PDA

View Full Version : Spot Welder Timer



RobDee
07-16-2006, 11:56 PM
I have one of those Hobart/Miller spot welders. There isn’t a lot to how it works and when I first used it I could tell immediately that timing is the key to good welds.

I ran a simple timer circuit with an LM555 chip figuring I didn’t need anything especially sophisticated just something that would give me a consistent time for each weld once I got the time right on a test piece.

Not true! One weld would come out perfect and the next burn through. I gave it some thought and figured that, while, my timing was accurate it wasn’t locked on to the pulses from the power company, sixty per second in my country. So what I needed wasn’t a timer but a pulse counter.

How the circuit works:

The circuit is designed around an Atmel uC that saves a lot of components. The chip sets a display to an arbitrary number from a rotary encoder, selected by the user. As the encoder is turned it increments or decrements the display from 1 to 999. Once the number of pulses is set one can merely push a button to count down the mains frequency. One more thing. A sine wave, what comes out of your wall outlet sixty or fifty times a second, is only half the story. The welder sees every pulse and that means two pulses per cycle or one hundred twenty a second.

The case is from a 3/16” thick 2” wide piece of aluminum. (See my article in the Mar/April 2004 HSM to find out how I build them.)

How does it work? I can weld a piece of aluminum flashing .010” thick to a piece of stainless .040” thick without any burn through.

Here’s a photo and the schematic.

http://www.dsgnspec.com/Hobart_uC.html

By the way I’ll be using the same uC program and display for a coil winder I’m building later on this year.

Rob Dee

CCWKen
07-17-2006, 12:08 AM
Wow, you must be welding some thin stuff to be counting cycles. I use mine on 20 & 18ga. CRS most of the and get by with a jimmied up HB3A timer. Have you tried to regulate the voltage too? Longer tongs will also lower heat (amps).

Forgot to mention: Your clamping force has a lot to do with it too. The clamp on these is not the most repeatable. ;)

RobDee
07-17-2006, 12:17 AM
I've found that with one half cycle accuracy I can do just about anything I want without changing the current.

I do a lot of thin stainless and aluminum. You have to see every weld the same as the previous in stainless. It's poetry and tong force becomes less of an issue.

Rob Dee

RobDee
07-17-2006, 12:19 AM
ps. I thought about setting up the tongs with a pressure measuring system but I don't believe I need it at this point.

CCWKen
07-17-2006, 09:08 PM
PS. Your link shows up as a blank page with X's where the pictures should be. :eek:

My applications don't require that much accuracy. I get by with 1/10 second increments. The pressure cell sounds interesting.

smagovic
07-17-2006, 10:14 PM
I have the same problem, welder and the schematic does not show, only empty rectangles. Vic Smagovic

RobDee
07-18-2006, 12:28 PM
I think it might work now. I'm a little rusty at HTML as I haven't done it for awhile and i haven't tended to my web page.

If I was going to do a pressure sensor I would use a simple spring set up where the pressure adj. mechanism is now.

Rob Dee

smagovic
07-18-2006, 01:00 PM
Are you selling a kit for this? Vic

RobDee
07-18-2006, 01:39 PM
Perhaps if there is enough interest and my board house doesn't kill me for circuit boards. I never thought about a kit for it. I wonder how many of these things are around?
My opinion, once you use a circuit like this it changes what you can do, especially detailed work.

I could probably run out programmed chips if individuals were handy enough to do their own boards.

If anyone wants the PCB layout let me know. I think Mouser has all the parts, nothing difficult to get.

Here's my site:

www.dsgnspec.com.

Send me a note if you like.

Rob Dee

RobDee
07-18-2006, 05:18 PM
Here are some welds with the new circuit for those of you who have enquired further.

The first weld is 0.010" flashing welded to 0.032" stainless.

The second is the same flashing welded to 0.075 aluminum plate.


http://www.dsgnspec.com/Hobart_uC.html

Rob Dee

CCWKen
07-18-2006, 08:05 PM
Your site came back inactive. :eek:

RobDee
07-18-2006, 08:25 PM
Server is down.

Mercury in retrograde!

Rob Dee

http://i95.photobucket.com/albums/l122/RobDee50/Spot2.jpg

http://i95.photobucket.com/albums/l122/RobDee50/Spot1.jpg

Dumped on to bucket.