1
\$\begingroup\$

I'm having trouble controlling the heat from running a bridge rectifier (Similar to http://www.amazon.com/1200V-Single-Bridge-Rectifier--QL100A1200V/dp/B00EZ8BUMU/ref=sr_1_2?ie=UTF8&qid=1459556648&sr=8-2&keywords=100amp+bridge+rectifier) at high current for fast electrochemistry. The low voltage (9V) high current (40-50A) keeps heating up the bridge rapidly and will lead to thermal runaway if not managed by cutting the power by a switch. As my bridge rectifier comes with a large heat sink, similar to the one in the amazon link, I'd like to know some other options available for controlling the heat generated by the bridge to increase the time I can use it without cutting the power.

Options I know about:

Air Cooling (Essentially having a fan (the bridge is in a box I made out of medium density fiberboard, bolted) push air through the box through a few holes, cooling the bridge) (Uncertain about effectiveness given the intense heat buildup)

NOTE: The bridge is a full wave bridge rectifier. The bridge is mounted 90 degrees from how it would be sitting on a table and lies on the side of the box. (Which isn't necessarily the best for convection, but it will lie directly in the air current path of a fan)

Water Cooling: All Options are by thermal conduction.

1)Indirectly: The mounting bolts stick out of the box by about 3", this leaves more than enough space to attach a great thermal conductor, such as aluminium flat bar (Looks like: https://www.google.com/search?q=aluminium+flat+bar&source=lnms&tbm=isch&sa=X&ved=0ahUKEwj9l6CX3O7LAhXIlh4KHbaXCyYQ_AUICCgC), to the mounting bolts and having them in a container of cold water underneath. The idea here is that the aluminium will try to equalize the temperature of the bolts which are being heated by contact with the bridge and thus, cool the bridge as the bridge will spend heat energy trying to heat the bolt.

2)Directly: Immersing the bridge in water (I have no idea if this type of bridge is water tight, hypothetically it should be) a quarter way up to the base (which is rather high) (Done to ensure no shorts due to water), which seems reasonable. The water level can be maintained by putting it in a container with holes drilled at the 1/4 mark of the bridge and refreshed by a carefully placed supply hose.

3)All In: Total submersion in reasonably pure distilled water (might work, I'm not sure if the two AC nodes will be able to conduct all that well in relatively pure distilled water at 120V) (I'm aware of distilled water contamination)

\$\endgroup\$
2
  • 1
    \$\begingroup\$ Immersion cooling is often done in mineral oil. \$\endgroup\$
    – Daniel
    Apr 2, 2016 at 2:18
  • \$\begingroup\$ The obvious approach is to throw away the rectifier and to get a high current pre-assembled power brick. You only need 500W so there's an ample variety of cheap units to chose from; you may be lucky enough to even get a fanless one. If you could get your voltage up to 12V sourcing one quickly would be even easier. \$\endgroup\$
    – oakad
    Apr 2, 2016 at 9:34

2 Answers 2

Reset to default
3
\$\begingroup\$

If the bridge rectifier and its heatsink are in an enclosed box, you are definitely defeating the purpose of the heatsink, as little heat will escape from the box.

Adding vent holes in the box, and a fan to move air through the box and over the heatsink should definitly improve things.

Whatever you do, you must remove heat from the vicinity of the heatsink. Just immersing the heatsink in a non-moving water bath will only delay the overheating.

\$\endgroup\$
4
  • \$\begingroup\$ So, you're saying that whatever water environment should not be stagnant and should be stirred/replenished by a larger water tank? \$\endgroup\$
    – Sarah Szabo
    Apr 2, 2016 at 1:27
  • \$\begingroup\$ Yes - not just stirred, but circulated. You must move the heat away from the rectifier/heatsink. You could pump the heated water through a radiator that is then cooled by moving air (like a car radiator). \$\endgroup\$
    – Peter Bennett
    Apr 2, 2016 at 1:40
  • \$\begingroup\$ Yes, the whole point of a heat sink is to provide a method of absorbing the heat so that you can move it away from the device. That's why they generally have fins, to increase the surface area for which to contact and transfer heat between mediums. This medium could be air or water, but it must have a flow that leads the heat away from the device. If the medium is to be recirculated, you need a method of removing the absorbed heat before the next cycle. \$\endgroup\$
    – Jarrod Christman
    Apr 2, 2016 at 2:08
  • \$\begingroup\$ @JarrodChristman: I'd say that the purpose of a heatsink is to transfer heat from the source to a medium that can remove the heat from the area around the heat source. \$\endgroup\$
    – Peter Bennett
    Apr 2, 2016 at 2:54
6
\$\begingroup\$

Have you considered not making so much heat in the first place? You can use Schottky diodes which if well chosen will halve the power wasted. If you know what you are doing then look at some form of synchronous MOSFET rectifier. If you use low Rds(on) devices which at low voltages are not hard to find you will waste even less power.

\$\endgroup\$
1
  • 2
    \$\begingroup\$ Seconding synchronous rectification. LTC makes a bridge controller IC for precisely this job, even. That and four big NMOS parts will do the job, although you are at the lower limit of the normal operating voltage range of the LT4320. \$\endgroup\$
    – ThreePhaseEel
    Apr 2, 2016 at 3:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.