Climate control systems

Hi Everyone,

I'm interested in any experience with climate control systems like those that would be used in a high end metrology room. My space is going to be e.g. 25x25 with a 10 or 12' ceiling.. There are some items in that room generating some heat, and only occasional traffic in and out.

any idea what I'm looking at pricewise for fairly precise systems. the real meaning of "precise" is heavily biased to cost feasibility...

Thanks in advance!

Ron

In a recently-previous life, my company had the experience of building a new building, around the 1998-1999 timeframe. The climate control systems that were being installed at the time were typically computer-monitored and programmed systems, using wall-mounted thermostats, with a sensitivity of 1 degree F. Most of the controlled spaces we built in to the building (for manufacturing optical metrology equipment, glass grinding/polishing included) used this standard control method. It was good to about +/- 1 degree fluctuation, depending on the parameters that were allowed by the central computer (which was in MY office). We had a cleanroom as well, and later on, we added some finer control in some smaller spaces, working off the same basic system, that allowed +/- 0.1 degree sensitivity, and in practical terms, the fluctuation was about +/- 0.2. This kind of thing can be achieved relatively inexpensively, if the basic system is already in place. Limiting traffic in and out is a good thing, as well as looking at how the air distribution works, for location of the sensing devices, and all those other practical details.

Thanks for the response,

So what are the keywords i need to search for such a system? did you just hire a HVAC contractor to set up the plan?

My company is just a little one (me), and my space i'm looking at will be about 3500sq ft subdivided to have one room well controlled. +-0.2 or 0.3 would probably be really good. I presume that's forced air? and is there something like a PID control on flow, or is it very frequent on/off? any idea about vent density or location from measuring equipment in order to enhance stability?

I need to find the right company to talk to about buying a system and some idea of cost numbers in order to price the total of moving into this new facility.

Thanks again!

Ron

Just keep in mind where the CMM is going in the room. You do not want your ventilation system blowing directly on the CMM.

Mark Boucher

Ideally, the room should be in the middle of the building. Exterior walls result in temperature gradients, especially if there is a window.

Windows should not face exterior wall windows. Direct sunlight is roughly equal to 1000 watts per square meter. Nearly all of it becomes heat.

Avoid incandescent or HID lighting. The radiant heat causes temperature gradients. Use T8 (not T12) florescent fixtures with good electronic starters.

White walls allow you to use fewer lights.

Choose your HVAC contractor carefully. Most are incapable of designing a balanced system and simply install too much capacity. The result is strong wind and wildly fluctuating temperatures that aren't even comfortable for humans. We made this mistake. Our shop has better climate control than the metrology room.

Your proposed room shouldn't require more than 500 watts of heating/cooling if the building around it is 50-85 fahrenheit.

Don't forget a HEPA filter. Shop dust (and cigarette smoke) will destroy a CMM.

Don't forget humidity control and recording of both temperature and humidity. The labs I worked in were government so I have no idea of costs but I do know that design, particularly airflow, is critical to proper performance. Make sure that your contractors have a proven track record in such applications. Your local HVAC contractor is unlikely to install a satisfactory system. Be sure that you know the requirements for the level of services you will perform.

Don Young

Breakit:

Don't know if you have already done something about your climate control system. In answer to your questions:
We had a service contract with Climatec, which was the controls subcontractor when the building was built. There are of course other nationally-known firms that will have the knowledge to "do stuff" in this area. Your typical 9-person local HVAC company may not.
The building we built was 60,000 sq ft, in the Southwest, using outdoor chillers with pumps for circulating chilled water to a total of 7 air handlers inside (3 for cleanroom, 4 for rest of building). The air handlers, driven by VFDs, responded to the total load as represented by the individual thermostat settings reported to the controller interface. This is all just a little background info. Typically the air handlers always ran at some speed, so there was always some positive pressure in the HVAC duct system.

The specific gear for the tight control was PID control on the damper for air flow from the air handler (basically controls how cool it gets). In addition, there was a reheat coil in the damper assy to heat the air supply if/when required; the valve also was driven by the PID control. Tuning the room response took some time, probably a few weeks of observation, adjustment, observation, blah blah...It helps to NOT have the airflow directly hitting the sensor... If you engage a knowledgable company, they will be able to calculate the airflow required, and thereby the ducting cross-sections, number of vents, and the controls required for the space. The important thing is to have a good idea of what it is you want to achieve, and what your fallback is when you get the price....

The room in which we first did this was about 150 sq ft with about 9 ft ceiling, and there were 2 computer systems running most of the time, along with test interferometers for optical component checking. Without the control system, the general building systems would cause several degrees of temp swing, which gives some lousy results with tests where you are looking at fractional-nanometer results for process control ... The close temp control really helped.

As other posters have noted, if you are building a room from scratch, for this application, having no walls that are exterior is a big plus. Cost-wise, the gear to ADD this control to an already-existing basic HVAC system with the infrastructure already there was about $3500 including the plumbing run (~40') for the line to the hot water supply, some software adjustments, a new damper assy, and a good bit of work in a small room with a manlift at about 15' off the floor through the dropped ceiling. If your stuff is more accessible, some of the work could be less expensive.

We got better control putting the controller sensing in the return air duct rather than having a thermostat on a wall.

Thanks guys for the replies.

I like the sound of anything PID, so that's good info for me. I'm going to pass these comments on to th HVAC guys and ask for a serious reply if they're comfortable with the job.

I should have said that my application isn't really metrology, so I don't need a standard controlled temperature like 68 per se. I do need the stability like in a metrology lab so that's why i asked this here. I can accept a range of nominal room temp like even between 68-75 but with a tight (anything better than +/- 0.2F would be nice) stability over 1-3 week timeframes. The reason is I have sensitive displacement measurements running 24/7 during such periods but the initial condition can be offset.. temperature variations of a couple degrees are really embarrassing in the results. I think I am facing a combination of temperature cycles (definitely visible as the time the sun hits the building till it goes down in my current place) and electric noise or power fluctuations (also likely sun up-sun down cycles actually) which power the instrument conditioners.

and then again i have about 20kW of induction heaters continuously cycling metal from RT up to about 2000F in 6 minute cycles in this room too

Unfortunately I think I'm going to have an outside wall... I'm thinking about buying an industrial condo unit and would be doing pretty major build-out to get it right, but an outside wall is an advantage for other reasons. certainly i want to not miss anything and can be real flexible about the general system design since nothing is started yet.

3500 to add the control is what I definitely wanted to hear! thanks!

Ron