? Cooling T5's in a DIY Wood Hood

[Deano]

I recently gutted my Aquatraders x4 24" T5 light fixture and put the bulbs in my Wood Hood over my 20 Long. The bulbs are now in individual GLO reflectors. The Hood is 30" x 13" x 4". Bulbs are about 4" from the water surface.

 

With the splash shield in place by the end of the light cycle the bulbs are running on average 180 degrees at the bulb ends. I removed the shield and the bulb temp eventually dropped to 140 degrees. My research indicates the correct operating temp for T5's is 95 degrees.

 

So want are my options

1: Run no heat shield and don't worry about the temp.

2: Run no heat shield and add some fans.

3: Run heat shield and fans.

 

If I run fans where would the best placement for them.

 

Here is a sketch up of the hood. The fan holes are not in the hood I just sketched those in just to see if you have any opinions on those placements.

 

[Mustang Boy]

I would run 2 fans on the end with the brand name on the bulbs and have 2 exits on the opposite side and keep the splash shield

[Deano]

I would run 2 fans on the end with the brand name on the bulbs and have 2 exits on the opposite side and keep the splash shield

 

I assume you are recommending the fans to push air into the hood blowing over the labled end of the bulbs.

 

Would that be no fans on the exit just a opening for the air to exit?

[Mustang Boy]

yes that would have it setup how most of fixtures are designed, this is because the label end of the bulb is the one that is supposed to be measured at 95 degrees

 

if you have 2 fans blowing in the air will exit on its own without a problem, i would even make the exit holes a little bit smaller

[nibor]

I would run 2 fans on the end with the brand name on the bulbs and have 2 exits on the opposite side and keep the splash shield

+1

 

yes that would have it setup how most of fixtures are designed, this is because the label end of the bulb is the one that is supposed to be measured at 95 degrees

 

if you have 2 fans blowing in the air will exit on its own without a problem, i would even make the exit holes a little bit smaller

Air temp along bulb 95F, temperature of endcap at label end 113F, temperature reading taken on the endcap itself. (According to The Grim Reefer and I don't doubt his research).

 

Exit holes should be at least the same size as the fans if not larger. Axial fans are not an effective way of moving air over a pressure difference. Many manufacturers, not just of aquarium lighting but also of computers, home cinema amplifiers, games consoles...., have very poorly designed airflow. Size the exhaust ports correctly and you can use smaller/slower/quieter fans and still achieve good cooling.

 

As an aside, laptop computers typically have much better designed fan cooling than other appliances. They use radial fans which are much more effective at moving air over the pressure differences created by the slim case size and small vents.

[Deano]

+1

 

 

Air temp along bulb 95F, temperature of endcap at label end 113F, temperature reading taken on the endcap itself. (According to The Grim Reefer and I don't doubt his research).

 

Exit holes should be at least the same size as the fans if not larger. Axial fans are not an effective way of moving air over a pressure difference. Many manufacturers, not just of aquarium lighting but also of computers, home cinema amplifiers, games consoles...., have very poorly designed airflow. Size the exhaust ports correctly and you can use smaller/slower/quieter fans and still achieve good cooling.

 

As an aside, laptop computers typically have much better designed fan cooling than other appliances. They use radial fans which are much more effective at moving air over the pressure differences created by the slim case size and small vents.

 

OK this confirms what I had been thinking. BUT measuring on the end cap does make a big difference on the end temp verse measuring on the metal part of the end of the bulb. As I recall and I didn't write down that measurement since I didn't think it would be relevant, but I think it was around 115F without the splash shield so I am not to fare off.

 

Sorry I meant splash shield

[nibor]

For ambient temperature only, you will have to convert ^oC to ^oF.

http://www.theaquariumsolution.com/optimum...rature-t5-tubes

 

Just to be clear: by endcap, I meant the metal end of the bulb (not the lamp holder).

A page I bookmarked a while back you may find useful.

[Deano]

For ambient temperature only, you will have to convert ^oC to ^oF.

http://www.theaquariumsolution.com/optimum...rature-t5-tubes

 

Just to be clear: by endcap, I meant the metal end of the bulb (not the lamp holder).

A page I bookmarked a while back you may find useful.

 

 

Appreciate it.

 

I am thinking 2 Silnex 60mm 14CFM or do you think I should go with the 80mm 32cfm and adjust fan speed as needed?

 

35C is 95F

[Mustang Boy]

to play it on the safe side i would get the more powerful ones that you can turn down to get the temperature just right across the bulbs, this will help extend the bulbs life and will increase the output because if a bulb is running too hot it isn't putting out the same amount of light as it would if the temp was correct

[Deano]

Well when I got home from work I measured the bulb temps at the label ends without the splash shield on all day and the temps were average 120F. The ambient, this was measured near the bulb between the reflector and the bulb was 110F average.

 

I am thinking 1 80mm Silnex 32 CFM. With 2 60mm holes at the opposite end. This should give plenty of exit volume.

 

I really don't want 2 fans although I have used the Silnex 80mm's before and they are dead silent. Noise is an issue since this tank will be in my office.

[Deano]

I was checking out Newegg for fans and a speed controller. Then I ran across this NZAT Fan Controller and temp monitor, looks kind of fun and interesting. What do you think? I could put the temp probes on each bulb and one for ambinate. You can let the thing control the fan speed automatically based on a preset temp on one of the temp probes.