I have a bulk tank, with weigh scale, that needs calibrated. Operating capacity is 3400lbs. The current calibration method is to empty tank, tare and calibrate using increments of 100lbs up to 500lbs. I know that standard calibration practice should be to calibrate with increments upto 3400lbs, but the equipment structure itself doesn't have the space to load the tank with 3400 lbs of weights.

What calibration method would you propose?

I have a bulk tank, with weigh scale, that needs calibrated. Operating capacity is 3400lbs. The current calibration method is to empty tank, tare and calibrate using increments of 100lbs up to 500lbs. I know that standard calibration practice should be to calibrate with increments upto 3400lbs, but the equipment structure itself doesn't have the space to load the tank with 3400 lbs of weights.

What calibration method would you propose?

Do you use the full capacity?

Yes. We have considered the idea to tare, calibrate, add 1000lbs of chemical, tare, calibrate, add... repeat this until the capacity is maxed out. But I am concerned this won't verify gage linearity.

We have also considered the idea to tare, calibrate, add 1000lbs of chemical, and calibrate again without tareing, calibrate, add... But the chemical added becomes a calibration standard which may not be exactly 1000lbs.

Our calibration service feels that the 500lbs method is sufficient. I'm not convinced. Accurate scales are critical to recipe control when we blend chemicals.

Put 407.43 gallons of water in it that weighs 3400lbs.

Put 407.43 gallons of water in it that weighs 3400lbs.

In this case, the water becomes the calibration standard. How can I show traceability to a nationally known standard for water?

Actually, the idea of putting a certain volume of water is a pretty good idea. I would not have thought of that. You could use a calibrated flowmeter and measure the amount of water going into the tank. Use that as your standard.

Use an instrument that is calibrated to a national standard to measure the water weight of the water supply you use.

I am still trying to figure out your setup here. Is this tank on load cells? Or, is it permanently affixed to a scale? Like Jim asked, what capacity do you use this to will assist in determining how high you need to calibrate this.

Just considering the details, because I think the idea is great also. For folks who know more about water than I do...can the weight of water vary from water district to water district based on the amount of contaminants in the water?

The weight of the water will depend on the specific gravity. Namely, the specific gravity with water is correlated to the temperature.

I guess all this depends on what requirements you have and how accurate you are expecting this to be.

I am still trying to figure out your setup here. Is this tank on load cells? Or, is it permanently affixed to a scale? Like Jim asked, what capacity do you use this to will assist in determining how high you need to calibrate this.

We have a series of tanks. Some sit directly on load cells, some sit on scale platforms. Either way they are permanently installed.

With your water supply, if you have a known supply that you will use just take a sample to an accredited lab and get the water certified for weight and gravity and once you have your supply certified then periodically you can sample and check it against your standard certified water supply.

Just audit you lab to make sure their instruments are calibrated and traceable!

I have seen all kinds of setups. Some load cells can be pulled off and have weights placed on them. The output of the load cells can be measured.

Let’s face it… It’s hard verifying the accuracy of tanks/large weights.

Well folks I was getting goose bumps for having a solution with using the water idea. The concern that came back from our process engineer is that after calibration, the water will be contaminated due to residual chemical from the tank and will have to be disposed of as hazardous material. Also, after calibration, any residual chemical still in tank will be saturated with water which will have adverse effects on future chemical blends.

You could use the same setup with the chemicals you put into the tank.

It all boils down (to me) how accurate you really need to be. You could tank measurement dimensions and go off of volume, if you didn't need to be that accurate.

Then use the weight of the chemical as the standard.

I'm thinking the following:

Make chemical blend as normal
Drain chemical into secondary tote(s) sitting on calibrated scale
Pump chemical back into same tank in metered amounts
Calibrate tank scale at various increments


Does this sound legit?

Well folks I was getting goose bumps for having a solution with using the water idea. The concern that came back from our process engineer is that after calibration, the water will be contaminated due to residual chemical from the tank and will have to be disposed of as hazardous material. Also, after calibration, any residual chemical still in tank will be saturated with water which will have adverse effects on future chemical blends.

What is the chemical? Flow meter and specific gravity is still the way to go. What is the required accuracy? Over what range, with what needed precision? How do you currently clean the tank? Can it be heated?

Well folks I was getting goose bumps for having a solution with using the water idea. The concern that came back from our process engineer is that after calibration, the water will be contaminated due to residual chemical from the tank and will have to be disposed of as hazardous material. Also, after calibration, any residual chemical still in tank will be saturated with water which will have adverse effects on future chemical blends.

Are the chemicals water soluble? Why would a minor amount of chemical contamination make the water a hazardous waste? Would it violate a discharge permit?

If the chamicals are water soluble and there's no problem with disposing of the water, using a calibrated flow meter would be the way to go.

This issue would be moot if the calibration company would do what you asked them to do - is there some reason they won't other then they think you don't have to? That's not really for them to decide.

You could write an SOP for calibration companies to follow when they calibrate your equipment.

Here is a table I found for the temperature dependence of density for pure water.

Temp (°C) Density (g/cm3)
30 0.9957
20 0.9982
10 0.9997
0 0.9998
If you can measure the temp to within a couple degrees, then the density will be correct to within 0.1%

Also, I don't know specifically about dnsity of tap water, but seawater is about 1.025 g/cm^3. Since this has much more dissolved solids than tap water, the different in density between pure water and tap water will be much less than this 0.25% difference.

And I just found another source at http://www.simetric.co.uk/si_water.htm
that suggests that tap water is practically idential in density to pure water.


On the other hand, why even bother with density? Wouldn't it be easier to just weight the water, rather than measure the volume?


Tim

Tanks can be tough.
There are several ways to go.
Dead weight is the best however it is usually the most impractical.
Hanging the weights on fabricated hooks at an equidistance, or a makeshift platform over the tank takes a lot of effort.
If you choose to use weights you can use product as a substitution weight as long as you don’t exceed the amount of certified weights you have. 3 substitutions are legal.

Legal for trade accepts 12½% of capacity for calibration.

I suggest you use a calibrated flow meter that will accept your product without damage to the meter. They can be found rather readily but are not cheap. You will of course have to know the absolute density of your product to calculate the weight.
You need to know that the best accuracy of any flow meter I have found is 1%.
Make sure you also consider the accuracy of the tank cells.
Do not set your tolerances too low.
If you require very close tolerances, the best bet is to add product via a batching/adding scale.

This may be considered a non standard calibration method depending on your scale calibration company. If you need 17025 calibrations, you may need to sign a waver.

Hope this helps