Laboratory Environmental Controls - Temperature and Humidity

R

Ryan Wilde

Mike,

I would say that it would depend on the type of calibrations that you are performing. If you are doing dimensional calibration to within a micron, it is a huge problem, but if you are calibrating a handheld multimeter, then it is no problem at all. (I can't tell by your spec'd temperature range, because dimensional labs should be 20°C [68°F] with very tight temperature constaints, while most others are supposed to be 23°C [73°F] and have greater tolerance for temperature swings.)

Also, he may be correct about "The laboratory shall ensure that the environmental conditions do not invalidate the results or adversely affect the required quality of any measurements". It would be prudent to point out to him that if he goes with a blanket generic statement in the manual then the specific temperature range and rate of change would have to be specified in each individual calibration procedure.

What it comes down to is this:
*What effect does temperature have on the uncertainty of the measurements I am performing?
*Is this effect acceptable?

A metrologist has to ask these questions and calculate the answer. If you go to your boss with hard numbers, you may change his mind, or you may just find that it truly is insignificant to the measurements you are making.

Ryan
 
E

energy

Mike,

Why don't you have your chart set to read once an hour, per your procedure? Or, change the procedure to say 15 minutes? Becomes a non issue. Then, ask your Boss to tell you how you can insure validation of your measurements in that environment. If he/she does that, make a space for him/her to sign the Calibration Certificate. Sounds to me like sand in your eyes while he/she kicks you in the kneecaps.
JMHO

energy


[This message has been edited by energy (edited 11 July 2001).]
 
M

mike1042

We do a large number of dimensional gages and I am aware of the requirement of 68 degrees but with the temperature swings we have it is difficult as best to perform calibrations. It has been a big shock to see how some companies handle their calibration programs. This is one of them. We are certified to ISO 9001 & AS 9000 but how we maintain these certifications is beyond me. I guess I need to sit down and do the math to show him the errors of his ways.
 
M

mike1042

Laboratory Environmental Controls

I have a problem with my boss and his interpertation of what constitutes a stable calibration environment. Our quality manual states 68-72 degrees F with a rate of change of no more then 2 degrees F per hour. Well our environment is changing approx. 2 degrees every 15 minutes. The chart looks like a saw tooth. All my training in Metrology says that this is a problem. He says it isn't because it isn't greater than 2 degrees. Also he wants to spec changed to a read "The laboratory shall ensure that the environmental conditions do not invalidate the results or adversely affect the required quality of any measurements". This does not sit well with me. I have tried to explain the problems with this but I guess I'm not doing it correctly. Any thoughts on this?

------------------
Mike Zimmermann
 
R

Ryan Wilde

Mike,

Something about this has really been bothering me, and that is where you said it was a sawtooth. It's been eating away at me, and I just can't take it anymore. If you have a sawtooth, then your environmental system is adequate, but your temperature control (thermostat?) is not. This is a VERY cheap fix. If the environmental system is capable of bringing the temperature back to nominal in a short time, then a very simple PID controller with RTD sensor will give you a beautiful environment.

It seems with a 2°F swing, you have a thermostat that is ±1°F, in other words, set at 69°F, on at 70° and off at 68°. You have the makings of a great system it seems. The systems that are hard to fix are those that are slow to recover. There are a few companies out there that sell temperature controllers for a few hundred dollars that would make your environment fantastic. If you need info, feel free to email me - it's in my profile.

Whew, I needed to get that off of my chest. I know it wasn't the point of the thread, but it is a fix...

Ryan
 
R

rock

Hi Mike,
You're really trying to measure the temperature of the material you're calibrating. Try placing the sensor in the hole of a four inch gage block and see if the spikes go away. They should. Air changes faster than metal. There's an ANSI or ASTM spec that deals with this. Sorry I can't give you the number. Rapid air fluctuation around a point is better than slow fluctuation.
Mike
 
Top Bottom