Question about plexiglass ducting and static electricity

revanrules

Junior Member
Aug 21, 2007
23
0
0
If I build ducts for the inside of my case out of plexiglass, should i worry about a static charge building up? Or if the duct is attached to the chassis, will it be grounded? It makes sense that it would be grounded and I should have no reason to worry, but what do you guys think?
 

BonzaiDuck

Lifer
Jun 30, 2004
16,341
1,891
126
Bonzai-Duck-tster here . . . .

As he said -- no problem. Foam-art-board is also similarly inert, easier to work with, but not as pretty.

I had to build some rather complex ducts for my system, and the foam-board was intended just for proto-typing. Even so, I put the project on the back-burner for using Lexan, and painted the foam-board in primer and silver enamel for the time being.

Also -- either plexi-, acrylic, or Lexan will work on this, but Lexan is more durable -- and possibly more scratch resistant. It may be that Plexiglas or acrylic is cheaper. You tell me.
 

BonzaiDuck

Lifer
Jun 30, 2004
16,341
1,891
126
Yesterday's post for EEYore appears in this thread:

Discussion on ducting as it may pertain to the Ultima 90

There are more pictures, with [as Arlo Guthrie's song once said] "with circles and arrows on the back of each one." I'm not all goosey and narcissistic about how good my ducts appear, or even this (very bad) photo of my case mod itself. For instance, foam-board can become slightly warped if left in storage of various temperatures and moistures for some time, and you may notice that one panel of my upper duct is slightly warped. But -- it works. You can do better than I with time and attention.

The inspiration for me was a single post here at Anandtech around 2004 by someone who had read this article at OverClockers.com:

John Cinnamon's "Cheap as Free Air Cooling," OverClockers, October, 2004

The basic principles:

1) The idea by itself that putting two fans of the same size or throughput on a single cooler (serial deployment) helps very much - - is a myth. But it is not a myth of the intake fan is larger than the exhaust fan, or if serial deployment uses restricted passages and pressure from higher CFM intake fans to feed air to additional fans, and this also holds true for the passage between a cooler fan and case exhaust.

2) It is important to find ways to restrict the volume of air circulation over targeted components which also increases air speed and pressure around those components. "Leaky" ducts also serve a purpose, if for instance air intake for the CPU is forced down on a motherboard, while the main case chamber's higher pressure is also forcing air through the leaks around the motherboard.

3) The best strategy is to target air at specific hot components and then force the air quickly from the case chassis. Blocking off other sources of leakage in the chassis itself -- excluding the deliberate placement of intake fans -- will marginally improve the results.

4) High efficiency CPU coolers which are already deployed with fans directing cooler-fin throughput to an exhaust draft may show the least improvement, but some improvement will be obvious nevertheless. . . .

5) As always (if you check Cinnamon's remarks), the essential ingredient is a duct or panel that blocks off the motherboard from the main case chamber, allowing air across the motherboard in restricted volume and high speed, for quick exhaust. Some boards are outfitted with high-efficiency heatpipe "necklace" coolers, but these contraptions get the least attention by enthusiasts -- even if the motherboard comes with special fans for parts of those heatsink assemblies. Further, motherboard components are often missed by water-cooling advocates, who must make a tradeoff between simplicity and additional waterblocks on the Northbridge chipset, but attention to Mosfets and other components (to my knowledge) has no water-cooled solution. Mostly, those components have an air-cooling solution, and therefore ducting would help even for a water-rig.

MY OPINION: As you design your ducting solution, it is a good idea to carefully calculate an intake-exhaust imbalance inside the case by virtue of rated CFM. The higher the imbalance, the better, and attention needs to be made to the adverse effects of intake filtering. But one should probably use these considerations as an opportunity to pursue a program of reducing the total number of fans, or deploying them with the most strategic efficiency. This also balances with an equal consideration that more fans at low rpms reduce overall noise level, and that the size of a computer case limits one's opportunities for additional fans more severely.

HOLISTIC APPROACH: Look at all the heat-generating components in relation to each other first. This is almost like a board-game (pun! :D ) -- you can get double, triple and quadruple "whammies" by simplifying and channeling the air-flow across several components, using a limited number (or reduced number) of fans for the greatest overall effect. You will also want to anticipate three or four key issues: the simplicity of the design, the design's relationship to ease of removal and replacement, how you might need to build separate duct components that fit together, and a means of securing the assembly to the inner chassis or even CPU coolers, VGA cards or other parts. Often, with light materials, a precision "snug-fit" is enough to resolve that latter problem. I suggest measuring in centimeters and millimeters, and getting a reasonably precise metal metric ruler or a couple of them in various sizes.

You can make ducts from foam art-board, or you can use Lexan, acrylic or Plexiglass, and you can make hybrids of art-board and Lexan. Lexan will not assist much in mitigating noise, but it looks "prettier." It also takes more time and effort and must be cut precisely, and one must pay more attention to how edges of panels fit together. It can be bent or shaped with a heat-gun, but there is a danger that the bends or shaping results will not be accurate enough for other fitting considerations. For straight bends, you might try scoring the panel first and applying the heatgun as precisely as possible so that the bend has a robust edge and is reasonably precise. So it is both easier (for precision) and harder (for the work) to use separate flat panels and numerous joints to fit together.

You can enamel either the art-board or clear plastics, but if you enamel clear plastics, make sure to purchase a can of primer-paint for plastics and apply it with attention to sanding and prepping the surface. Paint will melt the edges of exposed foam if applied too much at once, so the rule is "thin coats."

As for glue, Poly-Zap for Lexan or some other glue that welds Lexan, acrylic or plexi -- is the only real choice, and as a matter of permanence. For foam-board, some glues that "work" nevertheless will melt foam edges, and you need to assure that they set quickly. The best glue is also the most expensive: "Zapo Foam-Safe" from the maker of Poly-Zap. For double the price, you can also get a fixative the promotes "instant setting," but I've found it unnecessary -- completely unnecessary.

Every glue-joint of reasonable quality in foam-board seems to promote the rigidity of the structure you are building. The material can also be "finished," patched or even joined with fiberglass auto-body plastic.