Liquid metal slip/rotary contact

[girts]

are there any already existing devices on the market or possible ways in which one can make a liquid metal slip contact like a mercury contact in which the mercury can be held at a specific location within a confined space with the help of magnetic field?
I know that mercury itself is very weak or for all practical purposes non-magnetic and only responds to B fields if a current runs through it but are there any other ways to make a reliable connection by holding the mercury at a particular place given that it would be subject to rotation?

imagine the mercury inside a torus shaped tube and as the tube rotates the mercury should ideally be held at a particular place in the tube against the forces of friction from the inner surfaces of the rotating tube and also against gravity, is this doable?thank you.

 

[anorlunda]

You might find a ferrofluid https://en.wikipedia.org/wiki/Ferrofluid with good electrical conductivity too. Then it can be held in place magnetically.

 

[girts]

Hi anorlunda, but ferrofluid to the best of my knowledge is nothing more than some sort of fluid in which metallic dust or little metallic pieces are mixed in correct? I assume the metals can be of various types but if one wants the fluid to be magnetic then it must be some ferrous metal type, the problem is which is also my question can one get low resistance path through ferrofluid? I assume that this is a problem because at least with mercury the resistance is as low as that of a copper wire which is good.

 

[anorlunda]

I assume that this is a problem because at least with mercury the resistance is as low as that of a copper wire which is good.

Yes that's correct, but the next step in answering your question is to give the actual requirements.

What is the required resistance?
What is the size/weight restriction? (because more fluid compensates for less conductivity)

Your question in the OP does not mention any numbers at all.

By the way, I just searched for electrically conductive ferrofluids. I found several patents, but nothing for sale. So maybe they are not available yet.

You may also be able to accomplish your goal without any kind of fluid in the switch. Would you like to describe what you're trying to do?

 

[jrmichler]

Mercotac makes slip rings with less than one milliohm resistance. http://www.mercotac.com/. I've used them, they work well.

 

[girts]

thank you for the replies,

well the idea is quite simple actually, imagine a normal commutator like that found in universal and DC motors, only in my case it needs to be more specific and due to the layout located on the outer circumference where angular speeds are much higher than closer to the center of a rotating shaft which would make the brush contacts wear out a lot faster and cause additional problems.
just a curious motor project.

so my idea was to do the commutation by using something like a closed circular tube made from some dielectric isolating material (read any plastic for low voltages) and having some electrical contacts that penetrate the tube and then using say mercury inside the tube held at a particular place (where needed) and as the tube together with the contacts would rotate the mercury would electrically connect the particular contacts near it as they move by.
low resistance is very important here because the voltage is also low and so higher resistance like that from a ferrofluid probably would cause a lot of inefficiency and loss of power.

surely another way of doing this would be using semiconductors like IGBT's or else but due to the number of contacts I would need many transistors and that would be costly, also there would be a current limit.
probably one way of doing this with mercury would be taking use of the current that would run through the mercury and applying a B field at the correct angle which would then have a net force on the mercury as it would become magnetic due to the current through it?

 

[sophiecentaur]

so my idea was to do the commutation by using something like a closed circular tube made from some dielectric isolating material

The way things are going, I would think that brushless technology will take over from regular commutator motors. Perhaps that would solve your problem.

 

[jrmichler]

surely another way of doing this would be using semiconductors like IGBT's or else but due to the number of contacts I would need many transistors and that would be costly, also there would be a current limit.

IGBT's are made with pretty impressive current limits. This one is rated for 1700 volts and 7200 amps peak current: http://www.newark.com/infineon/fz3600r17hp4b2bosa2/igbt-module-n-ch-1-7kv-3-6ka/dp/49AC0230. There are probably larger ones, that's just the first one I found in a two minute search.

Trying to use mercury in a commutator creates another problem. Commutators switch current on and off under load, which causes arcing. Arcing in mercury vaporizes the mercury. And mercury vapor is not something to be taken lightly. I suggest that you spend some time studying the health hazards of mercury.

 

[girts]

as for the mercury vapor, that isn't a problem because first of all this is not meant as a typical commutator, here the mercury acts more like just a constant contact which only needs to rotate, the problem isn't in switching currents the problem is in transferring current from a rotary part to another rotary part and using a connection that provides as low of a resistance as possible while doing so, secondary the mercury would be in an enclosure much like a mercury vapor lamp aka "daylight" tube, so it poses no danger.

I was more interested in how one could maintain the mercury at the specific needed spot given that the whole assembly is rotating, my idea was by using a B field.
well imagine a torus shaped circular tube and mercury inside it, as the tube rotates the mercury needs to be at a specific place to make the contact, now since the current goes through the mercury theoretically I could add a B field that runs perpendicular to the mercury and so as the tube would start to rotate and drag the mercury along the inside of it the mercury should stop and stay at the specific place where current runs through it because that current would then interact with the B field applied and hold the mercury there as long as there is current through it correct?
also, @sophiecentaur can you give me some evidence for saying that we will have brushless motors because as far as I know there is no sign of ordinary copper commutators or slip rings not being used both in small DC and universal motors as well as large ones and also generators.thank you.

 

[sophiecentaur]

also, @sophiecentaur can you give me some evidence for saying that we will have brushless motors

I have a number of cordless hand tools with brushless motors in and I believe that higher power motors are used in electric and hybrid vehicles. As you don't yet seem to have answered the question about your power requirement, I can't tell if that is those powers would be suitable for your application. If you Google "Brushless Motors" there are many hits (including a Wiki link) which will give you a flavour of the availability and applications of brushless motors.
Brushless technology is not cheap, of course but, if you are considering such a totally alternative configuration for a motor then you are obviously going to have to spend money.
If brushless is being used for automotive applications then you can assume that it is very much with us.

 

[jrmichler]

All industrial servomotors are BLDC (brushless DC) motors. They are made in sizes ranging from the size of your thumb up to several hundred horsepower. Other variable speed applications use induction motors with VFD (variable frequency drives). No modern electric vehicles use motors with brushes for their traction motors. Siemens makes a line of electric motors for aircraft. None of those motors have brushes.

My ceiling fan has a two phase induction motor (with a series capacitor to get the phase difference). Low efficiency home forced air heating systems use induction motors for the blowers, while high efficiency systems use ECM (electronically commutated) motors.

I think automotive heater blower motors may still have commutators. My corded electric drill has a commutator. But those are cost sensitive, low performance applications where short life is acceptable. Commutators and brushes are obsolete technology.

Mercury is a problem. I designed a system that had a Mercotac in it. We had to change to slip rings and brushes in order to sell it in the EU.

 

[girts]

yes folks, I am aware that more and more BLDC motors are being used in DC source applications, but knowing more we know that a BLDC is nothing more than a single or (usually) three phase AC induction motor, there the mission is a bit more simple as they use semiconductors to switch DC within the stator coils,
in my case my coils (if I can call them that) are located on a rotor, and I need to switch parts of wire on in series and off as they pass a certain point in which a B field cuts them perpendicularly.
this is the reason I was thinking about a liquid metal contact that can be within a tube and the whole assembly could just rotate.
another way ofcourse is to use semiconductors but then I would need one semiconductor for each wire part that needs to be switched on and off.speaking of brushless motors, I assume they are used in automotive because the torque vs longevity balance is much more in favor for longevity since passenger cars are lightweight, but I imagine for applications such as trains and locomotives or other high power applications like handheld power tools for example where current is also provided to the rotor coils they will still use copper commutators since even though they wear out but are rather cheap and robust with respect to overcurrent and voltage spikes etc.

 

[sophiecentaur]

@girts I don't think we can usefully continue with this until you come clean about the power and dimensions of your design - we have no idea other than your statement that automotive motors are implied to be under powered for your application. That implies your commutator contacts will need to handle hundreds of Amps. It does make me wonder why nothing novel has worked for motor commutation for decades yet you feel you have a solution. I think you would need a pretty expensive research set up if you seriously want to develop a replacement brush at this stage.
It would seem that your design stands or falls on the availability of suitable 'brush' replacements. It seems there is nothing that will do the job so perhaps you should consider a completely different approach.

 

[anorlunda]

One way to describe a brushless motor is "AC synchronous motor with a permanent magnet field on the rotor" The "AC" is generated by switching the DC in a pattern.

They are not just small motors, here is a 20 KW one. https://www.alibaba.com/product-detail/High-Efficiency-BLDC-motor-20KW-electric_60500835073.html

 

[sophiecentaur]

They are not just small motors, here is a 20 KW one.

Unfortunately, the OP doesn't make it clear whether 20kW is 'low' or 'high' Power compared with his application. (?)

 

[girts]

I know it seems a little off putting that I don't specifically say my exact application but in this case it is because I am still thinking through the specifics of it.
the application is rather complicated and I do now have the skills to make a good technical drawing using a software like Solidworks or CAD or others, so I don't want to stir unnecessary confusion by uploading a poorly made MS Paint drawing which would be so hard to understand that it would rather derail any meaningful discussion altogether.
My friend is an engineer a mechanical one he said he would be able to do the drawings etc.
I would like to continue this maybe after I have a detailed schematic.@Sophie as for what you said why no one has come up with a novel way to commutate motors except copper and brushes and semiconductors but then the rotor is without a coil. Well I think that is because in a motor application in order to achieve higher output power for a given size motor one has to supply the current to the rotating part of the motor aka the rotor which in the case of liquid metal makes it complicated because one part of the contact must be stationary while the other rotating, my idea was to use this setup that I was talking about in generator, you see in a generator of this type (the one still in progress) I don't need a stationary reference for the contact , just that the liquid metal can be made to flow independently of the tube in which it is in.
so in theory I can use a sealed tube and mercury within the tube so there are no seals or parts that rotate with different speeds, so it can't get out.

I envision that I could use semiconductors as well but I would need many of them and they would introduce the need for additional heatsinks etc so mercury would be the best choice.

 

[sophiecentaur]

As I have already commented, if you cannot state what sort of power you are interested in then it's hard to contribute further. Suffice to say, solid state commutation is superior in pretty well every way to the old mechanical version. (Except for cost of cheap implementations) We moved away from Mercury Arc Rectifiers and Mercury Arc Invertors for all but possibly the very highest powers (and I could be corrected on that).

I don't need an engineering drawing of the actual layout of your motor idea in order to have some idea what this is all about. I guess you are excited about what you consider to be a potentially good and novel alternative to a copper commutator but it would need to be pretty good to do better than present and future Brushless Motors. When you consider that solid state switching can handle kW's of power with no trouble, it would be hard to improve on at this stage.

I envision that I could use semiconductors as well but I would need many of them and they would introduce the need for additional heatsinks etc so mercury would be the best choice.

Are you aware of the power limits of a mercury based contact system too?