without the math, from the posts of mine you referred to above. And as is easy to see there's not much difference between the real-world efficiencies of Greenbeams and Gavita PRO DE, yet, Greenbeams is vastly better than Gavita in terms of irradiance uniformity over the canopy:

Greenbeams with Philips Green Power CMH 315W

-- Greenbeams reflective material reflectivity, according to Cycloptics: 95%

-- Relative amount photons emitted by the lamp that exit the reflector after a single bounce, according to Cycloptics: 95%

-- Relative amount of photons emitted by the lamp that exit the reflector, according to Cycloptics: 92.3%

-- Relative amount of photosynthetic photons emitted by the lamp (400-700nm) that exit the reflector: 76.84%

-- Reduction in photosynthetic photons (400-700nm) exiting the reflector as compared to emitted by the lamp: 23.16%


Gavita PRO DE HPS 1000W

-- Gavita 'HortiStar' reflective material reflectivity, according to Gavita: 96%

-- Relative amount photons emitted by the lamp that exit the reflector after a single bounce: unknown

-- Relative amount of photons emitted by the lamp that exit the reflector: unknown

-- Relative amount of photosynthetic photons emitted by the lamp (400-700nm) that exit the reflector: 80.95%

-- Reduction in photosynthetic photons (400-700nm) exiting the reflector as compared to emitted by the lamp: 19%

 

[Luther Burbank]

I've joined the 860w CMD club. Bulb came with a few glass shards rattling around inside the outer sleeve as well as what looks like green patina'd corossion where the glass is annealed to the socket. Wanted the expert opinion before I went plugging it in. I'll get photos later.

 

[Jhhnn]

I think that the DE & the CMH lamps spring from different heritages. DE lamps were developed as supplemental greenhouse lighting. The sun still does most of the work, so uniformity of irradiance isn't a terribly big issue. They have no reflective surfaces around them to exploit. It starts from brute force & is refined from there. Installation & maintenance are simplified- fewer of everything. They simply exploit the high ceilings & normal venting of greenhouses for cooling.

CMH has to come at it from the other direction, simply because it doesn't have brute force to begin with. Technical limitations of the materials prevent it. They were also developed as lighting for people rather than plants where high cri is very desirable. Open work spaces & public areas like supermarkets, parking lots & garages at moderate intensities. Their usefulness for horticulture was an afterthought. Cycloptics saw it & have developed reflectors to exploit their qualities in growth chambers w/ reflective walls to achieve very uniform irradiance & virtually no shade because of multiple sources & diffusion. They also minimize photon loss in the reflector. Very low emissions in the infrared combined w/ lower temps at each lamp combine to allow closer canopies, increasing effectiveness. It's a different approach, necessarily so. It takes more finesse & it's also more expensive to install & probably to maintain.

Each is very good for their designed purpose. In a large grow, like a warehouse, the DE's are an attractive option. In smaller grows, particularly wrt scientific study, CMH has its own advantages. It's most like natural sunlight & the strong blue component gives short internodes & bushy growth, particularly in combination with diffuse light & close canopy placement.

Meh. I really want to use some 315's, but I'm too cheap to pay the freight on new stuff. I had a line on some used supposedly 315w highbay fixtures thru Craigslist, but the guy never called me & the ad ran for over a month. CL clearly has more than its share of fakes & flakes, that's for sure.

 

[Beta Test Team]

Great post, Jhhnn, and I agree totally. Though a huge warehouse full of Greenbeams sure sounds nice (however, the million dollar plus price tag for them not so much).

Personally I'm really interested in those 1000W MH DE from SolisTek, the 6K model looks interesting if only in terms of spectrum, and the uniformity can be worked with in a good reflector (more units, dimmed to a lower output for greater uniformity but no loss in PPF at canopy).