Building a PID melting furnace for under $40

[Big brass ones]

This is my first official thread!

Why so long?

Because I was too busy reading other threads and acquiring knowledge. the more I read here the less I feel I know about casting and guns in general. Thanks guys for keeping me humble.

I had to share my PID experience with others so that they might learn from my mistakes.

I choose to work on my Lee 4-20, because I have seen other peoples work and saw it was nothing new.

My goal was a cheap casting pot with temperature control and little modification to the existing pot should I desire to change is back. This was a bare bones setup and something anybody could do.

I choose a rex-c100 series pid because they are cheap on ebay <$20
A good k theromcouple < $5
A 12a SSR because it was also cheap < $6
A used cpu heatsink (free)
home depot parts 15a outlet, handy box, and a few odds and ends.

Here are the results!

http://i1048.photobucket.com/albums/s375/Cmpezzoni/meltingpot.jpg

I used PVC as the building material for the box so that I could glue the whole think together. the backside has the 15a outlet and box as well as the power cord.

http://i1048.photobucket.com/albums/s375/Cmpezzoni/melting.jpg

http://i1048.photobucket.com/albums/s375/Cmpezzoni/pot.jpg

I mounted the thermocouple inside the pot by drilling a hole next to the pour rod. I can't take credit for this as another fellow caster did this first and it works great! Any other method I tried had poor results compared to this one. I removed the pour rod so that the thermocouple could been seen (notice that it is at an angle). the probe does not touch the rod, but comes close. I used the home depot silver solder and it didn't work too well. I ended up using a low temp brazing rod for aluminum, but brass might work too.


Lessons learned:

Cost was under $40 but you get what you pay for.

The PID was suppose to be either C/F but is only Celsius (I even tried to change the program to accept F data).

I originally ordered thermocouples from china and it took over a month![smilie=b: thank God I bought extras!

The cpu heat sink is a great idea (drilled the heat sink and mounted the ssr)

Update: I forgot I opted for the larger heat sink (unnecessary), but the cpu heat sink works too!

the 12a SSR on ebay was the best deal of all $6 bucks and plenty of power.

In closing:

try the 12a SSR and cpu heatsink.

Don't buy A PID from China unless you like the metric system or know it is a C/F model!

thermocouples should be bought close to home.

[40Super]

Would you be able to post some pics of the thermocouple location and some other angles of the install. I am setting up my system and would prefer to go this route so I can figure what temps work best for me quicker and keep closer records on my alloy melt temp.

[Big brass ones]

new pictures added.

[snuffy]

OMG, get that PID out from under that pot! The base of a lee gets quite warm, but that's not the major concern. Lead dropping/drIpping on that PID controller could short out the electronics if the front is not sealed very well. Heat build up will kill that SSR, you're already drawing around 5.8 amps, you don't need heat to possibly overload that SSR.

[Mal Paso]

A SSR running a 120 Volt Lee 4-20 only generates about 8 watts of heat. Not a lot to worry about. Looks like air will vent past the j-box fine. The base overhangs the PID. You get extra points for stealing the thermocouple idea. I Like It!

[Big brass ones]

yes the shadow is from the base hanging over the pid so no lead can splash onto it. I thought about the problem of the heat dissipation and have yet to have any problems.

8watts!!! I didn't even know it was that much

I also used ceramic insulation around the heating element in the bottom of the pot to help with hold the heat in. This might also explain why it hasn't been a problem yet.

They say imitation is the most sincere form of flattery!

[Mal Paso]

That's 8 watts only when the element is heating. Once the pot reaches temperature the SSR takes a break every time the heater does.

I bought the big aluminum Auber box and just bolted the SSR to the inside back between the air vents. Works fine, doesn't even get warm.

I love turning my pot on, seeing 68 degrees and watch it climb. No more cranking it up to melt, watching like a hawk, then finding out where the Knob needs to be today, for whatever level the pot is full.

[Big brass ones]

I think my next project will be A ultra compact Pid controller with a single plug, switch, and thermocouple socket. Also considering a computer power supply as the box, and adding on to it. I'll also use a larger ssr and heatsink maybe even a fan so that I can use it for heat treating and warming up molds.

[40Super]

Does anybody know of a way to hook up to a laptop, like have the laptop as the control? I'd still need a relay so the power is taken from a seperate power source ,but rather than the PID,the laptop could control the temp(and hopefully not get molten lead on it).

[Big brass ones]

Yes I do know of a way to do that, but a software engineer would need to write the program. you could use the USB drive to power the SSR, but a more effective way would be to use a pic micro-controller to drive SSR as well manage the thermocouple. Being a flash based microprocessor they are cheap, and do a great job for applications like this.


The Program on the other hand would be rather complex being all three part equation

http://upload.wikimedia.org/wikipedia/commons/thumb/4/43/PID_en.svg/300px-PID_en.svg.png

you can see by the diagram that one would require an extensive knowledge of calculus, more specifically differential equations, and maybe even thermodynamics and that's not even the programming:sad:

I'd go into more depth but at the risk of sounding like Scotty on Startrek i'll take a pass.

save yourself the :killingpc and buy a PID.

[Big brass ones]

I'm sorry if my last post was a little know it all sounding. Please excuse the enthusiasm in this topic as I gave many good years in college to this very subject.

[nanuk]

and I can't even spell PID!!!!!

[hunter64]

I think my next project will be A ultra compact Pid controller with a single plug, switch, and thermocouple socket. Also considering a computer power supply as the box, and adding on to it. I'll also use a larger ssr and heatsink maybe even a fan so that I can use it for heat treating and warming up molds.

Did that.

Here is the Power Supply with the lid off.
http://img338.imageshack.us/img338/3976/picture004sc.th.jpg (http://img338.imageshack.us/i/picture004sc.jpg/)

With the guts out.
http://img31.imageshack.us/img31/4013/picture010lw.th.jpg (http://img31.imageshack.us/i/picture010lw.jpg/)

My Old Cell Phone Charger.
http://img30.imageshack.us/img30/2900/picture006ca.th.jpg (http://img30.imageshack.us/i/picture006ca.jpg/)

Cell Phone Charger with top off. Two wires on right were 110v and left 5v DC.
http://img99.imageshack.us/img99/4568/picture008lz.th.jpg (http://img99.imageshack.us/i/picture008lz.jpg/)

Side view. You can see the power supply on/off switch and plug in for the computer power cord.
http://img443.imageshack.us/img443/3051/picture012g.th.jpg (http://img443.imageshack.us/i/picture012g.jpg/)

Top View. On the left you can see the fuse holder and in the middle is the connection block.
http://img3.imageshack.us/img3/3280/picture013ka.th.jpg (http://img3.imageshack.us/i/picture013ka.jpg/)

Right side view. On the left is the PID controller, below it is the tie wrapped in power supply and on the right is the SSR.
http://img8.imageshack.us/img8/8759/picture014tx.th.jpg (http://img8.imageshack.us/i/picture014tx.jpg/)

Front view. Top left is where you plug in the Lead pot/Lube sizer etc. Below it is the thermocouple connection box and to the right is the PID.
http://img84.imageshack.us/img84/9483/picture016xp.th.jpg (http://img84.imageshack.us/i/picture016xp.jpg/)

Since this picture I have changed the cell phone charger to a 12v version and changed the small SSR to a big 25A version but it still fits in the box and works like a charm.

[40Super]

I actually do have the "means" to get the program written, I was looking to see if anyone did do it to save myself a lot of time and searching for the parts.

Hunter: I am looking to not even use a PID. I am more interested in making it way too complicated so later I can say,"Why the H**L did I do that."

[felix]

For those seriously interested, this is a good description of the PID functionality and mapping a design to the application. ... felix

http://ece.ut.ac.ir/classpages/S84/Mechatronics/PID%20Controller/pid.pdf

[Big brass ones]

Nice work Hunter!

Felix that is perfect! you must be a ham radio guy! I owe you a slide-rule! :kidding:

[popper]

40Super - Lots of code written and available for the PID - but then you get to get the code for the LED and buttons and thermocouple. If you use a laptop, you get to write code for the USB , printer port or serial port and build an interface.

[40Super]

I'm thinking a hvac control program can be used ,just would have to get it to go through the usb or ethernet port.It wouldn't be a given though, just doing the thinking part right now and picking some brains.:coffee:

[Frozone]

The easiest control port I can think of on a lap top is the USB.
Check out the USB to RS232 converter chips that provide a virtual RS232 port. They run ~ $10 complete.

They will provide 4 lines you can control from the windows op system Outputs are DTR (you can use this to steal power too.) and RTS. You can Poll DSR and CTS states as inputs.
This leaves Sin and Sout as null modem.
The biggest problem with using a PC is how to read the thermocouple. I don't think there is an ADC available.
You would still have to use a PIC to read the TC and convert it to a number then transfer that number back to the PC. But that is not hard and the code is simple. The ~$1 8 pin PIC would do the job nicely.
I'm sure you could steal enough power from the USB to run everything fine.

So for hardware you're looking at a USB/RS232 chip, a PIC, and a SSR. pretty cheap, < $25 I'd guess. Of course you still need to program/burn the PIC, Pic burners are in the $200 range ;-)

USB/RS232 (http://apple.clickandbuild.com/cnb/shop/ftdichip?productID=93&op=catalogue-product_info-null&prodCategoryID=83)
PIC (http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_223790_-1)

[nanuk]

The easiest control port I can think of on a lap top is the USB.
Check out the USB to RS232 converter chips that provide a virtual RS232 port. They run ~ $10 complete.

They will provide 4 lines you can control from the windows op system Outputs are DTR (you can use this to steal power too.) and RTS. You can Poll DSR and CTS states as inputs.
This leaves Sin and Sout as null modem.
The biggest problem with using a PC is how to read the thermocouple. I don't think there is an ADC available.
You would still have to use a PIC to read the TC and convert it to a number then transfer that number back to the PC. But that is not hard and the code is simple. The ~$1 8 pin PIC would do the job nicely.
I'm sure you could steal enough power from the USB to run everything fine.

So for hardware you're looking at a USB/RS232 chip, a PIC, and a SSR. pretty cheap, < $25 I'd guess. Of course you still need to program/burn the PIC, Pic burners are in the $200 range ;-)

USB/RS232 (http://apple.clickandbuild.com/cnb/shop/ftdichip?productID=93&op=catalogue-product_info-null&prodCategoryID=83)
PIC (http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_223790_-1)


How do you spell PID again???

thanks...

Wow... I have absolutely no idea what you just said!

I think I need to buy a book!

[Frozone]

How do you spell PID again???

I spelled it right. I wasn't talking about a PID.


thanks...

Wow... I have absolutely no idea what you just said!

I think I need to buy a book!

A PIC is Not a PID , however, you can make one from it. [smilie=1:
The thing you can't do is make a PIC from a PID.

Now MY head hurts! :roll:

PIC = Peripheral Interface Controller

PID = proportional–integral–derivative <Controller>

If you're really hungry I have lots of alphabet soup.

[AR-15 Cowboy]

I'm building a PID for my use but I'm following a different logic. To me the PID should be the primary machine and the Lee 4 considered the throwaway. This is why I wont make my TC permanently attached. Or wire my PID permanently to the Lee. That way when the pot starts to leak too much, or burns out, I can just slip out the thermacouple, unplug it from the PID and throw it away. Buy a new one slip the TC in and plug into the PID and I'm in business again. Pot $60 PID $130. Do the math. Also I can keep the PID 3 or 4 feet away from the pot and avoid accidents.

[Mal Paso]

I somewhat agree but half the advantage for me is an installed probe so I can just turn it on. I leave it full now, ready to go. I brazed a couple $7 probes into each of my 4-20s and spent a couple bucks on plugs so there's no wiring to switch pots. So I guess I got 2 $70 pots and a PID.

[Frozone]

I'm building a PID for my use but I'm following a different logic. To me the PID should be the primary machine and the Lee 4 considered the throwaway.

I have yet to see the pot on a lee wear out.
The heater is replaceable as is the spout and both are dirt cheap.
Your logic is flawed.
The heating elements do burn out - usually in 20+ years.
If the pot wasn't dripping when new, if it starts dripping it has crud in the spout.
Do you buy a new car when yours gets dirty or do you clean it?

I prefer my sensor as a bottom mount but I have multiple pots. It's easier to just plug a different pot in then move the probe.
Immersion probes are easy to knock out of the pot and, if installed, get in the way during storage ect.

[AR-15 Cowboy]

"Your logic is flawed.
The heating elements do burn out - usually in 20+ years.
If the pot wasn't dripping when new, if it starts dripping it has crud in the spout.
Do you buy a new car when yours gets dirty or do you clean it?"

I wasn't talking about cleaning the furnace, I was talking about a major failure. How many people do you know got rid of their car when the engine blows up and their was no warranty. It depends how much you use the furnace that determines how long it will last, and also how good you take of it. My logic was about the money. One costs twice as much as the others. I only know one guy who's furnace quit and it was after three years. He sent for a new element but before it came he wound up buying a new one locally. He decided to keep and repair the old one just for smelting which is what trashed his in first one.

[Frozone]

The more you write the farther up the scale my BSometer goes.

The Pot itself will not fail unless you let it sit in the rain and rust it out.
Service from LEE is about a week maybe 2 for parts.

If your buddy had to spend $70 for a new pot instead of waiting a week for a $10 part - well we'll just let that go.

Again: if it leaks it's dirty. if it doesn't heat change the element the pot is good for 50+ years, unless you run over it with a truck. It'll likely still work then.

LEE hating is in vogue apparently, so you consider the pot as a disposable item - I consider your opinion the same .

[Roger Ronas]

Next pot that you want to throw away, I'd like to be on the receiving end please.
Just saying,
Roger



I'm building a PID for my use but I'm following a different logic. To me the PID should be the primary machine and the Lee 4 considered the throwaway. This is why I wont make my TC permanently attached. Or wire my PID permanently to the Lee. That way when the pot starts to leak too much, or burns out, I can just slip out the thermacouple, unplug it from the PID and throw it away. Buy a new one slip the TC in and plug into the PID and I'm in business again. Pot $60 PID $130. Do the math. Also I can keep the PID 3 or 4 feet away from the pot and avoid accidents.

[AR-15 Cowboy]

The more you write the farther up the scale my BSometer goes.

The Pot itself will not fail unless you let it sit in the rain and rust it out.
Service from LEE is about a week maybe 2 for parts.

If your buddy had to spend $70 for a new pot instead of waiting a week for a $10 part - well we'll just let that go.

Again: if it leaks it's dirty. if it doesn't heat change the element the pot is good for 50+ years, unless you run over it with a truck. It'll likely still work then.

LEE hating is in vogue apparently, so you consider the pot as a disposable item - I consider your opinion the same .

The more you write the more I think you are lurking around looking for someone to argue with. I'm not a Lee hater. I think they are a fine company. I buy most of my reloading stuff through them. And either I'm communicating it poorly or you are not grasping the point I was trying to make. The point is not that the Lee furnace is bad, but that the PID controller is the more expensive item of the two. The PID will actually improve the operation of the furnace and extend its life even further. A bit of advice, if you disagree with someones opinion, you don't have to try and run them down or belittle their opinion, just state your facts.

[Mooseman]

I am watching this thread...

Rich

[Rockchucker]

Thread Highjack

Hey Rich, Days getting longer up north, Wish I was there FISHING! Have a good one.:)

[yobohadi]

Hey Big Brass Ones,

How well does your budget set up work? You mentioned cost and the "get what you pay for", but once you do the F to C calculations to your desired temp is it doing well to keep the alloy at that temp or have you seen some big fluctuations in temp with the cheaper components?

Oh, and are you happy with the setup now or would you pay the extra to get the PID and components from Auberins if you had to do over again?

[Big brass ones]

The PID does what it's suppose to so in that sense it works fine. I just convert to metric units which is what most of the world uses anyway!

this pot it amazing now. I really like lee products when there work as advertised.

Update! silver soldering does not work out too well in this pot. Lead collects around it so it can be hard to completely clean the pot. Next time I'll brass braze it in there.

[Mal Paso]

I worried the brass melt point would cook the probe. I built up a button of brass on the bottom of the pot to add thickness, drilled an angled hole, and silver soldered from the inside. The silver solder wicked into the joint just like copper pipe. Next time I'll silver solder from the bottom and just wet the joint. No puddle on the inside.

[Big brass ones]

A fellow on this site said the k thermocouple that i used could take up to 1800 f which is well above the melting point for brass. Mileage may vary!

[Mal Paso]

Depending on Alloy Brass melts at 1652-1724 degrees F. It was a little too close for me.

[Big brass ones]

Good point. that's too hard to control with a torch. For some reason I thought it was lower (1400's). I used something for brazing aluminum (couldn't tell you the exact alloy) and it worked fine.

[Dave Bob]

A fellow on this site said the k thermocouple that i used could take up to 1800 f which is well above the melting point for brass. Mileage may vary!

What thermocouple did you use? Link?

Thanks,
Dave