Temperature/Humidity Controlled Enclosure

[wddian]

Hi,

I'm not sure if this is the correct forum to ask about this, but here it goes.

For a project that I am currently working on, we have some very sensitive electronics that need to be kept at around 20-35 degrees C and approximately 40% RH (non-condensing) to operate efficiently. I do know that these kind of environmental chambers can be purchased for $15,000. But does anyone know where I can get a hold of temperature control/humidity control components? Or how do I go about designing the system?

Thanks in advance.

 

[itsmoked]

Not clear on your needs as you have provided far to little description. Is this like lab instruments someone has to handle while they stay in this environment?

What is the natural humidity of the area this unit will be in? If normal humidity is 50% or less you can often do it far less expensively than otherwise.

Otherwise you would be using cyclical air dryers. These are systems you plumb plant air to and they leak air into the space you need dry until some time limit, then they switch valving and the other half does this while the first is dried out.

Depending on the heat load and the expected ambient you may need a refrigeration unit and a heater to meet your temperature requirements.

If you know what you're doing and the local environment, you may be able to use the door mounted air conditioning unit in concert with the heater to easily maintain your humidity, as air conditioning is a drying process.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[wddian]

Thank you for the reply itsmoked.

I'm sorry for the lack of information in my original post. The chamber will need to be approximately 10"H x 20"W x 25"D. This will house our electronics and leave sufficient space for air flow. The electronics weigh ~55 lbs and require 90-264 Vrms 47-63 Hz, 5.3 A AC power supply. The environmental chamber will require feed-through of ~ 1/2" in diameter for it's umbilicals and cables.
The electronics require 20-35 degrees C and 20-80% RH noncondensing. The unit will be running constantly in a saw mill so particulate will need to be kept out of the housing chamber. I am not 100% sure on what the natural temperature and humidity are at a typical saw mill, but I am assuming the conditions should be within the requirements for the electronics.

 

[itsmoked]

Well, you could spend years on this subject as it is very wide ranging. A box, who'd-a-thunk.

That is a strange shape indeed.

Saw dust hmmm? Suspended particulates really complicate things further. Building something electrical that will depend on maintenance of air filters is a bad idea. Most electronics are higher temp than what you're considering. 45 ~ 50C is much more common for industrial settings. Too bad.

Is this a one-off or a product you will build many of?
If it's a one-off can you say what state/country it will be in as this will tell me what environmentals are like.

I'm assuming 120VAC @ 5.3A. This means a heat load of 640W!! This is a lot to get out of a box that size. Tell me some of this is exported out of your box to the field. Perhaps running a motor or ?

The biggest factor here is the amount of cooling needed to keep your painfully low temp. Small differences can make large price and complexity differences. Faced with this task you should actually run the system with a watt meter to find the actual heat load and not rely on maximums reported name plates. The name plate numbers are used for sizing wiring and fuses not necessarily average power consumption. For example 640 watts in a box that size would mean the top half of the box would reach about 50C with the box in a 22C environment.

These are typical of the many variations available:

http://www.thermal-edge.com/aircon/

I caution you again about dust and what issues it might present to these types of devices and how much "heat load" impacts the possibilities.

You can google for "electronics cooling enclosures".

Keith Cress
kcress -

http://www.flaminsystems.com

 

[IRstuff]

The only effective way of dealing with that much power in that small an enclosure that must be fully sealed is liquid cooling, with internal circulation fans that have no access to the outside. You can get what used to be a Neslab cooler unit and have it deal with the external environment, while your box remains essentially hermetic.

As a side note, something that is intended to be in the field should be designed to be more robust than what this seems to be intended for.

TTFN

FAQ731-376

 

[wddian]

Thanks to all your input.

Yes I was quite concerned with the particulate issue. Maybe a fully sealed enclosure with liquid cooling like IRstuff suggested would be a more suitable solution.
Now the dimensions of the enclosure is definitely subject to change. I am aware of the amount of heat load the equipment is producing. There is an integrated fan at the back of this unit (fan Size 4 11/16 " roughly).
The enclosure is a one-off prototype. It's sole purpose is to accomodate the necessary electronics for testing at a saw mill located in BC Canada. Like I've stated earlier, I don't know what the conditions would be like at the saw mill, but I would assume it'll be around the 15-25 degrees C and 70-80% RH range.
I was actually thinking of using a NEMA 12 cabinet or enclosure with an AC unit.

 

[waross]

A couple of years ago we were building a substation on BC. The temp got over 100F most days for over a month. Some days it hit 110F. Check the weather for Oroville Washington. We were usually hotter.
I have worked in mills at minus 30F. I have seen just about every combination of temperature and humidity imagiable and a few unimaginable combinations. BC is bigger than Texas. It stretches from the Pacific ocean to the Rocky mountains and from Alaska to Montana (check Hwy 93).
Can you be a little more specific?
If this is going into the interior, I would consider buying a cooler with a solid state cooling unit and pirating the cooling unit. Bolt it to the back of the box and cover everywhere else with styrofoam. If it is near the coast, is dry instrument air available. A very slow purge with dry air as Keith suggested should be quite cost effective.

I have seen those pancake fans with filters in a few mills. The interior of the panel is usually very contaminated with a layer of very fine wood dust. I have moved more than one control transformer out of the panel so that the fan was no longer needed and the ventilation hole could be sealed.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

waross; 600+ watts makes a mockery of solid state coolers.

Really, this is a complex problem that I would expect to take at least a week of engineering to do correctly. It would correctly use a site survey to determine what would work best with the facility and staff. Is cooling water present or reliable plant air, would be important considerations. It would also include time to axe that 600W thru either finding the actual power dissipation measurements, or changing from an over sized, office temperature PC to something better suited. (if it includes a PC)

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

I don't mean to be a wet blanket here, but that is going to be a near impossible task. I have done a lot of work on electronics in sawmills; maintaining that narrow of an environmental spec is going to cost you more than the equipment is likely worth, and keeping it running for any appreciable length of time is not likely. Two factors will come into play:

1) Any enclosure that is going to keep out sawdust is going to need to be NEMA 12 or better. That means no air exchange whatsoever. A sealed box then is going to require air conditioning, such as a Hoffman / McLean industrial AC unit. The trouble with that is, they have an outside air circulation system for the evap coils and those have filters on them to prevent contaminant loading of the coils. In a sawmill, NOBODY will clean those coils after the first month of service and in 6 months the AC unit will overheat and die. It's just an inevitability.

2) Humidity will be a problem if you are looking at the arid eastern BC (Kelowna, Okanogan, Penticton etc.), and winter temperature will be a problem on the west (wet) side and north. East side will have days in the summer where the ambient RH will be as low as 20% already. Those will also be days where the temperature will be highest, so the AC will be running full blast, dehumidifying the inside even more. I have seen static destroy sensitive electronics in areas like that almost as much as heat. Remember, you can de-humidify electronics, but you cannot humidify them if they are too dry. In the west and north, your enemy will be extremely low temperatures in winter. They don't shut the mills down up there, they like to think of themselves as a hardy bunch. But when you run the AC to dehumidify, you will need a big heater to keep from freezing the insides as well. The balancing act can be very tricky to maintain. I have seen several systems fail to keep it consistent enough to survive for long.

My suggestion would be to rethink your electronics design. Most industrial electronics based systems need to be able to tolerate 40C as a minimum upper range, 50C is better and 60C is the gold standard. You lower range also needs to be capable of 0C operation and -10C storage, otherwise you will have a big problem when they power up after a power failure. RH specs tend to be 10 - 95% so as to tolerate being operated in air conditioned enclosures, but if you can get the temp tolerance higher, it also allows the use of heat exchangers that are not as prone to failure as AC systems in sawdust environments (but do not de-humidify). Conformal coat your boards to survive the humidity, heat sink all of your sub-assemblies and high power components to increase the temperature tolerance. Also, make sure you have a stout power supply section (or a good external PSU), because the line power has a lot of nasties in sawmills.

Good luck.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[burnt2x]

"1) Any enclosure that is going to keep out sawdust is going to need to be NEMA 12 or better. That means no air exchange whatsoever. A sealed box then is going to require air conditioning, such as a Hoffman / McLean industrial AC unit. The trouble with that is, they have an outside air circulation system for the evap coils and those have filters on them to prevent contaminant loading of the coils. In a sawmill, NOBODY will clean those coils after the first month of service and in 6 months the AC unit will overheat and die. It's just an inevitability."

You can start on that idea! A sealed enclosure that does not allow outside air; evaporator inside the enclosure and the condenser outside where it is easy for the crew to service it if it gets too dusty.
I am looking at my grandpa's refrigerator and the split-type HVAC!
My 2¢

 

[waross]

Hi Keith;
I should have mentioned, find a robust power supply and mount it separately, outside the enclosure. Then re-evaluate the cooling load.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

Yeah Bill, I suspect this is just a PC as those are typical PC operating requirements. So you would need a PC external supply which is not typically available.

There are industrial PCs available that would handily take these conditions.(heck I was on the design team for one) The one I helped design had no fans.

I think these guys sell it now:

http://www.vartechsystems.com/products/nema-4x-monitor-computer-workstation.asp

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Another option may be an industrial PLC and a data link.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[KiwiMace]

I don't get it. You say: the electronics require 20-35 degrees C and 20-80% RH noncondensing. This is like hitting the side of a barn for most aircon units. Right-size the ac unit for an insulated box and you'll get effective coarse humidity control as a default. Sounds like a heater is also required.

If i've missed something and you need a non-varying T&RH within this range, then that is something else altogether.

Check for dust explosion classification of where you are sticking the HVAC unit. It may need to go outside the building.

 

[itsmoked]

KiwiMace; You are missing that this is a saw mill!

Air conditioning evaporator coils and fins, do not co-habit in functional terms, with airborne wet, sticky, sap laden, multi-length saw dust fibers.

The McLean units arguably the best would require the model that is 19" taller than the OP's box. It would also draw 13.8A at 120V.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[KiwiMace]

I didn't disagree about the need for protection and filtration, that was someone else's value add. The aircon unit can be ducted from outdoors or the roof.

What I was getting at was that the criteria should be very easy to achieve. You could tape a popsicle to a desk fan and get within that range.

 

[wddian]

Thanks again for everybody's feedback.

The electrics described cannot be re-designed to match more closely to typical industrial standards. The environmenal requirements are set in stone, 20-35 C and 20-80% RH noncondensing. The only flexibility I have is in the design of the enclosure. I do realize this can be a very complex problem and coupled with my liimited knowledge in the particulars, which is why I greatly appreciate everybody's input.
The saw mill is at 100 Mile House, which is in the South Cariboo region of Central BC. The conditions there are typically ~20 C during the summer (very dry and evenings are cool), ~-15 C during the winter (starting in November and ends in March).

 

[waross]

That's global warming for you.
I spent a year at a mill about 60 miles North West of 100 Mile House a long time ago. -30 C used to be common.
If your enclosure is opened at -15 C on a dry day and the air changes out, you may end up way below 20%RH when the interior of the enclosure warms up again.
Is there any way you can mount your equipment in an office building or similar and link to the field?
Otherwise you may have to bite the bullet and settle for an expensive solution.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[davidbeach]

I'm curious why someone would put together environmental requirements for a piece of equipment that is totally divorced from the reality of the installation? Maybe I'm just spoiled by working with protective relays, but it would seem that any industrial electronics ought to work over the same temperature range as the relays do, -40C to +70C (displays often don't work below -20C). +20C to +35C sounds like something that was never intended for any real world use, just in somebody's lab somewhere.

 

[IRstuff]

If the designer gets to set the environmentals, then that's what you wind up with, which is something that hasn't any temperature compensation or correction, because it was easier for the (lazy) designer to not put that in.

TTFN

FAQ731-376