Tuesday, January 29, 2013

Probe Lifting Bracket

One of the physicists at work approached me with a problem recently. At first it looked like some kind of hoisting problem requiring  lots of oversight and engineering approvals. As it turns out once we discussed all the details its really an ergonomic problem. I'll describe the problem we are trying to solve.

The scientist has to test many samples of superconducting wire. Each batch of new wire requires a battery of tests and measurements to determine the allowable design parameters for fabricating larger magnets using the same wire. These samples are tested at liquid helium temperatures which is around 4 Kelvin (-452 F). The structure that supports the wire during the test is called a probe. If we just took the room temperature probe +452 Kelvin and dropped it into a cryostat full of liquid helium most or all of the liquid helium would turn to gas and be lost to the atmosphere because of the large differential temperature.

To minimize the liquid helium loss we pre-cool the probe structure in liquid nitrogen first to take as much heat out of the probe before it is inserted into the liquid helium. Nitrogen is cheap compared to helium and has a much lower thermal conductivity than helium. Not much nitrogen boils off as gas when a room temperature probe is inserted into a bath of LN (liquid Nitrogen). This pre-cooling gets us down to around 70 degrees Kelvin. Pretty close but not close enough for the wire to become a superconductor.
So the ergonomic problem we are trying to find a workable solution to is the insertion of the pre-cooled probe pictured above into the cryostat containing the liquid helium. In the past one of the technicians would manually lift the probe (~30-40 lbs) and walk up a step ladder carrying the cooled probe out in front of his body, and insert the probe slowly into the liquid helium filled cryostat. To compound the problem as the probe is lowered into the helium we pause every few inches to allow the probe to cool off so we don't lose too much of our expensive helium.
Here is Paul inserting a pre-cooled probe into a cryostat containing liquid helium. You can see the volume of gas coming out of the cryostat as he inserts the probe. This is going from 70 Kelvin to 4 Kelvin. You can get the idea what would happen if we didn't pre-cool.

After kicking around some ideas we came up with a couple of plans. The first was a bit agricultural and consisted of a small self locking block and tackle to support the weight of the probe but still allow the technician to manipulate it easily. This idea had the advantage of expediency and low cost but with less control than we needed.
 The second idea was more involved but a better overall solution. Using a small hand crank shop crane we can extract the probe from the LN pre-cool and then lift the probe and swing it over and lower it in a controlled manner into the helium bath. We have a few of these hoists around that are used to mount large lathe chucks. The models dedicated to the lathes have a block on the bottom that clamps right into an Aloris tool post. Our only problem was we didn't have one like the one in the image with a long vertical tube. All the models we had on hand were the shorty versions of the one above and we need the vertical height for the probe. A few weeks ago I was looking for a box to ship something and I went into a shipping room in another building. Low and behold there was one of the shop cranes we are looking at for our problem with a long vertical tube. I asked If we could borrow it for a test and hauled it back to our building.

Like all good stories this one started off with a minor thud. I bought four inch square aluminum to make the mounting bracket. I figured it was easier to make a smaller square out of a larger square than make a round bar square from scratch. Since I have become a four jaw chuck enthusiast I decided to use the four jaw on the LeBlond lathe in our support shop. There was a four jaw and it looked like most four jaws in most shops. Hardly ever used. No problem. A little cleaning and it will be ready to go. Hey, where is the chuck key for this? Problem number one. So we tear the shop apart looking for the chuck key without joy. I call one of the machinery repair guys and ask if they have a drawer of old chuck keys hiding in their building. No joy again. To Johnnies credit he offered me the chuck key off their lathe as a loaner. Thanks Johnnie.

I finally found one in another building that didn't have a chuck to go with it. I was one click away from ordering another one but decided to check one more place. So now I think I'm ready to go. Opps, chuck key is the wrong size. I had quickly scaled the socket and it looked like half inch. Turns out a bit too quick with the scale. The socket is a seven sixteenths square key. Dang.
 No other quick options than mill the existing chuck key down to fit. Tak took pity on me and milled it for me after hearing me whining about it. When we measured the square socket on the chuck we must have picked the smallest one. After milling all four sides of the chuck key it only fit one of the four sockets. As you can see this simple job is headed downhill pretty fast. Fortunately Tak was able to touch up the sides and corners on the chuck key and get it to fit. Phew, nothing has been done on the actual job yet. This is all prep work.
 The corners were the hangup. If you look at the female socket they have rounded corners. I guess their broach was wearing out when they did this chuck. Once Tak rounded the corners all was good.
Finally I get to make some chips. Four jaws hold like the dickens so I could make up some time once I got the darn blank in the machine.
 Around 600 rpm and .015/rev feed. With 6061 you need to feed aggressively to get the chips to break.
Ready for boring now. 2.560 diameter  x 5.00 deep. Flat bottom to make it more interesting.
I didn't get any pictures of the bore roughing. Since it was deep I stepped the bore out in phases. What I mean is I bored steps to increase the amount of room for chips to come out. For example I drilled a pilot hole large enough for the boring bar to very close to the finish depth. Then I hogged out to a depth of around three inches but increased the diameter to almost two inches before I went any deeper. This allows the chip that are formed deep in the bore a place to go that isn't jammed around the boring bar. You can feed more agressively and use a larger depth of cut with this stepping method. All the material needs to come out so you might as well make it come out smoothly. No rules that you need to keep the bore parallel during roughing.
Finished bore. I couldn't get the camera to focus on the bottom of the bore.For the final cuts I used .005/Rev feed for a smooth finish.
I switched back to the normal six jaw we keep on this machine to do the backside diameter. Face to length then turn up to another shoulder to create the mounting bracket blank.
Finished on the lathe for now. Yes I forgot a dimension on the drawing and had to pencil it in. Most of the mill work was normal operations. The large corner chamfer is of some interest.
The chamfer was not particularly precision so I just used an angle plate clamped to the vise jaw to set the square part at a 45 degree angle. To measure the chamfer I just zeroed the tool on the diameter of the cylindrical part and moved up the required amount. It was an additional dimension that needed to be added to the drawing but easy to extract from the model.
I had Matt, one of the new trainee's help with the drilling and tapping of the block that the bracket will attach to. This is the actual very first job he has ever done on a Bridgeport mill. That's part of the reason I took a few pictures of him on the machine. I wish I had a picture like this from when I first ran a mill. It would probably be printed on stone tablets.....
The mounting block that Matt modified back in position on the cryostat ready for the new bracket.
Looks like the computer model at least. Hope the hoist fits in the hole.
This is the hoist we borrowed from the other shop to test the system. If everything goes correctly we will order a stainless steel version of this hoist. We need stainless not because of corrosion but because of the magnetic field inside the cryostat. Deep inside the liquid helium bath is a 15 Tesla superconducting magnet. Any steel in the area wants to head to south into the magnet. When I say Tesla, I don't mean the cars.

Thanks for looking.

1 comment:

  1. Good content, I trust this is a good weblog about Wish to see refreshing content material next time. Thanks for sharing this publish with us.Battery Operated Ultrasonic Level Sensor