Question Anybody mix gas by weight?

[Akimbo]

I was at an industrial gas plant in Germany in the 1970s picking up some big cylinders of cal (calibration) gas and got a tour of the plant. I was able to watch them mix the gas and was surprised that they used a really big precision balance scale and mixed by weight instead of partial pressure. I asked them how accurate it was and they said that the gas chromatograph reports that ship with cal gases are usually within 0.2% of the target. I have been to a several gas plants since and all of them mix by weight.

Precision digital scales that can handle 50Kg/110Lbs at 0.1g/0.220 Lbs are available today for $500-850 US. Still expensive but not inaccessible. Does anybody use them?

 

[inquis]

It certainly removes the vagaries of non-ideal gas pressures, since there are a few different ways to model that.

Seems like a lower capacity might suffice, where doubles would therefore have up to 2x the error.

 

[Akimbo]

As I recall, the bottle was standing on the scale and they had a large double loop of HP hose hanging from the overhead. The scale would jump when the hose was first pressurized but was quite stable after that as the scale moved toward balance. It should be a lot simpler now by pushing the tar button on a digital scale.

It didn't look like there was much difference in price between 30 and 50 Kg capacity. I think I would invest in a set of calibrated weights to check the scale but you are only adding at a few Kg at a time to a set of doubles.

 

[Chris H]

A lot of specialty gases are mixed by weight. There are automated fill stations that add gas to a cylinder or cylinders. It is overkill for diving gas, but it might be fun to mess with. There are procedures for accounting for the flex in the fill lines.

 

[Chris H]

Here is an example of what the industrial setups are capable of. Sur-Fill iQ-s Automatic Specialty Gas Fill System

 

[broncobowsher]

When weighing refrigerant you have to keep the scale on a flat stable base and be careful that nothing even bumps the hose. When dealing with such small weights, everything matters. The comment that the scales jump around when the line is first pressurized, that cannot be overlooked. If the cylinder moves a little bit and the tension the charge line changes, that will mess up your fill.

But it would be sweet for not having to deal with temperature. But I bet it would be tricky to hit your target pressure. Especially on small cylinders. One gram is how many PSI on a 3L diluent bottle?

 

[Chris H]

But it would be sweet for not having to deal with temperature. But I bet it would be tricky to hit your target pressure. Especially on small cylinders. One gram is how many PSI on a 3L diluent bottle?

If you are filling by weight, pressure is not the target, the weight of the gas you are adding is. The software that I’ve used has specifications for many different cylinder types. You choose the cylinder, and the mix you want in it. If you choose incorrect cylinder specifications, you overfill or under fill the cylinder.

 

[TrimixToo]

Wow. Interesting. One of the difficult things with trimix gas blending is developing a feel for how much helium to top with (assuming O2 first) if you want to get fills done in a reasonable amount of time. It's not completely straightforward nor perfectly accounted for in any of the blending software I've yet seen. Doing it by weight would account for these variations in pressure and temperature en passant.