External Pulser Circuit for Miller 180SD TIG welder The pulser control circuit is a modified version of a PWM control, found on several web sites, credited to G. Forrest Cook (http://www.homepower.com/files/pwmhp75.pdf). The output section is modified from an LM555 pulser circuit built by Gary Coffman, and described in a posting on sci.engr.joining.welding. The control parameters were generalized form posts on sci.engr.joining.welding by Ernie Leimkuhler, and an email discussion with Ray Sommer. The circuit has the following features: External to the welder. Removable. Self powered. Pulse rate from 20Hz to 0.5Hz (0.05 Sec. to 2 Sec). Pulse width from 0% to 100% of the pulse interval. Background current: 100% to 10% of the pedal setting. The pulser circuit is powered from 20vdc the welder provides to the foot pedal’s On/Off switch. This voltage is mildly filtered (~3Vpp ripple) and unregulated. The pulser converts this to a filtered, regulated +15vdc via U1 to bring the 20vdc down to the DC-DC converters input spec’s of 14-16vdc. A triangle function generator (U2’s first 2 op-amps) creates the pulse rate. The rate is set by C2 and the potentiometer R16 + resistor R5. R5 determines the fastest rate limit. R16+R5 determines the slowest rate limit. The pulse width modulator (U2’s last 2 op-amps) sets the pulse width percentage. The actual pulse width will vary in direct proportion with the pulse interval. The trim pots (R8 and R10) and U2’s third op-amp adjust the amplitude and offset of the triangle signal This insures that potentiometer R17’s range is 0% to 100% of the pulse interval. The last op-amp creates the PWM and drives Q1, and LEDs D1 and D2. 0% Time would be the same as stopping the pulser. Also 100% Current would do the same.The red (full power) and green (background) LED’s let you visualize the pulse rate and duty cycle before you weld. The background current is set by a potentiometer R18, paralleled by a P-channel MOSFET (Q1). These are placed in series with the foot control’s pot. When U2's last op-amp turns on Q1, R18 is shorted out, and the welder gets a control voltage determined only by its front panel setting and the pedal’s pot position. This would be the high current part of the pulse. When Q1 is turned off, R18 is placed in series with the pedal’s pot. The welder now gets a control voltage determined by the front panel setting, and a ratio of R18, the pedal’s pot, and it’s wiper position. This is the low current part of the pulse. This is not a linear function so a log-taper pot was used to more closely approximate a linear % background current vs. pot rotation. Several approaches could have been used to add the pulser to the welder. One method would be to cut the foot pedal cable, near the welder, and add a breakout box. A cable from the pulser box could connect to this box. A SPDT switch could be added to the breakout box to bypass the pulser. Another method would be to run a cable from the pulser box directly into the foot pedal. This would require only minor changes to the pedal wiring. This is probably the cheapest, but the least convenient in operation. The best method is to install connectors, on the pulser box, that match the torch and welder’s connectors. This is the least invasive, but most expensive. Since I’ve been able to locate the correct AMP/Tyco connectors, at 1/3rd the price that Miller asks for them as repair parts, I chose this approach. The schematic layout was in ExpressPCB, a free PCB CAD program. Since there were many revisions, I only used the PCB portion at first to get general layout ideas for point-to-point wiring on Radio Shack perf- boards from. In the end I did have a PCB made for the final circuit. High frequency arc noise necessitated the ferrite beads and bypass capacitors on the wires running into the pulser box. The large number of header connectors, P1/J1 thru P7/J7, and the separate power supply board were to allow testing and alterations during the development of the circuit. Adjustments of R9 and R10 are performed with an ohmmeter and an oscilloscope. This can be done with the pulser connected to the welder or with a 20vdc bench supply. The first step is to disconnect R17. With the ohmmeter, measure the total resistance of R17. Now connect the meter between one end of the pot and the wiper. Rotate the shaft until the meter reads exactly one half of the total resistance. Reposition the knob, without moving the shaft, to the 50% mark on the front panel. Reconnect R17. Connect the oscilloscope to chassis ground and pin 14 of IC1. Adjust R9 and R10 to the approximate midpoints of their mechanical range. Set the sweep and the pulser’s rate control to display one complete high/low cycle. Adjust R9 to make the high and low intervals the same length. Turn R17’s knob to the 0% position. Adjust R10 to a point where the low pulse just disappears from the scope. Turn R17’s knob to the 100% position. The high pulse should just disappear at 0%. If you don’t get the low and high pulses to stop at the 0% and 100% marks, you may need to make a very slight adjustment of R9 and then R10. Repeat this until you get the correct response. I made my 0% and 100% panel markings at slightly less than the mechanical stops on R17, to allow for a little over travel. I wanted to make sure that the pulses stopped at the 0% mark. The AMP/Tyco Connectors are available from Electronic Industries Inc. P.O. Box 266 Oshkosh WI 54903-0266 (920) 235-8930 http://www.electronicind.com Part numbers: 1 each: AP-213571-2…Plug 1 each: AP213570-1…Receptacle 1 each: AP206070-1…Plug Shell w/Cable Clamp 10 each: AP-213603-2…Pin, 18-14, Tin, 10 each: AP-66601-2…Socket, 18-14, Gold Total (w/ shipping) $27.82 I stopped making changes when I got it to a working form. If I were making another I would make several more modifications. 1. Eliminate R1 &R2 and tie U2's pin 1 to the +/- power supply's commom. 2 Eliminate C3, C4, C5, ,and U1. 3. Replace the DC-DC converter with one that will handle the 20vdc from the welder and has a regulated +/-12v @ >40mA output. November 14, 2006 Ken Moffett Scientific Instrumentation Macalester College 1600 Grand Avenue St Paul MN 55105 651-696-6383 moffett@macalester.edu