knna

-Heatsinking-forced cooling: what I tried to explain on my post about heatsink is they are required, independent of if you use forced cooling or not.

A heatsink is required because its what sink the heat released by the LED. After that, you need to dissipate it from the heatsink, of course, and there is where the type of cooling enters the game. Any heatsink has a thermal resistance depending of its caracteristics, that determines how much it warms depending of the heat that it supports. Its usually given on K/W (Kelvin degrees, same scale than ºC, that the heatsink goes over an ambient temperature of 25ºC for each watt of heat loaded on it).

That figure on K/W is given for the heatsink on open air, thus heat dissipation is done by convection (it heats the surrounding air). If you apply forced cooling to a given heatsink, you may get a way lower figure of thermal resistance, meaning you are going to be able to keep the heatsink cool still loading it with many more watts.

The lower the heatsink surface area for each Watt loaded, the forced cooling must be stronger. If you use a copper pipe to mount your LEDs, its the heatsink . Although it would depend on how hard you run your LEDs and how are the spaced on the pipe, a half inch pipe has little surface area. So the only way to keep it cool enough is by using forced cooling. Depending of the density of power installed on it, it may be aircooling (blower attached to the pipe, so air going thought the pipe cools it) or water cooling (or another liquid).

Water has an high heat specific capacity, so it can rid off large amounts of heat. With water going continously into the pipe, for sure its going to keep cool enough for your LEDs. If I always advice to use a heatsink with larger surface area is because if not, any failure may damage seriously your LED array. If you design the system in a way that eliminates that possibility, go on.

I dont like the idea of using water to waste for it. I think green and water is a very valuable resource on our planet. But its an ideology statement, not a technical one. I suppouse in Canada water is not a problem (In Spain, it is)

-Thermal adhesive-kapton tape: there is no opposition for both. You can use both at a time.

Kapton should be used always on DIY LED mounting directly to heatsinks. Its task is to electrically isolate heatsink from LEDs. On your case, you should use kapton on the copper pipe (if you mount LEDs directly to it) or on the narrow aluminiun sheet (if you use it to mount LEDs). Specially with water involved, you wont want any electricity flowing on that pipe.

Kapton may be used as a continous layer below the LED, adviced when the bottom slug of the LED is electrically conected, or as minimun, below the electric contacts.

For mounting over a pipe directly, using Crees would make sense, as their electrical contacts are on the sides, while the botton slug is already isolated. With solder paste, you could solder them directly to the copper pipe, as far as you can put it at 230ºC. As its curved, the electrical contacts will get on the air, and you will can solder wires to them, once soldered to the pipe. Probably, in this way of mounting, you wont need kapton (although I would use it anyway: as its unsolderable, you can limit the area to be soldered with it, and ensure it not reach electrical contacts).

-Replacing a 1000W Hortilux. If you want to do it just with LEDs and want to keep yields constant, you are going to need to install about 500W of top bin LEDs. Its going to be expensive. Although at 150$ of bulb each six months (why so fast replacement? have you measured lm drop with age? If so and it goes below 85% in less than six month, there is a problem of ballast-bulb compatibility), you will recover it relativelly fast, just on bulb's replacement savings.

Im not advicing to replace large HPS for LEDs right now. I think its wiser to start with a small LED system and check if its able to produce what you need before replacing a large HPS system. One of the advantages of LED setups is their scalability. You can use the same light density on a small cab than on a large room. What about starting an small scale prototype? Say 100W of LEDs on 1/5 of the surface area that the 1000W is covering. So you can test how the water cooled copper pipe works and if it produces what you need to replace the 1000W. For the time you have it clear, LEDs will have dropped in price.

An interesting choice for those using expensive blue enhanced HPS as the Hortilux is to install some blue LEDs around it and use a standard 20$ HPS. Blue LEDs are way more efficient emitting blue than the enhanced HPSs, and with the current price gap between blue enhanced (horticultural HPSs) and standard ones will does it very profitable, paying itself on the first bulb life.


knna

__________________


Bulb Analyzer Tool

scottcanada

Dear KNNA:

Thank you kindly for taking the time to answer my questions in detail. I sincerely appreciate your efforts, support your vision, and want to help you any way I can (buying from you in 2 weeks when your website is done would be a good start).

Hopefully this post will address some of your questions and won't require a detailed reply from you.

Water being wasted. Yes, I completely agree with you that water to "waste" is bad, but there are many applications where the water could be recovered. For example, you could design a system in your house where the LEDs from your grow room were being used to heat the hot water tank for the house (makes the energy savings of LEDs even better). In my application the water I feed my plants needs to be warmed most of the time. My tap water can be as cold as 7 degrees in winter and I usually want it to be in the 18 - 20 range for feeding. I grow in large dirt containers so it takes approx 500 litres of water every 3 days to feed so much of this could act as a heat sink for me. In Canada the government is already licensing small grow operations (up to 3 patients) and the courts have just forced the government to allow large scale legal grows (for more than 3). If a large scale LED system could be set up it could heat the water for a nearby apartment building.

Another important thing to consider is that the water does not have to be clean, it just has to be cold. Why not use water that is already considered "waste" water if it is cold or even room temp (you could collect rain water off your roof in barrels and pump it through your system).

Lastly, keep in mind that although a pipe may be 1/2 inch thick, you don't need to have a high flow rate. Just a small trickle of cold water down a copper pipe should be plenty to cool the LEDs. I am interested in water specifically because I already have so many water lines in my space that the water appears to be the best heat sink for me (air where I live gets VERY hot and humid in summer so is a bad heat sink). For me I have lots of water but need to bring my power down. I am already running 8,000 watt of HID + air conditioners, dehumidifiers, etc so you can see why I have an interest in bringing my power and heat down.

The other major concern people have is about "penetration". I think LED could replace 1000 watt HID under certain conditions. For instance, a combination of a really good heat sink like cold water, lots of LEDs not being run too hard to get good efficiency, and training the plants to grow flat. If it can be done DIY for about $1000 and match yield of 1000 watt HPS then it will take off. If it cannot match 1000 watt HPS under any conditions then there is no future for it. This is why I am bent on matching 1000 watt HPS because this is what everyone wants to see. Maybe water cooled could be the solution for those who need the 1000 watt results.

I think Low Stress Training or SCROG is very important here. If we can even get 1 foot depth of excellent light use evenly distributed then we should be able to match yield of HID. I think some people try to grow large 5 foot tall (1.5 meter) plants with LED and can't match HID so give up on it, but I have yet to see a really good LED system used on well trained plants that do not require more than 1 foot of penetration for max yield.

I should mention that I have been in contact with the manufacturers of the SmartLamp and SmartBar. When I asked them about the spectrum of their systems they would not give me specifics, but they did say a test of their product was just done and the study will be released in a few weeks. This is what they said in an email to me:

"The CIDES Experts en serre - Experts in greenhouse that is an independent researchers community made a study on our product and they are very impressed. Their studies about our product will be published on Theoreme Innovation, CIDES and GoLedGrow sites in few weeks . Click on that hyperlink to see the CIDES site in english. A thing that they found among others is that the agronomic gain of our product is 1.55 per watt. Compared to the HPS lamps which is 0.75 per watt. This mean that the reason for choosing our lamp is not only energy saving. Is to give the best results on the market for plant growth."

What does "agronomic gain" mean? I have never heard this term before. If anyone has purchased this system before, maybe they could let us know what combination of bulbs were used. It seems to me that everyone is trying to "re-invent the wheel" when it comes to the spectrum. Has the plant sensitivity curve not been determined a long time ago? Why is there still so much debate over which spectrum is best and why is it something the company feels they need to keep secret? This is something I want to know before buying it just like Hortilux puts on their site.

Sorry for the long post. You must be very busy and I hope I have not tied up too much of your time. I just want to say that I sincerely support your vision of providing affordable access and unbiased information about LED setups for DIY people. I will do only the same with the info you provide me and will certainly refer people to your website for purchases. Thank you again and bye for now.

scottcanada

This post is to provide more context on my grow (mainly my use of water), and to see if KNNA or anyone else with a good technical background can comment on water cooled air conditioning or hydronic cooling. I am sorry if this is posted in the wrong place, but hopefully others on this forum may find it interesting.

I recently purchased 2 used machines off a buy/sell website and I'm trying to understand how they work. The guy told me it was a water cooled air conditioner or a hydronic cooling system. It consists of a huge fan (at least 2000 CFM) that pulls a huge amount of air through a radiator with fins (radiator is about 2 inches thick, then 2 feet x 2 feet square. The radiator has a water in and a water out line. The person who sold it to me said you run very cold tap water through the radiator and the fan moves tons of air through it. It has a draining system to remove the condensed water that will form. He said it only works in high temp and humidity so it great for grow rooms. Now I am trying to understand exact how this works, if it even works, at what temp is best, and if it can do both cooling and dehumidification. In my google searches I think I have now understood it, but want to check with someone who know about heat transfer in detail. The main advantage of this system is that it does NOT use a compressor so therefore uses less electricity (mine uses 5 - 6 amps at 110 volts to power the huge fan, compared to 12 amps for a 12,000 BTU air conditioner).

I have posted to a server an image of a psychrometric chart with lines and dots added to it by myself in photoshop to illustrate my understanding, and to allow others to correct me if I am wrong. Here is the chart: http://s94391194.onlinehome.us/free/ocn/chart.jpg

My understanding is that if I can get the temperature of the water below the dew point of the air in the room, then the machine will work at least as a dehumidifier, and hopefully it will cool the air as well (not sure about this). My understanding of why this type of machine is not used in most applications in my country is that getting the water temp low enough would be hard, and then again the issue of wasting water due to the lack of effective water recycling systems in existing buildings. This is why they use a compressor with a refrigerant to achieve the very low temps required to cool a room already at 20 degrees - correct? But getting back to the chart linked to above:

Am I correct in my understanding that point A on the chart represents 20 degrees C and 30% RH, and that at that point the system would not change the air even if the water temp in the radiator was only 5 degrees C.

Point B: This point represents 30 degrees C and 30% RH. At this point the system would cool and/or dehumidify the air if the water temp in the radiator was 5 degrees, but not if the water temp was 15 degrees.

Point C: This point represents 30 degrees C and 50% RH. At this point the system would cool and/or dehumidify even if water temp is 15 degrees C, and the efficiency of the system would be greatly increased if the water temp could be brought down to 5 degrees.

Am I reading this right? I have the equipment but purchased it in the winter and the temps are not quite high enough yet here to try it. I just want to check the theory on it. I also don't have reliable data yet on the temp of my tap water on the hottest summer days. It may go above 20 which would make this useless, but regardless I really want to understand the theory behind this because there are other applications like pumping cold water from the bottom of a lake in summer to get the cold water.

If anyone who knows something about heat transfer or cooling systems could comment on my chart to let me know if I am getting it, and if you could advise on if the system could actually cool the air and dehumidify and how exactly to calculate this.

Thanks, and sorry if this is posted in the wrong place.

random

Hitting that thank you button wasn't enough.. THIS THREAD RULES! Thanks so much for all the information man.

-random

knna

Very good observations, scott. I was working on a design of a LED greenhouse water cooled and reusing the water for the own greenhouse. More than to irrigating plants, what I though was recirculating the heated water in pipes below the ground, to raise soil temperature for greenhouses working on cold environments.

Preheating water for irrigation is a great idea, too. If the water is going to go through the pipe just once, you wont need a high flow rate at all. I love the idea of using the heat from a large LED system to preheat water for a building requeriments. Near 100% efficiency of the electricity burned!

About the air conditioned system using water. Those systems are having a fast development on the last years, as they are way less energy consuming than those using compressors. But their use is limited to certain situations: cold and very humid, achieving heating and dehumidify, or hot and dry, achieving cold air and humidify.

Think that vapor water releases heat when it condenses, while water evaporating consume heat. If you condense water (dehumify), the system wont release cold air, but the opposite (air temp over ambient temp). I live on a hot and dry climate, so I use this system myself, as I get increased humidity and cooled air at a time. I really only have seen these systems used on this situation, and I dont know very well how they work with high humidity.

Let me remark some important about LEDs: they only limit height of plants being grown when using only a flat lamp on top of plants. When using side lighting, you can grow plants way taller than any HID can do (producing buds, of course). Im building right now a "demo" cab, 45*45cm (1 1/2'), 2m tall (6' 8"), with LEDs along all the walls to show this concept. Hpefully Ill do a journal soon with it, growing a sativa.

We have join an experimental LED group on my spanish forum. 20 growers trying different ways of growing with LEDs: different spectrums, different styles of growing. But most of them are using side lighting. As grows goes advancing (most are now just starting, on veg stage), Ill link pics so people may see how they works.

I fully agree with this. I think a 1500$ system would be cost effective too, but no much more. On the current state of the art of LEDs, its very difficult to achieve it, almost impossible currently. Due that I still dont suggest to replace large HIDs for LEDs.

But with a drop in prices about 20-25% each year, and perfomance/price of LEDs increasing at a rate over 30% a year, its only a matter of time. Maybe as soon as at the end of next year, its going to be feasible. Right now, I think LED lighting is only cost effective on small grows and as supplementary lighting.

I would like to know it, for sure , as there some different ways to define it.

On one hand, it can be defined as the ratio between yield and energy burned (g/Watt). This would be the definitive ratio to look for.

But if they want to let out the equation the energy efficiency of the lamps, they can define it as ratio between yield and light used (either on uE or PAR Watts). This would inform of the agronomic efficacy of the light spectrum used.

Last week I had as discussion with John from Theoreme Innovations. He defend that the TI may produce 4x for each uE than HPSs. I told him to prove it, as the highest differences in spectral efficacy reported on the botanist literature is 2x, and not respect to HPSs. Respect to HPSs, Ive found enhancements of 50% max (on a uE basis).

Im still waiting for the answer, so I think we will find the answer on that study. But I hope people at the CIDES improves a little their methodology and documentation. On a study from 2005 on their site about a comparision of HPS vs HPS+LEDs, they calculated results based on a energy efficiency of LEDs of 97.6%, that is completely out of reality.

The TI Smartlamp uses a lot of different colors on it: near UV, blue, yellow, orangem red, deep red and far red, building an MH like spectrum but with way less green emission. But the problem of this lamp is not of the spectrum, but of the energy efficiency, way below half of HPSs

What is very well stabilished long ago is the photosynthetic response of plants to isolated wavebands of light, that is pretty similar for all superior plants.

But each plant specie respond differently to light having all or some mixed colors on it. For example, cucumber prefer a yellowish spectrum (perfect for HPS lighting), with near 40% of light in the middle region of spectrum (500-600nm) and no more than 30% of red, while tomato (probably, cannabis too) prefer a spectrum dominant on red and blue, with just 20% of energy on the middle region and at least 60% on the red.

With tomato, giving 30% of the energy to the red from the green/yellow doubles the yield respect to an spectrum that is perfect for cucumbers. Wheat, on the other hand, dont mind at all the light spectrum. It produces in direct proportion of the photons received, irrespective of its wavelenght.

Nobody has found the optimal spectrum for cannabis yet. Much less if it varies along its growing stages. And still much less about the response of the resin production and its composition to light quality.

With LEDs, fortunatelly we can perform that investigation. For sure, nobody is going to do it for us But we are just starting to do it, to have it defined when LEDs becomes a cost effective solution for any grower on the next years.

(Due that I called the group on the spanish forum as "experimental": we are trying to find the best spectrum for MJ growing)

__________________


Bulb Analyzer Tool

torchthatbird

Wow, this is good stuff. I'll have to come back and read it when I'm not stoned.

__________________


Torchs' 4th of July Mutistrain Grow. KongXChemo/Grapefruit, Jock Horror and Bubblicious. Complete

Torchs' Winter Rebound Grow:Nirvana K2, Skunk Bagseed, G-13 Shop KongXChemo/Grapefruit(complete)

Sensi Star: Breed Seed Project....take two(no seeds/much weed)

Jouney to the Cup(members only)

Chocolope from DNA(complete)

TTB's Halloween Multistrain Grow: Chocolope/Afghooey/Tasman Haze(complete)

Revisiting K2, The Red Diesel/G13Haze/Jock Horror Summer Fun Breed Seed Project(members only)

NEW STRAINS! KG26 and Fruity Haze

scottcanada

I found a textbook on google books that deals with this. The whole book is excellent and I want to buy it, but specifically related to this air conditioning system is pages 849, 850 and 851. Only problem is page 850 is missing which is why I don't really understand this concept. I will have to go to the library on monday, but if anyone wants to see what I was basing my chart on, scroll to page 850 here: Refrigeration & air conditioning ... - Google Book Search

It seems to indicate that for people like me with both high temp and high humidity that this would remove both (they refer to them as latent heat and sensible heat). Just want to confirm from someone who knows. Not super urgent or super important tho cause i will have a test within a few weeks, just kinda curious to chat about it in advance

delta9nxs

“optimal spectrum for cannabis” = holy grail! It would be great if it turns out 1 or 2 wavelengths of led light will grow large, dense buds. But, cannabis has evolved under the full power of the sun, and there are many studies showing that adjacent or accessory frequencies off the chlorophyll A and B peaks are also able to contribute light energy to the photosynthetic process. Has it been demonstrated that these accessory wavelengths are absolutely necessary? If so, in what ratios? For that matter, does anyone know with certainty what the proper ratios of red and blue should be for cannabis? Also, there is the entire issue of light at certain wavelengths and intensity acting as biological triggers at different stages of growth. Leds may present an opportunity to manipulate growth and flowering in ways that could accelerate the time frame to harvest, as a grow report on another site suggests. They were using red leds during the “dark” period and claiming that there was no tendency to revert to veg and that photosynthesis was still taking place, thereby pushing the finish date up. And, what about claims that more blue in flower increases potency? is cannabis, having evolved using full spectrum light which is unbroken along the along the spectral curve, able to utilize efficiently intermittent sources of light that seem to me to be more staccato like or interrupted by gaps? Could leds, which are instant full power, be strobed rapidly enough to trick the plant into believing it is getting a continuous stream and possibly further reduce electrical consumption or extend diode life? We haven't touched on the IR issue either. Do leds provide the right amount of IR for optimum growth?

I really don't expect answers to all of these things. Anyone who feels like commenting on any of this is more than welcome. The possibilities of this technology appear to be endless. I would be willing to participate in a series of small experiments attempting to quantify some of this stuff. It would not be too much of a burden on any one experimenter if many people were to individually tackle different aspects in a coordinated way. Similar to what knna is doing in the spanish forum. Perhaps they could be joined somehow with knna as the coordinator. Language is not a barrier to sharing for the common good. I speak and understand quite a bit of spanish, abeit not fluently. I think I might be able to understand the writings of the spanish language group, but I know I would not be able to participate in a technical discussion in spanish. Well, that's all folks!

choader

You've suggested that to replace an HID with LEDs you need 50-60% of the HID wattage. What about replacing CFLs, even big ones, with LEDs? Any guidance? Or LED W/sq ft?

Thanks.

Choader

__________________
SS Bubblegum & DP Mazar Small Cab - Complete
Choader's 600W HPS Grow - Complete
Small Cab Grow - Complete

READ ME!! Posting Guidelines
Male and Female Pre-Flowers

knna

Excelent post, delta.

I think we are going to be able to find the answer to all those esential questions, as LED lighting goes settling on the growing scene, as preformance/price improves.

I really would like we could find them before LEDs replaces massively HIDs (4-5 years? sooner?). As more people using LEDs and trying different things, more possibilities of we can find them. Organized groups may accelerate the process. There is no need to direct the investigation, but simply coordinate it and stabilish a general agreement so results of each grower are meangliful to the others, so we can compare different setups results.

For me, at least two things are required to do comparable different setups:

1)Amount of light used well determined, on any radiometric unit, preferably, uE, but PAR Watts should suffice. For this task, its a must to use LEDs with emission known and well caracterized. Both absolute amount of light used, im uE (same as lm, but way more meangliful and universally valid independent of spectrum used), and the average density of light used : uE/m2 or uE/sq ft should be enough, but info based on volume of plants would be way better (uE/cb ft).

Average spectra info would be desiderable, but its easy to calculate once LEDs used and their current draw are known.

2) Standarized way of reporting yields. After Cannastats experience, I suggest to use total fresh weight and/or bud's fresh weight. As more info the better, if its given together with dried weight, we will be able to study deeper how the way of manicuring and drying affect it and the differences between growers.

3)Ideally, we should have info of the resin production, aditionally to the bud weight info. So we can study how light affect resin production. But standarizing ways of extraction is difficult, so probably we only may compare results from same grower, or check general trends.

Im sure only little growers would help giving info of 3), but it should provide unvaluable info to all the MJ community. Cannabinoid profilling would rock , still using an aproximative method as Cannalyze.

On the absence of cannabinoid profilling, shared smoking may help on qualitative reports On the spanish group, we are planning to join together at the end of the year to share some LED grown MJ .

As more people involved, faster development, but I believe still an small group of people with good will may advance a lot.

__________________


Bulb Analyzer Tool