I'm not sure this forum is the right place for this post but it's probably better than others.
Makes me think we might need a "Measurement" sub forum?

In the shed I have two, bulky, good quality, old school, bench top lab balances, a Mettler 10kg to 1g, and a Sartorius 1kg to 0.01g. I also have two budget digital balances (1kg to 0.1g , and breast pocket size 250g to 0.01g).

Recently I gave SWMBO the 250g balance to use for her craft work and used that as an excuse to buy something small I could still use inside and something with a bit wider weight range so I purchased one of the cheapest ($10) 500 g to 0.01g balances I could find on eBay. It's advertised as a kitchen scale so I was not expecting much

Around the same time I purchased a budget ($20) set of calibration weights from Ali-express.
Quality calibrations sets are usually made of SS (ebay sets start at around $50) or special alloys or even precious metals but this one was nickel plated steel for the larger weights and Al for the smaller weights.
Over time Al corrodes and changes weight so cannot be use with much reliability as a calibration weight unless they are recalibrated.
Anyway, whatever they are, I figured I could recalibrate these weights using a certified 0.1mg balance at my former pace of employment and then would use the recalibrated weights to check out my home balances.

IMG_4180p4.JPG

The accuracy claim on the calibration weight set (200g - 10mg) was accurate to +/-3mg

Checking these weights out on the 0.1mg balance at work I found that of the 19 weights supplied all bar 5 of the weights were within spec.
The 5 weights that were out were, the two 2g weights, and the 1g, 0.1g, 1g and the 10g weights, but they were only outside spec by no more than 4.5 mg.

The calibration weights <50mg are within spec but in % terms(up to 23% outside spec) they are not that accurate.

Given the calibration set only $20 I was pretty impressed. In practical terms it doesn't matter that they are out as I have now recalibrated them and can use them to check balance accuracies across the weight range to within ~0.2mg (the accuracy of the balance at work).
Another practical aspect is that given they are all within 4.5 mg I figure I can use them without any correction to check a balance that can at best display 10 mg.

Next I used the recalibrated weights to check out the $10 balance.
There was no stated accuracy claim for the kitchen/budget balance by the vendor but lets assume it is twice the smallest displayed digit (ie 20mg).

Using this criteria all the balance measurements between 20g and 50mg are all within the 20 mg spec.
For 50g and above the there appears to be a systematic "over weight" being displayed by the balance,
60 mg for 50g (ie 50.0g)
120 mg for 100 g (100.12g)
200 mg for the 200 g (200.20)

For less than 50 mg the balance irregularly displays a weight - sometimes it displays close to the "correct weight", other times it displays "a smaller weight" and mostly it displays ZERO. It appears as though the balance does not trigger consistently for weights <50mg. It's interesting to note than most cheap sets do not supply calibration weights less than 50mg probably because of the effort required to make the small weight accurately.

Another feature I see on this cheap balance (also common on many other cheap balances) is that the weights displayed can differ depending on "where" the weight is place on the balance top with slightly lower than expected weights displayed when the weight is placed towards the outer rim of the balance top plate. The safest place is always the middle.

I'm not complaining about either the calibration weights or the budget balance - in fact I believe I got more than I paid for.