thermocouple mounting???

[cutter_spc]

I would like to mount my thermocouple, for my new PID project, though the bottom of my Lee 20 pounder. I have a K type like this one

http://www.auberins.com/index.php?ma...8744cf47b7d39d

How are you guys mounting this through the pot? I see no way of making this sensor leak proof, as the threaded furle is seperate from the sensor itself.

Can anyone help me on this?

Thanks, cutter

[calaloo]

Here is what I did. The end of the thermocouple is simply twisted together and fused. I cut the chrome plated brass bolt of the wire. I drilled out a 1/4 X 1/2 stainless bolt. Drill most of the way yhrough with a larger bit but the last little bit should be drilled so the wire is a fairly tight fit. Let the wire barely protrude past the bolt and silver solder it in place. I made a washer from copper to use as a seal and have had no problems.

Now, here is what I would do if I were to do it again. I would cut the bolt off the end, twist and fuse, then silver solder the thermocouple to the outside of the pot. Much easier to do and no chance of a leak.

[cutter_spc]

calaloo,

It sounds like you are making it into the cheap bead type thermocouple, is this correct? How are you fusing them together, silver solder?

Thanks, cutter

[theperfessor]

I use a probe type for my pot. I just made a bracket for the top of my Lyman pot with a hole in it and drop it through and down into the pot. That way I can remove it easier when I want to flux and skim, etc. No chance of a pot leak either. The screw in types I use for my molds to control hot plate.

[calaloo]

All a thermocouple is is two specific alloy wires fused together. If you want to call it cheap that's OK with me. The wire is fused not soldered. Twist the wires together and use a torch to fuse them. Silver solder the fused wire to the outside of the pot. There is nothing in your way when you cast, flux or whatever. Simple, elegant, and cheap.

[rtracy2001]

Why not drill a hole through the bottom housing (not the pot itself) and use a couple of jam nuts to adjust the tip of the thermocouple to be in contact with the bottom of the pot? Should get a pretty good reading that way with no leaking.

[cutter_spc]

perfessor, I could go that rout, but I really like the idea of having the top of my pot wide open. I read a post of yours a while back about controlling the hotplate via PID, I plan on doing the same thing. Are you drilling and threading your molds? Have you done iron molds?

calaloo, when I said cheap, I really meant simple, wasn't trying to offend. I looked up fusing thermocouples and it seems simple enough, there is even a "pop" method that just requires arking the wires. You have nailed what I am looking for, nothing in the way, simple, elegant, and cheap!

rtracy, hadn't thought about that, I'll have to check to see if there is enough clearance.

Thanks for the suggestions guys, cutter

[cutter_spc]

Well I went with rtracy's suggestion and jam nutted the thermocouple through the bottom cover of the pot with the tip of the thermocouple resting very snugly against the pot itself. I plugged the pot in regular first until it got up to about 400, then I ran it through the PID. Right away I saw I had a small problem, the PID was reading about 200* cooler than my Tel-True. The good news is it seem the be pretty constant, where ever I set the the PID the pot was running about 200* hotter. I can live with that as long as it's consistent. The more good news is, it's an absolute dream to cast with this set up. I can't explain how much more enjoyable this makes casting for me. Set the PID and get to casting and not having to worry about the pot temp allows me to concentrate more on other aspects of the game.

Thanks for the help guys, cutter

[lurch]

Cutter,

I went through a similar situation with the thermocouple mounting on my Lee 4-20. If you look at what is happening with the jam nuts on the bottom cover you will find that you have made a heat sink for the thermocouple.

As I envision what you are doing, it looks like this:

• The thermocouple junction is located in the little nib at the end of the bolt.
• The nib is held in contact with the bottom of the pot by the bottom cover.
• The bottom cover is tightly mechanically (and thermally) connected to the mounting bolt by the jam nuts.

There are a couple of things here that are conspiring against you:

• The thermocouple is not really very tightly pressed against the pot crucible. The bottom cover is simply not strong enough to exert much force. This results in a fairly weak or high thermal impedance connection.
• The tight coupling of the mounting bolt to the bottom cover is actually a pretty good thermal connection and can draw heat away from the mounting bolt.
• Since the thermal connection to the pot crucible is not really that good the heat being drawn away by the bottom cover results in the thermocouple junction temperature* being significantly lower that the melt temperature, resulting in the low reading on the controller. The more heat that is drawn off (or rather through the whole assembly) and the higher the thermal impedance is from the lead to the actual junction, the lower the reading will be.

I thought about this a lot after discovering the same thing when I did mine. How I resolved it was to thoroughly insulate the thermocouple mounting hardware from the ambient air. I never tried the jam nuts on the bottom cover solution as I was pretty sure I could not get sufficient mounting force to make good thermal contact that way. What I did was drill a hole in a piece of 1/8" flat stock approximately 1 in square that the thermocouple mounting bolt would pass through. I then welded a 1/4-20 nut to this flat stock. The flat stock was then very carefully tack welded to the crucible. This allowed the thermocouple to be tightly held against the crucible and have as low as possible thermal impedance without permanently attaching the thermocouple to the pot. I put a clearance hole in the bottom cover for access to the nut when the pot was put back together. What got me in trouble was that the nut and flat stock were tall enough that they were bearing against the cover of the assembled pot. Even though I had a lot of mounting force, there was enough heat draw away from the thermocouple junction that I was seeing a significant difference in temperature readings between the PID controller and a standard casting thermometer. I wound up enlarging the hole in the bottom cover so that there was no contact with my thermocouple mounting assembly and then putting some insulation around the thermocouple itself. This cut the heat draw away from the thermocouple to a point where the PID controller and the casting thermometer had readings that while not exact, were at least in the same zip code.

Silver solder as mentioned will work if you keep close watch on the pot temperature. It would certainly be less work and provide a very good thermal interface that would probably not need much if any extra insulation. I would think it would be possible (probably not advisable though...) to get the pot hot enough to melt the solder while . If that happens and there is no secondary control stop for the temperature, the pot will go "open loop" and get as hot as it can possibly get. I left the Lee "so-called thermostat" in circuit in my pot for just such eventualities.

Of course, if you are happy with what you have then by all means use it. If it get you consistent results and you know how to "fudge the system" then that's really all that matters in the end. Neither the boolits nor the targets will care if the PID controller was reading the actual melt temperature or not...



* It's not technically the junction temperature that make a thermocouple work, but rather the temperature gradient along a wire causing a corresponding voltage gradient - the Seebeck effect. That's a different discussion though and for practical purposes the actual junction temperature will suffice here.