4
\$\begingroup\$

I am producing a system with aluminium enclosures all connected to a central unit which is connected to Earth. All the enclosures of the system have their chassis ground connected together via cable shielding. The system will be used in academic and industrial environments. For ecological purposes and to increase recyclability, I would like to avoid using paint and instead laser mark the sheet metals with the company's logo and the connector names above the connector slots.

Paintless (raw) enclosures seem rare, are they bad practice? If so, why? Electrical safety? EMC?

\$\endgroup\$
13
  • 4
    \$\begingroup\$ Most of the aluminum we use for chassis' is anodized. This process provides a much harder and durable finish than raw aluminum. One shortcoming is the anodizing creates a somewhat insulating surface, and needs to be removed if good electrical conductivity is needed. \$\endgroup\$
    – SteveSh
    Feb 18 at 17:59
  • \$\begingroup\$ Plain aluminum doesn't look appealing. It doesn't oxidize evenly, and it scratches easily. Hammertone finishes used to be popular. Anodized would be my choice. \$\endgroup\$
    – Mattman944
    Feb 18 at 23:48
  • \$\begingroup\$ Is stainless steel an option? It has worse electrical and heat transferring properties, but a whole lot better look and durability. It can also get cheaper overall (well, this is rare) \$\endgroup\$
    – fraxinus
    Feb 19 at 8:21
  • \$\begingroup\$ @SteveSh OTOH unfinished diecast aluminium boxes are common (RS). It's not the best look, admittedly; the bare aluminium mechanical parts I get made tend to look better (anodising isn't an option here, or at least not a financially viable one as we'd be sending one-offs to an external contractor) \$\endgroup\$
    – Chris H
    Feb 19 at 9:36
  • 1
    \$\begingroup\$ @peter - Again, no. Maybe we're mixing up returns (for power) and chassis connections. Power distribution is usually done as you described - star from the primary power source. Chassis, on the other hand, are many times connected together through structural members, with very low (mohm) resistance connections. That structural path should not, in a properly designed system, carry any currents except for perhaps small amounts of leakage current. \$\endgroup\$
    – SteveSh
    Feb 22 at 19:29

5 Answers 5

7
\$\begingroup\$

Aluminum is soft; a bare finish is not very durable, prone to scratching, scuffing, etc. A bright metal finish is attractive when new, but a dingy surface quickly gets unappealing, and a printed surface could quickly become illegible.

The contrast between etched and bright metal also might not be very good; on the upside, a little paint could be used in the etching, retained by the depth below normal surface. Possibly, a "black" treatment could be applied, whether by etching in the first place (optically rough surface), or by electrolytic treatment (nano-nickel something or other? I forget what all processes are available).

Anodize is the most common bare-metal treatment; the resulting oxide layer is hard, moderately tough (depending on grade), and being porous, it can be dyed in any color. The process is environmentally friendly (mainly sulfuric acid and electricity).

Note that anodize results in at least a somewhat frosty or matte appearance, even if starting from polished metal; it probably wouldn't be a good choice where etching alone is used to render graphics.

There might also be environmentally-friendly (carbon neutral? naturally sourced materials?) paints available; I don't know offhand I'm afraid.

Paint or anodize should be masked or ground away where connections are made: around bulkhead connectors, ground lugs, EMI spring strip, etc. Probably, star washers are sharp enough to dig through most anodizing (maybe not hard anodize), but most EMI spring stock probably isn't.

\$\endgroup\$
2
  • 4
    \$\begingroup\$ After anodising with a dark dye, laser-etching provides very good contrast. Engraving should show up[ just as well, but it's an extra machining step \$\endgroup\$
    – Chris H
    Feb 19 at 8:52
  • 2
    \$\begingroup\$ FWIW, a good looking matte finish is a result of pre-bath surface finishing step(s), as much as the anodize itself \$\endgroup\$
    – Pete W
    Feb 19 at 15:49
9
\$\begingroup\$

Aluminum is a somewhat active metal, and all finish options reduce that to varying degrees. Whether unfinished aluminum is a bad idea or not, depends on the environment and application, but a manufacturer of general purpose enclosures can't know that in advance. That's why you're not going to see enclosures for general use without a finish.

re: surface oxidation / conductivity, and grounding

A pretty common finish, which maintains surface conductivity, in contrast to anodize or paints, is chromate conversion coating (aka iridite / alodine / chem film). It's cheaper than anodize, too.

Another very common practice for conductivity for grounding lines is using star washers, with one-time-use, that bite through any surface oxidation and most coatings.

\$\endgroup\$
6
  • 2
    \$\begingroup\$ Some applications specifically prohibit the use of things like star washers because of the possibility of FOD creation. \$\endgroup\$
    – SteveSh
    Feb 18 at 18:56
  • 1
    \$\begingroup\$ Press-in fasteners are also popular, giving a solid metal-on-metal connection; combined with a star washer, have seen ground connections made this way many times. \$\endgroup\$ Feb 18 at 18:57
  • \$\begingroup\$ @SteveSh - good point, updated the wording to not say 'universal' \$\endgroup\$
    – Pete W
    Feb 18 at 19:23
  • 1
    \$\begingroup\$ Chem film is also a good idea for EMC reasons. If delivering the product to Europe or other jurisdictions with RoHS-like restrictions, be sure to specify a hex-chrome-free coating. \$\endgroup\$
    – The Photon
    Feb 19 at 16:25
  • \$\begingroup\$ @SteveSh "FOD"? \$\endgroup\$
    – Yakk
    Feb 20 at 14:36
6
\$\begingroup\$

Paintless (raw) enclosures seem rare, are they bad practice?

Paintless does not imply raw, untreated aluminum. Yes, you can sure buy untreated aluminum enclosures, but they are not a finished product and must have some surface treatment applied before use. They look nasty and only turn nastier over time. Bad idea, very unprofessional impression.

If painting is "unecological"*, then anodizing is the other low-resource-use option. All it does is add some oxide to the surface, and optionally some mineral pigment that poses no recyclability concerns.

Anodizing is an "interesting case". It has split personality:

  • it is considered a conductive surface for the purpose of protection against electric shock, but

  • it is considered a non-conductive surface for the purpose of attaching PE conductors

In other words, for every EMC and PE ground connection, you'll need to either grind off the anodized layer, or, more properly, mask off the areas that need to contact other conductive parts prior to the anodizing bath. The former approach is OK for early prototyping. The latter approach is how you'd be doing production:

  1. Have raw aluminum enclosures made,
  2. Mask them,
  3. Anodize them.

Steps 2 and 3 are done by metals finishing shops, per your drawing, or by the custom (or customized) enclosure vendor who provides step 1 too.

*How many kilograms of paint do you think you will use over the lifetime of the project? Now look at how many resources, amortized, are needed to support your life, on average, for one hour. There's an awful lot of paint that is the resource-equivalent of a day's worth of engineering time. Don't throw the baby out with the bathwater. Literally a few hours of you considering this "optimization" at work will use up more resources than you're trying to save.

To further illustrate the point of dubious savings, consider this excerpt from Joel Spolsky's essay "A Field Guide to Developers".

Let me, for a moment, talk about the famous Aeron chair, made by Herman Miller. They cost about $900. This is about $800 more than a cheap office chair from OfficeDepot or Staples.

They are much more comfortable than cheap chairs. [...] The ergonomics, especially of the newer models with lumbar support, are excellent.

They last longer than cheap chairs. We’ve been in business for six years and every Aeron is literally in mint condition: I challenge anyone to see the difference between the chairs we bought in 2000 and the chairs we bought three months ago. They easily last for ten years. The cheap chairs literally start falling apart after a matter of months. You’ll need at least four $100 chairs to last as long as an Aeron.

So the bottom line is that an Aeron only really costs $500 more over ten years, or $50 a year. One dollar per week per programmer.

A nice roll of toilet paper runs about a buck. Your programmers are probably using about one roll a week, each.

So upgrading them to an Aeron chair literally costs the same amount as you’re spending on their toilet paper, and I assure you that if you tried to bring up toilet paper in the budget committee you would be sternly told not to mess around, there were important things to discuss.

\$\endgroup\$
5
  • \$\begingroup\$ Thank you for your reply. Would step 4 be painting the connector names above the slots? Does paint hold well over anodization? \$\endgroup\$
    – Tommy95
    Feb 19 at 18:55
  • \$\begingroup\$ Or would laser etching after anodization be the way to go? \$\endgroup\$
    – Tommy95
    Feb 19 at 19:03
  • 1
    \$\begingroup\$ Please don't lecture people about how the environment doesn't really matter that much. Every bit matters - we got it to the state it's in now with only a little damage per person, times eight billion people. Anodization sounds pretty good. \$\endgroup\$
    – user253751
    Feb 20 at 0:26
  • \$\begingroup\$ Environment does matter, and that's precisely his point. It actually matters enough that it's worth the effort of considering globally the impact of our activities, not doing some minor-effort negligible-impact feel-good little-bit (not that I'm suggesting the OP had in mind either). 8bn times very little does not matter, it's far from enough. OTOH, one should start by estimating one's impact and work on it : if OP's impact, all things considered, is dominated by his enclosure paint use, then by all means he should reduce it. \$\endgroup\$
    – Nicolas D
    Feb 20 at 13:07
  • \$\begingroup\$ It's about trying to keep designs to their essentials and not neglecting the small potential improvements just because they only represents 1% of the total impact. The other 99% are presumably already optimized. \$\endgroup\$
    – Tommy95
    Feb 20 at 18:41
5
\$\begingroup\$

Not a good idea. Bare sheet aluminum has a grain to it, and stuff gets down into the grain - fingerprints, oil and grease, etc. Bad for cosmetics.

The real reason is electrical. Bare aluminum oxidizes. This can affect the quality of the earth ground connection to the case. Another issue is the way aluminum reacts when in contact with other metals, such as copper. Aluminum is used in both residential and industrial power wiring, and all connections with other things are dipped in an anti-corrosive gel.

An alternative to paint is a pre-paint surface treatment called chromating. This is an acid etch bath that leaves behind a surface coating that is more resistant to fingerprints and corrosion. There are two main chemistries, hexavalent and trivalent. Hexavalent is a better coating, but frivalent is less toxic and more eco-friendly. We had two baths, one for clean-chromage and one for a gold-tone finish.

https://en.wikipedia.org/wiki/Chromate_conversion_coating#Aluminium_and_its_alloys

\$\endgroup\$
4
  • 2
    \$\begingroup\$ But "the quality of the earth ground connection to the case" will be a concern whether painted or not. Aluminum is a popular material, and is often grounded. Can you explain this discrepancy? \$\endgroup\$ Feb 18 at 18:45
  • 2
    \$\begingroup\$ bare Aluminum will develop a few nm oxide before passivating only. because the metal itself is soft, this is really easily breached, which makes bare Al virtually retain a conductive surface for engineering purposes. \$\endgroup\$
    – tobalt
    Feb 18 at 20:38
  • \$\begingroup\$ @tobalt: Fasten a couple pieces of aluminum foil to a flat surface, hook up a continuity tester, and place a ball of crumpled aluminum foil so it bridges them. If the continuity tester uses a low voltage, the foil will typically read as a very good insulator unless or until a higher voltage is placed between the two flat electrodes. That higher voltage need not have any significant energy behind it, though: even a piezoelectric stove lighter operated near the apparatus will often suffice. Clapping near the apparatus will then reset it to non-conductive. \$\endgroup\$
    – supercat
    Feb 19 at 16:35
  • 1
    \$\begingroup\$ @supercat "Al foil" especially the kitchen commodity type, is sometimes coated and essentially always annealed leading to thicker thermal oxide. If you take a piece of Al hardware and saw it (to obtain an actual bare metal surface), you will have a really hard time ever reading anything other than a short on it, even with ever so minute force, and even after very long exposure to air \$\endgroup\$
    – tobalt
    Feb 19 at 18:01
4
\$\begingroup\$

A paintjob (or powder coating, anodization, ...) makes EMC more difficult if anything, because cable shields can make much better circumferential chassis contact on a bare metal wall.

On the other hand, paint is usually not regarded a reliable insulation layer either, so cannot be counted to provide safety by double insulation. That means that chassis earthing is mandatory anyway, if your devices are mains powered.

The reason for paint is probably that a bare metal chassis could be regarded as cheap-looking.

\$\endgroup\$

Your Answer

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

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