High Accuracy Weighing

Yeah, don't forget to compensate for tides!

Thanks for input. Surely we will calculate :D

Yes, I know dear

Quite. Potable water is impure to around 0.05% and gravity varies across the Earth's surface even more than that!

This equipment search sounds like a bucket-full of nonsense here....

Actually, using two platforms would actually be better than you specifications. Your resolution would be 50 grams. If the errors of each scale (Repeatability and Linearity) are random (not identical), the combined expected error would not be twice the error of each but sqrt(2) = 1.414 times the error of each.

why?

That's +- 100 CC in 6 cubic meters.

I demand to know what application would require such a small measure?

Take a VAT of a known volume and plant a wee little graduated cylinder on top.

...But seriously? What's the application!?!?

You wouldn't want your winemaker to sneak a glassful out of the vat?

I wouldn't want a whinemaker at all, but the winemaker is entitled to sample the wares, as he earned it. I think the OP is making an ultra standardized solution of something in something, with the other something dissolved in it.

Perhaps he's making homeopathic medicine ��

No, it's actually a solution of cocaine.

Sometimes cocaine is a solution, but not often.

On the contrary.. I want my winemaker to sample to his/her livers content.

Wait? I have a winemaker? I must be drinking. Happy St. Patrick's day bragh

He is making a trap door to his secret treasure trove. Besides, he probably works for India National Bureau of Standards, and is making a STANDARD solution. They do that in the U.S.A. - large vat of standard - all the same to be distributed nationwide.

The $800 fluke meter is to show the customer just before handing them the bill.

I for one am happy as a clam with my generic meters. I can step on them, drop them, spill liquids on them, even blow them up or loose them.. and none of it matters.

My most frequently used digital meter was $2.99 new. Of course it was much more expensive if I didn't get it on sale.

..You must be right about his 'requirements' not matching his actual needs. GA! even if yer wrong.

I use those $2.99-on-sale meters, too.

The site sparkies choke when they see me using it, but 80% of my work is checking 4-20mA circuits, 10% whether the 24Vdc supply is still functioning, 7% continuity, and 3% is checking that the 120Vac outlet is still hot. It works just great for all of those tasks.

And the fact that these meters always read 0.02mA low and there's no adjustment doesn't matter. If I need a calibrator, I use a calibrator. When I don't, I use my cheapie.

When it breaks or its cheap probes break or it falls down a flight stairs, I shrug and cut the plastic shrinkwrap to extract another one from its packaging. The rest of the world loves disposable cell phones. I love my disposable cheapie DVM meters.

Some years ago I worked at a local place that sold both fluke and Mastech meters. Mastech impressed the heck out of me given we ran them head to head on every possible real life measurement to see what the difference was between the flukes and the Mastech.

Well for one it took three flukes worth ~$1000 to cover almost all the functions the single ~$50 Mastech did.

Next, Mastec had a 3 year warranty, fluke 2.

Same Cat II and Cat III ratings.

Plus in head to head the Mastech meters matched the flukes within an at most+- 1 - 2 counts on the least significant digit if not dead on reading. 120.24 Vs 120.22 - 120.26 reading variance. (same reading variances typically found when running two flukes side by side 'but that didn't mean anything' to the 'experts'.)

And last according to sales records Mastech had near zero returns for warranty Vs fluke having about 5% for number sold.

Now the most impressive part was the guys who came in and bought Mastech meters typically knew they wanted a good mid range multi function meter and likely knew how to use it.

Whereas the guys who bought the flukes well, they loved the autorage features, being most didn't have a clue what they were doing (Typical contractor types with more money than brains), and thusly also had no clue what the numbers showed when doing said real world head to head comparisons for them.

fluke came with a lesser warranty, an outrageous price for the few near self operating functions it had and in day to day work told them the same information a $5 harbor freight meter did but it said fluke on it and cost 10+x the price (despite a poorer warranty and very well documented track record of having all around more warrantied repairs and crap service and support) therefore it still had to be the best.

Personally I have a Mastech 4 1/2 digit unit I have beat on for several years now and it still works perfect despite having been left out in the rain and snow multiple times on top of being severely abused in ways any self respecting fluke owner would poop themselves over.

And yes it's really nice knowing my house power is at xxx.xx (+- 10% - 15%) at all times according to its readings. Those last 3 digits make all the difference in the world in my knowing whether the circuit is turned on and within the two digit percentage range of working limits or not.

Mastech 8265. Great mid range 4 1/2 kick around meter for under $50.

Their 50,000 count true RMS auto ranging fluke killer for ~$200

Mastech 8218

Where did you steal buy that $2.99 one? Eburglary.com?

Just kidding! Sounds like a deal to me, especially when quick and dirty is good.

Harbor Freight! Not always, but every now and then.

Get a weighing and load cell expert company on board right away.

Give them complete specification for your X level of precision, and Y level of accuracy, and make them stick with that before custody transfer of the device you are buying.

Expand the load range of your system so that you never ever can reach the damage threshold. Say you make it able to weigh 7500 Kg +/- 0.05 Kg (very nice set of tank scales indeed), and since you will have an overflow (won't you?) on the tank at a preset level before 7500 Kg is reached, you will never push the load cells past the 7500 limit, much less a limit of 110% (8250 Kg), since you plan on stopping test at precisely 7000 Kg gross weight.

Caveat Emptor: Let the buyer beware

Vos Modo Id Quod Est Reddere: You only get what you pay for.

So pay attention, and you will get attention. See?

1. Yes, the tank will be tared eevery time. The tank will be made up of Stainless steel.

2. It is for calibration rig as per ISO 4185.

3. The aim is to make it National standard level measurement.

4. The weight will be converted to volume using high accuracy density measurement

Why not just use high precision physical measurements of the tank roundness, diameter, and depth, then combine that with precise and accurate density measurement (temperature if water), to arrive not only the volume, but the mass contained? I think this could be very precise.

Consider that not even a stainless steel tank is perfectly round. You need a roundness parameter.

Controlling the dimensions is not so easy.

What about continuous dimensional measurements? Shirley says that cannot be all that hard.

Actual facts: Even a tank of that size will change dimensions on temperature, but will also demonstrate uneven gains (and losses of effective radius) along the height of the cylinder (if that is the symmetry or supposed symmetry). There can be warping due to differential heat inputs from insolation for example.

Tidal forces will enter in at some point, seriously.

You need to make sure whatever the choice, you can deliver on your promise.

Gsuhas: what is the application? If batch you can transfer to a smaller weighed tank to dispense, if continuous can use a flowmeter.

Iris: +1 on the great Minebea Intec loadcells. But you forgot they were Phillips before Global Weighing.

Much is already saird about the challenge you are going to take: But I cna give you some hints too.

1) company - look also into Hottinger Baldwin Messtechnik GmbH they have also excellent laoad cells and mesurment systems.

2) The tripod spolutiin for your tank is best and gives very good resultes. It might not be required to have 3 loadcells - If the design is well done might be 2 load cells are sufficient. I remember to have seen that in Hottinger Baldwin Catalog - which gave lots of hints how to install the load celles correctly

3)not mentioned are the foundations of the platform where the loadcells and tank will be located. With the rquested precisition please understand that already trucks passing close by can inflict vibrations to your system which can cause high jitter on your measurments.

4) You will need special cables and excellent shielding to suppress any electromagnetic noise from workers running around with smart phones and tablets with bluetooth. The inputs of your measurement gear has to have excellent RF suppression (low pass filtering) to avoid disturbence of your measurements. A fact some times totaly over looked!

I had load cell based crane load supervision units installed in the cabin of several tower cranes. I was lucky as the RF interference (GSM, microwave) from a nearby telecom exchange was already considered in my design. Other cranes working close by using a simpler design from a korean had different readings depending of the relation of the crane tower to the antenna tower and activities of the GSM radios and walkitakies the crane operator used. - So the cranes did not pass the crane inspection as the measured load was not really readable and jumping around and not within the required 5%! limits.

So happily I could sell some (not so cheap - more expensive ha,ha) low pass filter modules to the competition to get these cranes through inspection. The parts were cheap - but there was lots of KnowHow to pay for!!

5) other pont is the calibration. If you use a liqid you have problems as it is very difficult to get the right weight for calibration. For the cranes we could use rebar steel bars which - despite the fact that they look so crude - have an astonishing accurate weight. I can not really remember, but it was much better than our required tolerance. And as cutting of rebar is done by sheering there is no loss of weight when it is cut to more handy pieces. Ask a rebar vender. And during calibration they are lying around without any movement - They are not inflicting unwanted noise into your measurement system.

To calibrate we basically lifted the steel only some meters from the ground. But even then you could already see the impact of the wind to the ropes and the crane arm.

And the software had to cater for the load of the rope as this was either runnig over a pair of rolls or with twice the length over a pully- There was a diplay which showed the SWL (safe working load) which was in the pully mode naturally lower as due to the height of 120m a considerable lenght of the cable had to be considered.

Given the degree of accuracy required this has to be a laboratory application, possibly clean room environment. Not many fork trucks in a lab. He is going to need calibration certificates traceable back to his local National Standards and a calibration regime that requires re-certification every three months at most. Most load cell cables are shielded with resistance compensated short runs between the cells and the amplifier, so RF should not be a problem. Temperature I agree will be a major headache. That is why I think he will need to put this in an environmentally controlled space.

(From the original post): I need high accuracy weighing platform for weighing 6000 kg of water in a tank (Tank Tare weight around 1000 kg)

2. There are 3000 Kg capacity weigh platforms with Resolution 50 grams

If I put such two platforms side by side, put my tank on both of these in sharing, and add up the outputs, can I get what I am looking for?

No. Two 3000 Kg platforms would handle a 6000 Kg load. You would need to use 3 of these platforms to handle your 7000 Kg load.