I am designing a product that requires internal mixing of a sealed container. If you need to visualize the container, imagine some quasi-oil drum that holds about 20 L of fluid.

I am looking to use as little electrical input as necessary so I will be using a manual hand turner with some sort of paddles attached to mix the liquid appropriately. However, the seal on the joint or input valve connecting the manual crank to the tank must be as airtight as possible.

How do I make the hinge connecting the crank to the tank as airtight as possible?

  • $\begingroup$ Why not fasten blades to the inside of the container and rotate the whole thing, like a cement mixer? $\endgroup$ – Dave Tweed May 25 '16 at 14:42
  • $\begingroup$ That was the original plan, but there are going to be two output and one input valve and it would cause an issue of both torsion of the pipes and getting liquid into unwanted output valves (one for gas produced one for liquid). $\endgroup$ – tyler a May 25 '16 at 15:10
  • $\begingroup$ Hinge is not the correct term for what you want. You're looking for a seal or gland for the mixer shaft. $\endgroup$ – Dave Tweed May 25 '16 at 15:31

What you're looking for is a mechanical face seal, which are used to allow rotating shafts to penetrate into a sealed volume such as a pump, mixer, compressor etc.

One important thing to remember about these seals is that all such seals leak to some degree. The exact seal you need will be driven by your specific requirements such as:

  • How much leakage of the process fluids can you tolerate?
  • What is the process fluid?
  • What are the pressures and temperatures?
  • Operating speed
  • Budget
  • etc.

On the higher end are engineered and stock mixer seals and cartridge seals like these from Flowserve. Those can easily run into the tens of thousands of dollars depending on your specific requirements.

There are also less expensive OEM seals designed for things like swimming pool pumps. Such as these, also from Flowserve. Again, depending on the specifics of your application you might get away with a sub $100 solution.

At the low end of the scale is simply using one or more radial lip seals that ride on the shaft, usually on either side of the bearing for the shaft.

You can also eliminate the need for a seal completely by using a magnetic drive mixer like these from Magnasafe. I have never seen a hand operated model though.

I can give you further guidance, but not without more detail from your end.

If your project involves any significant gas pressure, I strongly suggest that you following the ASME code for pressure vessels or contract with someone who is qualified and incorporating safety features to limit the possibility that the vessel could become over pressured and turn into a bomb. People continually underestimate the energy contained in relatively small volumes, like 20l, of pressurized gas and the consequences can be lethal.

Disclosure: I used to work as an product development engineer in the Seals division of Flowserve. I cited their literature because I'm more familiar with their products.

  • $\begingroup$ Thank you so much for the detailed response. Basically what I am designing is an anaerobic digester. So the liquid inside would be semi-viscoius, more so than say water, similar to oil for example. The pressure within the container should not be incredibly substantial, but it will definitely be a concern and is something we are looking into. I have recently stumbled upon the design for how a magic blender works, and its basically two O-Rings so that could possibly do the job (same thing used on the Cupola of the ISS shutters really). I hope this makes sense $\endgroup$ – tyler a May 26 '16 at 7:13

If the torque requirement isn't too high you can use a magnetic coupling between the outside crank and the inside paddles, like a magnetic stirrer.

  • $\begingroup$ I wonder if the magnetic approach is practical for viscous fluids or non-homogeneous mixtures—sludges and the like. I.e., can it scale up from a stir bar to a full paddle? $\endgroup$ – Air May 25 '16 at 16:38
  • $\begingroup$ @Air that's why I mentioned the torque requirement. $\endgroup$ – ratchet freak May 25 '16 at 18:49
  • $\begingroup$ Thought about some sort of magnetic coupling as you mentioned, but yeah the torque requirement is something we are still looking into. Thank you though! $\endgroup$ – tyler a May 26 '16 at 12:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.