LED Help Please!!!

[rives]

The 860's are designed as a retrofit lamp for 1000w metal halide magnetic ballasts, and are vertical-mount only. I would think that they are probably also compatible with a 1000w hps mag ballast with the ignitor removed, but I really haven't looked into them that closely. There is at least one company building an electronic ballast for them, but the CDM lamps are all low-frequency only lamps, so they will not work on standard electronic ballasts.

Your point about the excessive PAR levels of some LED fixtures is right on the nose. That is what I was alluding to in the post above when I mentioned using multiple low-wattage fixtures spread evenly over the top of the canopy. What I, and several others, found a couple of years ago was that the trend toward higher-wattage LED fixtures with the same footprint was counterproductive. For instance, the Lumigrow ES330 has the power to bleach the shit out of the plants immediately below it, but fails to reach the edges of a 30x30 tent. Spreading (4) of the ES165s over the top of the canopy would be way more preferable than running (2) ES330s over the same area. So what did Lumigrow do? They obsoleted the ES165 and kicked the power up on the ES330 in the same size envelope.......

 

[Smoking Lamp]

The sun is about 2200 umol's @ the surface of a plant, the AT is not overkill, maybe @ 3-6" inches.

One thing I wonder but has stopped me from the CMH is how efficienct they are? I mean an HPS is 31% efficient overall, so that remark has me a bit perplexed. The last thing I would call an HPS is efficient, great light source don't get me wrong.....power factor correcting and ballasts will always be it limitation or is that not correct?

 

[rives]

The sun is about 2200 umol's @ the surface of a plant, the AT is not overkill, maybe @ 3-6" inches.

One thing I wonder but has stopped me from the CMH is how efficienct they are? I mean an HPS is 31% efficient overall, so that remark has me a bit perplexed. The last thing I would call an HPS is efficient, great light source don't get me wrong.....power factor correcting and ballasts will always be it limitation or is that not correct?

I think that one of our problems in evaluating lighting is that the tools currently available to us really only work when comparing light sources from within the same family. The figures that I've seen on a Lumigrow ES330 is that it will put out 26.5 μmoles/m2/s of blue and 169.1 μmoles/m2/s of red. Without knowing offhand what the output of the sun and the Lumi are at a given distance, I think that it is safe to say the ES330 puts out a small fraction of the light that the sun does. However, they will destroy a healthy, mature plant in pretty short order if run at full power inside of a foot or so from the plant, and I've never seen the sun do that.

The same problem with metrics carries over to things like "efficiency". Most middle-of-road LEDs chips themselves are roughly equivalent to HID performance. There are some worse, and some substantially better, but they are typically within a few percent. However, it seems that nobody ever addresses the rest of the circuit losses with LEDs. Many of the power supplies available have dismal efficiency, and these losses stack. Add in cooling fans and the losses go up even more.

At any rate, the 315w CDMs put out between 105-120 lumens per watt depending on the lamp color temperature and the ballast used. But again, trying to equate the two light sources in lumens doesn't work, and PAR values aren't a great deal better. I think that any measurement that you pick, the ES330 is going to look superior on paper to the 315w CDM. However, my real world results tell me that there is something else going on because the 315 easily outperforms the 330 by a substantial margin in the same tent, with the same strain, and the same nutrients.

I don't understand your last question - how are "power factor correcting and ballasts" limitations for an HPS? In an industrial setting, HID fixtures help the overall power factor by partially correcting the power factor disruption (in the opposite direction) that is caused by running induction motors. In a residential setting, there is no penalty for a poor power factor - it takes special metering to measure it.

 

[FLAgreenthumb]

i haven't had any first hand experience with any of them. but your lead in comment confused me. you said you favored Lush but wanted to spend as little as possible. Lush has some pretty expensive lights!

lol... your right they are pricey! of course I want to spend as little as possible! how is that confusing? lol

but i would rather spend 3500 and get something that is going to perform than spend 2000 on mediocre crap...lol
i know i mentioned it before but I am buying lighting for a 4x8 canopy.

I have narrowed it down to lush lighting 2 dominators 2x's on a lightrail at approx 26'' from the canopy.
I honestly hadnt even considered ceremic metal halide....
I just thought it was something like the old 430w sonagro hps
or the sunmaster warm deluxe halides which I have used in the past,
I didnt know its was a new technology... I have to say I am impressed by the spectrum of the philips bulbs that I looked at, but its still an HID and while it is more effecient and does put out more PAR per watt it still is going to put out more heat that has to be exhausted which calculates into higher initial cost..... which i know going with led is going to cost way more up front but the closer the 2 numbers get the more i lean towards led....
3 of the kind L600's are still there too....idk
need to do something soon...lol

 

[FLAgreenthumb]

I wanted to say thanks to everyone for all the info and advice... I have definitely learned a lot about lighting.....

 

[rives]

I didnt know its was a new technology... I have to say I am impressed by the spectrum of the philips bulbs that I looked at, but its still an HID and while it is more effecient and does put out more PAR per watt it still is going to put out more heat that has to be exhausted which calculates into higher initial cost

Not to try and persuade you one way or another, but you need to understand that if an LED fixture is 30% efficient and an HID is 30% efficient, then they are BOTH converting 70% of the energy used into heat. The heat is released a bit differently, but it is there nonetheless.

Additionally, the CMH & CDM lamps do not "cast" heat like a HPS does. I've had buds grow within an inch or two of the glass on the hood with no ill effect. As I pointed out in the above posts, my tents actually run cooler with the the CDM fixtures than they did with roughly equivalent LED fixtures because of the vented hoods - the hoods take a pretty low volume of air to stay cool (Bell says 10-15 cfm for a 1000w lamp), and the loop can be both drawn from and released outside of the grow area.

Every LED fixture that I've seen simply blows the hot air into the grow area, upping the requirement for the air exchange that all has to go through the carbon filter before it can be released.

Best of luck with whatever you decide to use!

 

[flat9]

The sun is about 2200 umol's @ the surface of a plant, the AT is not overkill, maybe @ 3-6" inches.

One thing I wonder but has stopped me from the CMH is how efficienct they are? I mean an HPS is 31% efficient overall, so that remark has me a bit perplexed. The last thing I would call an HPS is efficient, great light source don't get me wrong.....power factor correcting and ballasts will always be it limitation or is that not correct?

Firstly, the sun is about 2000 or a bit more at full strength, no? So this may only occur for a max of few hours each day, and the average is much less.

Secondly, you're presuming that cannabis actually is able to use all 2200, which it likely isn't. See the following links, which suggests no point going beyond 1500 (I believe the light used was a flat, full spectrum, so with LEDs maybe that means 1000 roughly ... will explain below):

http://link.springer.com/article/10.1007/s12298-008-0027-x
http://link.springer.com/article/10.1007/s11627-008-9167-5

Finally, as Rives mentioned, not a fair comparison given the different absorption rates at different wavelengths (this is called the DIN standard 5031-10 curve). PAR is actually still not a great measurement in that it doesn't take this into account. A better measurement would pre-weight the amount of photons per unit area per second by the relative rate of absorption for photosynthesis at that particular wavelength, then add it all up. I believe in the literature this is called "yield photo flux density," but it varies for each species of plant since actually the absorption rates differ for each plant as well. The DIN 5031-10 curve is just an average over plant species, as I understand it. Attached is a slide from an OSRAM presentation explaining this concept with nice visuals.

Anyway, for LEDs, because they put the energy where plants most respond to it, I believe you could get away with less PAR (given its equal weight to each wavelength) and still rock the house, so to speak. Given the response curves, I'd guess you could get away with 2/3 to 3/4 of the PPFD ... so no point in going beyond roughly 1000 to 1150 for LED.

 

[flat9]

i haven't had any first hand experience with any of them. but your lead in comment confused me. you said you favored Lush but wanted to spend as little as possible. Lush has some pretty expensive lights!

lol... your right they are pricey! of course I want to spend as little as possible! how is that confusing? lol

but i would rather spend 3500 and get something that is going to perform than spend 2000 on mediocre crap...lol
i know i mentioned it before but I am buying lighting for a 4x8 canopy.

I have narrowed it down to lush lighting 2 dominators 2x's on a lightrail at approx 26'' from the canopy.
I honestly hadnt even considered ceremic metal halide....
I just thought it was something like the old 430w sonagro hps
or the sunmaster warm deluxe halides which I have used in the past,
I didnt know its was a new technology... I have to say I am impressed by the spectrum of the philips bulbs that I looked at, but its still an HID and while it is more effecient and does put out more PAR per watt it still is going to put out more heat that has to be exhausted which calculates into higher initial cost..... which i know going with led is going to cost way more up front but the closer the 2 numbers get the more i lean towards led....
3 of the kind L600's are still there too....idk
need to do something soon...lol

If you got 3500 to spend, don't waste your time buying 2 mega powerful lights (see discussion above regarding distribution). I'd seriously recommend getting 8 of the BSLED SPD112D 315-watt models. Each will rock a 2 x 2, and you'll have the most even blanket of light on a 4 x 8 space you could possibly imagine. 8 of these, with shipping, can be had for probably around 3200 but you'll have to contact BSLED.

Also check out the COB models from Houyi. Those also seem pretty amazing.

 

[flat9]

Not to try and persuade you one way or another, but you need to understand that if an LED fixture is 30% efficient and an HID is 30% efficient, then they are BOTH converting 70% of the energy used into heat. The heat is released a bit differently, but it is there nonetheless.

Additionally, the CMH & CDM lamps do not "cast" heat like a HPS does. I've had buds grow within an inch or two of the glass on the hood with no ill effect. As I pointed out in the above posts, my tents actually run cooler with the the CDM fixtures than they did with roughly equivalent LED fixtures because of the vented hoods - the hoods take a pretty low volume of air to stay cool (Bell says 10-15 cfm for a 1000w lamp), and the loop can be both drawn from and released outside of the grow area.

Every LED fixture that I've seen simply blows the hot air into the grow area, upping the requirement for the air exchange that all has to go through the carbon filter before it can be released.

Best of luck with whatever you decide to use!

Rives how do we calculate efficiency for an LED vs for HPS, etc.? One of the things I'm very confused about is how to look at a given lamp and determine how many watts are actually lost as heat. You can get some idea based upon the number of watts in the far infrared, but other than that I'm lost on this point. Please explain... thanks ahead of time.

 

[flat9]

The 860's are designed as a retrofit lamp for 1000w metal halide magnetic ballasts, and are vertical-mount only. I would think that they are probably also compatible with a 1000w hps mag ballast with the ignitor removed, but I really haven't looked into them that closely. There is at least one company building an electronic ballast for them, but the CDM lamps are all low-frequency only lamps, so they will not work on standard electronic ballasts.

Your point about the excessive PAR levels of some LED fixtures is right on the nose. That is what I was alluding to in the post above when I mentioned using multiple low-wattage fixtures spread evenly over the top of the canopy. What I, and several others, found a couple of years ago was that the trend toward higher-wattage LED fixtures with the same footprint was counterproductive. For instance, the Lumigrow ES330 has the power to bleach the shit out of the plants immediately below it, but fails to reach the edges of a 30x30 tent. Spreading (4) of the ES165s over the top of the canopy would be way more preferable than running (2) ES330s over the same area. So what did Lumigrow do? They obsoleted the ES165 and kicked the power up on the ES330 in the same size envelope.......

I've seen that electronic ballasts have much better PAR measurements for HPS and MH bulbs. Any ideas as to why this is so, and would this also be the case for CDM?

Sorry OP for kind of hijacking this thread, by the way! But there's some great discussion in here....

... now back to hijack! I'm running vertically mounted HPS 600 watters in an 8 x 4.5 as such now:

x-x
-o-
-x-
-o-
x-x

o = plant, x = light. Caveat: corner lights come on for 6 hours only (a la Heath's latest tree grow thread).

I'd like to reduce the power consumption a bit or at least the heat. Got no problem w/ the vert bulbs, anyway, but the 860 watts would be overkill. Given all the discussion of YPF, PPFD, etc., do you think I could get by with 400 watt CMH in place of the 600s? My other idea was just to replace each of the X's with an LED (two in the center, though) that are vertically hung ... or rather hung at a 45 degree angle.

That's a very frustrating trend with respect to making the most monstrously high-powered LED units, by the way. Huge waste...

 

[PetFlora]

I find leds not to generate as much heat AND, whereas hps/cmh are reflecting ~ half the light emitted, leds send 100% to the canopy. Basically, a 200w led is the equal of 400w hps/cmh- BASICALLY!

The problem with most consumer grade leds is the diodes themselves are the weak link. High quality diodes are available, but for whatever reason, only a few use them- A 51, Apache Tech come to mind

I think that is about to change, as we know better what we need, we demand better (thanks to the many DIYers), and at the end of the day, for the manufacturers, it's all about supply and demand. An educated consumer wins the day, an uneducated one (me in the past) gets hosed. It seems the manufacturers drank the same RB Kool-Aid and are having a hard time coming to terms with the false info. Computers can be easily rebooted, but humans? Not so much!

Not to try and persuade you one way or another, but you need to understand that if an LED fixture is 30% efficient and an HID is 30% efficient, then they are BOTH converting 70% of the energy used into heat. The heat is released a bit differently, but it is there nonetheless.

Additionally, the CMH & CDM lamps do not "cast" heat like a HPS does. I've had buds grow within an inch or two of the glass on the hood with no ill effect. As I pointed out in the above posts, my tents actually run cooler with the the CDM fixtures than they did with roughly equivalent LED fixtures because of the vented hoods - the hoods take a pretty low volume of air to stay cool (Bell says 10-15 cfm for a 1000w lamp), and the loop can be both drawn from and released outside of the grow area.

Every LED fixture that I've seen simply blows the hot air into the grow area, upping the requirement for the air exchange that all has to go through the carbon filter before it can be released.

Best of luck with whatever you decide to use!

 

[rives]

Rives how do we calculate efficiency for an LED vs for HPS, etc.? One of the things I'm very confused about is how to look at a given lamp and determine how many watts are actually lost as heat. You can get some idea based upon the number of watts in the far infrared, but other than that I'm lost on this point. Please explain... thanks ahead of time.

You are getting at my point, exactly. Our current measures are based on incomplete knowledge, and then extrapolated out based on the capabilities of our instrumentation - they only work well within a given family of light sources. We can easily compare the portion of light that is generated in the areas that we perceive and measure, but that is obviously different than what actually makes the plant function. We have a little knowledge about the reaction of certain compounds within the plant to various portions of the spectrum, but again, we keep finding more intricate relationships that we had no knowledge of before (green light absorption after red and blue are saturated, for instance).

Because of the single-bandwidth nature of the light generated by LEDs, it is fairly easy to come up with a measure of how efficiently the LED processes electrical power to make that segment of the spectrum, whether it be lumens/watt, par/watt, etc. However, broad-spectrum light sources like HIDs are pumping power into areas that are outside of our perception and theoretically could be outside of what our instruments are measuring or what the plant is processing. HPS for instance, shouldn't work as well as it does according to our understanding of plants - the preponderance of light that they make is the wrong bandwidth, they generate to much IR, etc, etc. A well-designed LED fixture should have the ability to easily outstrip them as a plant light source - however, we have seen that this potential is rarely realized. There is apparently something else going on.

The CDM lamps make this even more evident. My experience shows that they will have good results at a wattage/square foot level that simply will not work with other light sources. While they have a very good lumens/watt ratio or whatever measurement you like, it really isn't enough better than the other sources to explain why it works so well.

I've seen that electronic ballasts have much better PAR measurements for HPS and MH bulbs. Any ideas as to why this is so, and would this also be the case for CDM?

Sorry OP for kind of hijacking this thread, by the way! But there's some great discussion in here....

... now back to hijack! I'm running vertically mounted HPS 600 watters in an 8 x 4.5 as such now:

x-x
-o-
-x-
-o-
x-x

o = plant, x = light. Caveat: corner lights come on for 6 hours only (a la Heath's latest tree grow thread).

I'd like to reduce the power consumption a bit or at least the heat. Got no problem w/ the vert bulbs, anyway, but the 860 watts would be overkill. Given all the discussion of YPF, PPFD, etc., do you think I could get by with 400 watt CMH in place of the 600s? My other idea was just to replace each of the X's with an LED (two in the center, though) that are vertically hung ... or rather hung at a 45 degree angle.

That's a very frustrating trend with respect to making the most monstrously high-powered LED units, by the way. Huge waste...

The variations in output that we've seen from electronic ballasts I think are attributable to the ability to tune the frequency of the lamp's input power to the individual lamp. Some seem to react very favorably to a high-frequency input, while others like a low-frequency. A magnetic ballast doesn't have the ability to change the frequency of the signal. The CDM lamps also have sweet spots in the frequencies that they accept - they are built to run on either magnetic ballasts or low-frequency electronic ballasts. The GEL Ceramatek ballast is the glaring exception to this - it drives the 315 with a higher frequency, and claims to attain a much longer lamp life, lower lumen depreciation, and higher light output than other ballasts.

I've never used them, but the 400w CMH lamps are apparently much less efficient for our purposes than the 315, and are going the way of the dodo as a result. Azeotrope had used them for years and loved them. I talked him into trying the 315 and he was amazed at the difference in his results with all of the other parameters held constant - same strains, grow area, nutrients, etc. Tenthirty has been using 315s for quite a while now, and finds that they are not only outyielding his 600w HPS fixtures, but doing it with higher quality.

 

[PetFlora]

HPS has been around for growing at least 40 years: leds only a few.

I stay up on what is going on via LED Magazine. I can tell you the near future is going to be mind blowing, and what is currently available (Apache Tech/A-51/ Hans panel) is being proven in quite few grows that I have followed.

COBS have changed the game for ever, well the foreseeable future, anyway. Cobs are available in 10 20, 30, 50, 75,100, 200w chips/cobs, though they are driven at ~ 70%.

80/90/110* angle reflectors fit over the cobs are ~ 4" diameter at the outer edge, dispersing the light to a fairly wide area over the canopy. The resultant quality & intensity depends on the quality of the cobs

COBs are ~ 1" x 1" chips mounted directly on a sunflower shaped heat sink + one parabolic reflector per cob

Pretty simple to DIY using highest quality cobs (Cree, Vero, Luxeon...), but upwards of $5/watt, and worth it

 

[rives]

I think that there has been a huge, unforeseen benefit from the continued development of LEDs - the market pressure has forced some unprecedented development in HID technology. As you pointed out, HPS has been around for a long, long time. With some minor tweaks here and there, it is largely unchanged over that period. CMH has benefited from the competition of LEDs moving into what has always been the HID arena - parking lot lighting, architectural lighting, etc. The result has been the development of the CDM lamps.

It will be interesting to see where this back-and-forth development takes us. One thing is for certain, it is beneficial to us if we remain open to exploring the benefits of recent developments.

 

[FunkBomb]

Is there anyway to incorporate the lifetime of the light source and it's associated output depreciation into the measurement of efficiency? Or is that not feasible?

The reason I ask is that LED's have a longer lifetime compared to traditional HID lights. I'm not sure if ceramic metal halide falls into the category of traditional HID.

This lighting discussion has been great. Lots of learning going on.

-Funk

 

[rives]

I think that there are probably too many variables to try and incorporate that, particularly since it seems that most existing efficiency measurements are of an ad hoc nature. HID light degradation is all over the map, from HPS lamps that should be replaced every few runs to the GEL-ballasted CDMs that are claiming a life comparable to LEDs with a 90% lumen maintenance at 50% life. LED fixtures are also highly variable - the best of them will probably meet or exceed the chip manufacturer's expectations, but many others have inadequate heat sinking, too little airflow, and try to live out their warranty period by reducing the current to the LEDs far below optimum.

 

[flat9]

Rives where does one purchase these 315 watt ceramatek electronic ballasts & lamps? How many watts per sq foot does one need? You can hang these vertically I presume?

Also, DHF in vert always preaches 50 watts/sq ft w/ bare bulbs, and that's watt I'm running (3 x 600 with 2 bulbs off at any given time). To be honest, I think going beyond it is pretty much hard to dial in the environment though according to my calculates I may be able to benefit from more watts (as per the 1500 umol number). It'd be nice to keep roughly the same number of watts or a little less (5 x 315) and give it a whirl.

Any documented grows on here comparing 315 watt CDM to 600 watt HPS?

 

[rives]

The Ceramatek ballasts are available direct from GEL, who has a retro-fit "kit" available which used to come with either a PGZ18 socket or a mogul-to-PGZ18 adapter. Cycloptics was selling them separately from their reflector package but I don't see them currently listed on their site. You might try calling them.

I think that the traditional 50w/sq.ft. level is only applicable to HPS - other lighting families are substantially different in their light output per watt. I think that I mentioned that I just finished a grow with just shy of 20w/sq.ft. with the 315 that exceeded my best run with a Lumigrow ES330 by 37%, and would have been very close to doubling it if all five plants had performed the way that 3 of them did (two should have been culled, but those were all the clones that I had at the time).

I'm not aware of any comparison grows on here.

http://www.gel-usa.com/products/

https://www.cycloptics.com/store

 

[PetFlora]

The measurement of a quality led light is ~1000-2000 umols/m*/sec over a 3 x 3 area ~ 2ft below the light

Most consumer grade panels have < half that

 

[flat9]

Cheers for your help guys! Now I'm having a real hard time deciding whether to upgrade my flower rooms to CMH or to LEDs.

I can get quality LEDs (at least they've worked extremely well for me and do have an 1800 umol/sec rating at 24" under the light) for about $1.20/watt (actual draw) delivered to my door.

Rives may I ask what you yielded, what the strain was, what the medium was, etc.? Forgive me if there's a grow log on here that I'm missing. Anyway, that's a great result in that you beat your ES330, and congrats on your grow.

PetFlora you seem to know a lot about COB reflectors. What would be the advantage of going with the lower degree reflectors ever, if at all? The point I was trying to make before was that I think that LEDs concentrate far too much radiation in one spot, and this is their biggest failing... do you disagree?