Heatsinks for DIY LED lamps

FYI, for anyone interested, one way to do this is to find the high points on both the supply and return legs and add a tee at each. Then simply run a short vertical piece which ends in a reducer into which you'd screw in a bleeder valve, looks kinda like a car tire air valve, and works the same. Just push in on the center pin to release the air until water squirts out ;?)

EDIT: On second thought, unless you're dealing with a very LARGE number of these, I don't think 2 would be necessary, just one at the highest point should suffice.

Shugglet said:

Move the lights around a bit after you hook up and power it up and all the air will be removed.

Even if the tubes werent completely filled with water there would still be plenty to cool it regardless, gravity is keeping the water in contact with the hottest part of the tubing.

Click to expand...

Well I know in home heating systems which utilize hot water, you need to install these at the high points because enough trapped air can cause issues. However these systems aren't going to have the loops and zones either, so may not be a problem, but it just makes me feel better to design these in so as to avoid having to fix it if it does become one. The solution is cheap, so why not ;?)

I have built the tubes.... Yeah 1/2"is fine for flow, yes u get air after the pumps shut down, no it doesn't matter as long as you get water flowing again. I have to advise putting the outlet higher than the inlet if a lil trapped air scares you. The thing is water will flow across the bottom of the tube regardless that also happens to be where the heat is......

A larger reservoir just holds more heat......i had a 35 gallon trash can I ran for a res and it wasn't any better than 5gallon.....i switched back to 5gallon. Picture this whole set up as a cars closed loop cooling system, low volume with high flow through small channels.....

NastyN8t said:

A larger reservoir just holds more heat......i had a 35 gallon trash can I ran for a res and it wasn't any better than 5gallon.....i switched back to 5gallon. Picture this whole set up as a cars closed loop cooling system, low volume with high flow through small channels.....

Click to expand...

If your radiator is efficient enough at removing the heat, yes a smaller res wont hurt. If you build your setup and find your radiator not enough to keep the temps from climbing, a bigger res can be an effective cheap solution.

Also, as long as you have a res, a bleeder valve isnt really that necessary as the air can escape from the system there and wont get to a submersible pump. To inline pumps it might pose an issue.

NastyN8t said:

A larger reservoir just holds more heat......i had a 35 gallon trash can I ran for a res and it wasn't any better than 5gallon.....i switched back to 5gallon. Picture this whole set up as a cars closed loop cooling system, low volume with high flow through small channels.....

Click to expand...

Well that was something I wasn't actually going to go with at all, as in no res, just the closed loop system. Again I was approaching this from the viewpoint of a plumber, and these systems are always pressurized, from city water if nothing else. That has caused me some issues because if I mounted the radiator and pump on the floor, then how do I fill the system and purge all the air?

One solution was to place the pump above everything else, as in on the floor above, which works out because of the location of the upstairs access panel to the bathroom plumbing - in the room opposite the upstairs bath there's an access panel, through which you can see through a hole into the basement. So I had been considering putting the pump and a filler tube through which I would add the coolant and the radiator would hang from the ceiling in the basement right below this opening. Yeah, it's a bit of extra work, but I didn't want to have to deal with a reservoir which present it's own set of issues.

Shugglet said:

If your radiator is efficient enough at removing the heat, yes a smaller res wont hurt. If you build your setup and find your radiator not enough to keep the temps from climbing, a bigger res can be an effective cheap solution.

Also, as long as you have a res, a bleeder valve isnt really that necessary as the air can escape from the system there and wont get to a submersible pump. To inline pumps it might pose an issue.

Click to expand...

Exactly, since my design incorporated an inline design, this was the challenge I was facing with the reservoir. I think that if my design is sized properly, a closed loop will work fine, but vamos a ver, lol.

fearnoevil said:

I didn't want to have to deal with a reservoir which present it's own set of issues.

Click to expand...

What issues would a res present? Also, coming from the plumbing perspective Im not so sure I would call a house a closed loop because it is pressurized by an outside, constantly refreshed source.

Your method seems like it would work no problem, but I feel it may just be a little extra work.

fearnoevil said:

Exactly, since my design incorporated an inline design, this was the challenge I was facing with the reservoir. I think that if my design is sized properly, a closed loop will work fine, but vamos a ver, lol.

Click to expand...

Itll work, but any sized res placed before the pump in the loop will eliminate a lot of your issues me thinks. Also, if you have no res you will have no choice but to add more cooling power to your system if the existing radiator cant keep up. In effect a res acts as a nice buffer and extra options to combat the heat if needed.

As to the system volume -vs-heat exchange capacity, that's something I'm not sure of. I've never done much in the way of HVAC work, so never really had to calculate that stuff, just plumb in the appliances, lol. So I've been thinking (always a problem, lol) about how the flow in gpm affects the ability of the radiator to dissipate the heat.

SO in a hypothetical situation, if you had a variable speed pump with an imaginary range of from 1-100 gpm, one might think that the faster you ran the pump the better it would cool, but for some reason I believe that for a limited sized exchanger, like a small car radiator and box fan, would it make more sense to run closer to the lower range in order to allow more time for the water to lose enough heat absorbed to maintain efficiency?

The way I see it, if you ran the pump at 100 gpm, wouldn't the water accumulate more and more heat if it passed through the radiator so fast that it barely had time to lose any heat?? Of course I could be completely off, and any advice from anyone with any expertise in these systems would be greatly appreciated ;?)

Shugglet said:

What issues would a res present? Also, coming from the plumbing perspective Im not so sure I would call a house a closed loop because it is pressurized by an outside, constantly refreshed source.

Your method seems like it would work no problem, but I feel it may just be a little extra work.



Itll work, but any sized res placed before the pump in the loop will eliminate a lot of your issues me thinks. Also, if you have no res you will have no choice but to add more cooling power to your system if the existing radiator cant keep up. In effect a res acts as a nice buffer and extra options to combat the heat if needed.

Click to expand...

Actually, most boiler systems used in residential heating are on a closed loop, with a low level detection control that adds more water as needed, whereas a circulator for potable hot water supply isn't, just an added loop to keep the hot H2o circulating.

fearnoevil said:

As to the system volume -vs-heat exchange capacity, that's something I'm not sure of. I've never done much in the way of HVAC work, so never really had to calculate that stuff, just plumb in the appliances, lol. So I've been thinking (always a problem, lol) about how the flow in gpm affects the ability of the radiator to dissipate the heat.

SO in a hypothetical situation, if you had a variable speed pump with an imaginary range of from 1-100 gpm, one might think that the faster you ran the pump the better it would cool, but for some reason I believe that for a limited sized exchanger, like a small car radiator and box fan, would it make more sense to run closer to the lower range in order to allow more time for the water to lose enough heat absorbed to maintain efficiency?

The way I see it, if you ran the pump at 100 gpm, wouldn't the water accumulate more and more heat if it passed through the radiator so fast that it barely had time to lose any heat?? Of course I could be completely off, and any advice from anyone with any expertise in these systems would be greatly appreciated ;?)

Click to expand...

If you ran it too fast it would go through the radiator with out cooling as much so there would probably be an optimal flow. How well the fan on the radiator pulled/pushed air through would have a factor as well.
It's like fine tuning fan speed vs flow rate...A reservoir would act like a buffer for pressures and heat.

Uberknot said:

If you ran it too fast it would go through the radiator with out cooling as much so there would probably be an optimal flow. How well the fan on the radiator pulled/pushed air through would have a factor as well.

Click to expand...

The thing we have to remember is without a res, your radiator MUST remove equal or more heat from the system than is being added to it. As long as that condition is met, increasing flow rate will increase efficiency and performance. If that condition isnt met, you need to go back to the drawing board regardless.

fearnoevil said:

Actually, most boiler systems used in residential heating are on a closed loop, with a low level detection control that adds more water as needed, whereas a circulator for potable hot water supply isn't, just an added loop to keep the hot H2o circulating.

Click to expand...

Ah yeah, I was more thinking general plumbing, not heating/cooling. lol

Shugglet said:

The thing we have to remember is without a res, your radiator MUST remove equal or more heat from the system than is being added to it. As long as that condition is met, increasing flow rate will increase efficiency and performance. If that condition isnt met, you need to go back to the drawing board regardless.

Click to expand...

oh yeah sooner or later all the stuff will heat up and every thing will equalize in temps for sure you want a nice fan!

The reservoir would require an additional tank, sealed to keep bugs/contaminants out, plus having to check it every so often to make sure it's not running low, adding water/coolant or having any other problems. The advantage of a closed loop is that once it's up and running, and through the "run in" period (after all the air has been removed with zero leaks), it becomes relatively maintenance-free, basically set it and forget it). Again, I'm looking at this from a different viewpoint then a lot of DIY folks, so it just goes against what I'm used to thinking of - i.e. a system needs to be able to run without the need for constant monitoring,

That doesn't mean that I'm dead set against this, just one way of looking at it.

I would think you could make a simple sealable res from a 5gal bucket and lid. Just thinking of other possible options for ya.

Also, as long as your radiator is removing more heat that is being added, your "optimal" flowrate will then depend on the ambient temperature around your radiator. Lower ambient temperatures would effectively raise the optimal flowrate leading to better performance.

I use 500gph pumps 5gallom reservoir with 520w

Shugglet said:

I would think you could make a simple sealable res from a 5gal bucket and lid. Just thinking of other possible options for ya.

Also, as long as your radiator is removing more heat that is being added, your "optimal" flowrate will then depend on the ambient temperature around your radiator. Lower ambient temperatures would effectively raise the optimal flowrate leading to better performance.

Click to expand...

Yeah, agree on the ambient air temp. although a bigger fan or doubling up on the radiators/fans, but that goes against trying to keep the cost of the system down. I have to keep reminding myself that that is one of the main points to switching to an LED system, to lower how much I have to pay the thieving power company, lol.

I'll have to consider the res idea and how to incorporate that into the design, I'm sure there's ways I'm not thinking of at the moment ;?)

Well you have to know what the radiator fin count and thickness is before picking out fans.

NastyN8t said:

I use 500gph pumps 5gallom reservoir with 520w

Click to expand...

Well if I get to where I'm planning on going, I'm thinking my total system will include 8 fixtures @ between 400w-500w each, although I still need to nail down the optimal number of cobs for my grow area. So not completely sure this system will do that, lol, gotta test the watters so to speak ;?D

fearnoevil said:

Yeah, agree on the ambient air temp. although a bigger fan or doubling up on the radiators/fans, but that goes against trying to keep the cost of the system down. I have to keep reminding myself that that is one of the main points to switching to an LED system, to lower how much I have to pay the thieving power company, lol.

I'll have to consider the res idea and how to incorporate that into the design, I'm sure there's ways I'm not thinking of at the moment ;?)

Click to expand...

I was getting at the fact that once you get your system to the point where it can keep water temps within a few degrees of ambient, nothing short of a chiller will help.

Id just recommend a 5gal bucket with inlet and outlet and one of the "higher grade" lids with a rubber gasket.