1
\$\begingroup\$

We have a product in which we are experiencing failures due to rain water (presumably) penetrating the case and dripping onto the circuit board. I believe we have the source of the leak identified and fixed, however, I'd like to perform some thorough testing. I'd like to simulate a few different situations...

  • High ambient temperature (~120 F). The other extreme would be nice but I can't imagine an economical means to generate low temperatures.
  • I'd like to cycle between wet/dry such that it experiences a deluge and then a drying cycle. I want to see if the gaskets sealing the enclosure are affected by cycling.
  • I'd like to create a humid environment as well.

At the moment I envision a simple 4'w x 8'l x 8'h enclosure with a basin of water. A pump would circulate water from the basin up to the top of the enclosure where it would feed through a PVC pipe with aerators/sprinklers pointed directly at the devices under test. Just above that (so as not to get wet) a few 200W infrared heat lamps would be directed at the devices as well. The whole enclosure would be sealed with a 30mil poly.

My budget is roughly $400. Any suggestions or thoughts on how to get the most bang for my buck?

\$\endgroup\$
  • 1
    \$\begingroup\$ Is venting required in the case? What environmental level do you need for water spray? I would put more money into instrumenting the inside with humidity sensors and a fan with some sort of telemetry. 500W Halogen lights are $20 or less. Dry Ice used to be around 1$/lb from the nearest dairy producer. Bricks or crushed ice goes pretty quick in large space, so use a box or small container. I used a picnic thermal case. You can cool cycle 1lb bricks to a metal case but it may not spread evenly so keep dry ice wrapped in paper and then store in freezer between heat cycles. \$\endgroup\$ – Sunnyskyguy EE75 Oct 25 '12 at 22:21
  • \$\begingroup\$ If you like HALT HASS, this works and add vibration. Car wash facility might seem tacky but inexpensive if you can get the method approved. This doesn't have to be an operating env test. just non-op. environmental test. \$\endgroup\$ – Sunnyskyguy EE75 Oct 25 '12 at 22:22
  • 1
    \$\begingroup\$ -1 Please provide standard units, as is it is a nuisance to read for non-US people. \$\endgroup\$ – starblue Oct 27 '12 at 7:14
  • \$\begingroup\$ @starblue Minus one for that semi-significant value huh? How about 122cm x 244cm x 244cm? \$\endgroup\$ – Jason Oct 27 '12 at 18:40
  • \$\begingroup\$ One important way that water gets into a supposedly sealed enclosure is the suction / pressure difference caused by rapid temperature changes. The typical scenario is a cold rain shower on a hot afternoon. The product goes from 50 C dry, to 15 C, so it will suck in more than 10% of its volume. And at that time it's covered with water and the air is humid. So, humid/wet temperature cycling is essential. \$\endgroup\$ – tomnexus Oct 31 '17 at 4:36
5
\$\begingroup\$

TEDIOUSLY LONG (BUT PERHAPS VALUABLE) HELP

In light of the absence of dimensions for the device you need to test, I will attempt to offer some advice on options. (I designed a gas-tight computer case the size of carry-on personal luggage, so I'll use that as my reference.):

Unless you are going to use the testing rig numerous times, constructing a chamber solely for testing one prototype is extravagant. Nevertheless, if you are going to use the chamber for an innumerable number of test, or if the test chamber will be used on other occasions, the sturdy construction is paramount.

Considering your the limit of your budget--and contrasting that with the requirements--one of the most inexpensive materials to build such a test chamber of those dimensions is concrete. To economize, you could divide the area of the walls equally and construct a frame in which to pour the concrete and build the chamber in sections (thus eliminating the cost of building a large frame for a single pour). If you use concrete with fiber reinforcement, you can forgo using rebar (provided the walls are of appropriate thickness.

Of course, using cement is labor-intensive and requires a foundation able to support such weight.

The option I would employ is to use a pipe as a testing chamber. Though the common consumer is not going to see this at their local hardware store, it is not impossible to find pipes constructed from ABS or other polymers or composites with diameters greater than two (2) feet in diameter.

There are several methods you can use to seal the pipe from one end (and serve as the 'floor').

If available, purchase a cap to seal the pipe from the same manufacturer (or make), so as to avoid considerable labor. Nonetheless, sealing the pipe with such a cap (usually a bolt-on flat disc for pipes in this size) will still require that the pipe be placed on a surface able to support the weight of the water and any other contents--preferrably a ground-level concrete floor. (It is important to note that most caps are NOT flat, but curve slightly outward. Should this be the case, you may instead place the pipe on sand or soil. CAUTION!: I believe it is crucial I note that not securing a pipe of large dimensions to a secure foundation or adequate frame can prove dangerous [if not reckless and potentially harmful], given that should such a structure fall to one side, the weight of the contents alone are enough to exact considerable damage to surrounding property or persons.)

Absent the option of purchasing a proper cap, you may construct a base with concrete. To facilitate the task of drilling into concrete (to secure the inserts into which bolts or custom cut threaded rod will be inserted), I suggest creating a template of the face of the pipe from plywood. Precisely map the position of the holes where the inserts will be placed, and drill and secure wood dowels of the size and depth required for the implantation of said inserts, allowing for appropriate depth of the threaded rods afterwards (the dimensions should be noted in the instructions provided with the box of inserts used in masonry).

After the concrete has cured--prior to removing the frame--(with the appropriate size bit), drill in the position of each dowel to remove the wood, assuring that only the concrete remains. Remove the wooden frame from the cured concrete. Place the pipe over the base to confirm the position of the inserts. (Make corrections as deemed necessary [if you measured precisely, the holes in the concrete base should match the holes on the face of the pipe exactly].) Wash/Brush and properly dry the concrete base (and holes). (One trick to getting the water out of the holes--if the holes are too deep to properly dry with cloth--is to blow air into them using the output of a vacuum cleaner [with appropriately sized adapter], or an air compressor, after which you can allow the holes to air-dry.)

You will need epoxy to secure the metal inserts to the concrete base. Following safety precautions and time envelope according to instructions, apply epoxy and place inserts in holes. Allow the epoxy the most generous time to cure in reference to instructions.

Once epoxy has properly cured, secure the threaded rods into the insert (that is, if you are using threaded rod; you may also find ready-cut screws threaded on both ends). (On the side of caution, I will mention the obvious: You will need nuts, flat washers, and split washers [to go atop of the flat washers] to secure the pipe to the concrete floor. Another obvious note: since you will be using water, it is recommended that all metal parts be made of stainless steel). (Alternately, you may use screws to secure the pipe from atop, once the pipe is positioned atop the sealant and concrete base [as noted in the pursuant paragraph]. [Note that by not using threaded rod you loose the advantage of aligning the pipe with ease and increasing the potential of making a mess of the silicone sealant.])

You will need silicone sealant (industrial grade) graded to seal concrete and plastics (found in most hardware stores). Apply the sealant generously around the contact area where the pipe and concrete base will meet. (Read instructions on the sealant container to assess if a drying period is necessary, and follow instructions accordingly.)

Once silicone is applied and ready, properly align and position the pipe atop the concrete base (preferably with the help of someone).

Prior to securing the pipe to the concrete base, ensure that the pipe is indeed resting properly on the base. Place the flat washer atop the flat pipe surface. Place the split washers atop the flat washers.

To secure the pipe in a balanced manner, tighten each bolt according to manufacturer recommended pattern. (If you do not have that information from the manufacturer, looking up the product number of the pipe may yield the desired documentation on the internet. Failing that, it will be sufficient to tighten each bolt on the opposite side of the pipe from the last bolt tightened [in a zig-zag pattern around the pipe]. This will ensure that the silicone spreads evenly as nuts are tightened.

Though it may not be crucial for your application, I recommend using a wrench able to release at a set pressure. This will ensure that all nuts are tightened to the exact spec (and reduce the possibility of a leak due to imbalanced tightening). (Personally, I don't recall ever having a leak problem by judging tightening pressure by my own cognisance.)

Provided curing time (if any) for the silicone is followed accordingly, your test chamber is now complete.

P.S. I just remember you also noted you need to test to 120 degrees F. I am not certain of the melting point of PVC or Polyethylene, but it either may be able to persist in said temperature without damage for prolonged periods of time (provided the temperature is accurately sustained). (I am assuming you do not need the chamber to be completely enclosed, in which case your most economical method for creating a sustained environment of said temperature may be to purchase or borrow a propane heating element [like those used in labs], or a manually operated hear, or something similar, in combination with a thermometer [of course].)

P.P.S. IF you need the chamber to be enclosed at all times, then your best course of action is to build the chamber out of concrete, or compacted earth, or rammed earth, or adobe, with a metal place incorporated into a suspened floor foundation, under which you would position the requisite heating element/source.

P.P.P.S. I admit I went into excruciating detail partially due to the fact that I may have to build something like this some time in the future (should I decide to pursue building MIL-SPEC electronics).

P.P.P.P.S. ABS explodes violently when thermal limits is exceeded (i.e., exposed to fire) (as opposed to melting like other polymers). Please query thermal capacity of the polymers used in any plastic pipe you may consider.

\$\endgroup\$
  • \$\begingroup\$ +1. By "wrench able to release at a set pressure" did you mean torque wrench? \$\endgroup\$ – Nick Alexeev Oct 26 '12 at 0:05
  • \$\begingroup\$ Duh. Thank you, Nick! (I just came down with yet another instance of flu just a few hours ago--those brain cells have evidently atrophied.) \$\endgroup\$ – Ivan Oct 26 '12 at 0:14
  • 1
    \$\begingroup\$ Are you an electrical engineer, or a civil engineer? :-) All of this sounds like way overkill for simple splash/heat testing! \$\endgroup\$ – Dave Tweed Oct 26 '12 at 0:43
  • \$\begingroup\$ For certain markets, product testing requires adherence to testing standards established to ensure minimum guarantee measurable performance targets/specifications. The knowledge I've accrued comes from readign mountains of data and some design and engineering I've put into practice. In conversations with formally educated engineers, I'm often asked where I studied. XD \$\endgroup\$ – Ivan Oct 26 '12 at 0:50
  • \$\begingroup\$ To put it into proper perspective, if you design a product you wish to introduce to market--say, the military--depending on the target environment, your product will have to pass the requisite standards (preferably PRIOR to sending to a lab for certification). If you do not do become knowledgeable of testing standards and perform your own testing, you can doom the product (not to mention any chance of actually securing a contract). \$\endgroup\$ – Ivan Oct 26 '12 at 0:59
4
\$\begingroup\$

We just used an old fridge for our environmental chamber. We were only going after humid/heat (being in Canada we could throw the unit outside on a good cold winter night and get -20C without much trouble).

The fridge did stink, but it worked well enough for us to get the testing done.

edit to describe process

There wasn't much to modifying the fridge: we stuck a small ceramic heater in it and set it to low (which was still too high for the small insulated space inside of a fridge. The heater had a fan which did a fine job of stirring the air in the fridge for us. For humidity we just had a big pot of water. We'd cut a hole and covered it with a piece of lexan, sealed against the fridge door with RTV; this allowed us to see the display of the unit under test as well as the temp/humi sensor.

If I was doing it over again I would use a temp/humi sensor module and interface it with an Arduino so I could do data logging more easily. I would probably also replace the pot of water with one of those garden ultrasonic misters to help rapidly change the humidity. Another hole cut in the side of the fridge connected to a generic HVAC damper and small fan would allow me to exchange the air in the chamber and perhaps do more rapid cycling. In fact the Arduino could control the solenoid feeding the water and power for the mister as well, allowing me to do fully automated, unattended heat/humidity cycling.

\$\endgroup\$
  • \$\begingroup\$ I am about to do a research and I am using fridge as an environmental chamber. Can you tell me a bit more details how did you simulate humidity and heat in the fridge? Thanks \$\endgroup\$ – Soldo Oct 30 '17 at 1:22
  • \$\begingroup\$ @Soldo I've edited the answer to describe the process. \$\endgroup\$ – akohlsmith Oct 31 '17 at 0:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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