A PEEK AT ALL THE FILES INCLUDED TO BUILD THIS WELDER

THE DOCUMENT BELOW  SHOWS WHAT ARE  INCLUDED IN THE DOWNLOADABLE  PACKAGE.
UPDATE: THE MOSFET PCB OUTLINE FILE (.GKO)CAN NOW BE DOWNLOADED HERE.

ADJUSTING PULSE 1

Pulse 1 on this welder is to burn away oil and dirt and pulse two makes the actual weld. I have many guys telling me that they set pulse one much higher to get two welds in one. Unfortunately, that's not how it works. Pulse one will join the two metals lowering the resistance and the energy of pulse two will take the shortest path to the other electrode doing basically nothing. In other words, If a 3ms pulse is enough to adhere the metals being welded and you set pulse one to 3ms and pulse two to 12ms then you will get one weak weld formed by the 3ms pulse and not the 12ms pulse.

To make decent welds with a dual pulse welder you should turn pulse one off and find a voltage and pulse 2 setting that gives you  good solid welds. Leave the welder on the determined  voltage and set both pulse potentiometers to 0, adjust pulse one to a low setting and make a weld. Keep on adjusting pulse one higher until the metals starts to adhere, decrease pulse one potentiometer about 3%. Set pulse two to the determined setting above and you should get really good welds.

The size of your electrodes and cable to your electrodes makes a big difference. A 4 feet long #0 welding cable will give you twice the current than a 4 Feet #8 welding cable. Do not make your cables shorter than 4 feet though.

UPDATED FIRMWARE AND WELDER FILES AVAILABLE SOON

I've been working on a new version 2.5 firmware for the Capacitor Discharge Welder and also a package that will contain the newest firmware, gerber files, Bill of materials, manuals and an extra 6 pages to explain how the circuit works and trouble shooting procedures etc. I've included Version 2.1 and version 2.5 in case a bug shows up. Version 2.5 has a change to the charging system, taking the stress off mosfet Q4, slowing the charge intervals and two points on the control board can be jumped with a wire to change the maximum voltage of this welder from 20 to 23.2V. This will increase the energy of the welder from 600Ws to  807Ws. You will need 25V capacitors to use this feature and Diode D10 has to be changed. UPDATE: I'm busy doing some final tests on the firmware. I did not get a very big response about making everything available so I'm going to start by making the updated chips available

If you don't know how to program a processor, you can buy one of these programmers.

 This one is $34.95  : K8076 serial programmer   

 This one is $69.95 : Cana USB programmer

It is as easy as starting the software supplied with the programmer, plug your microprocessor into the programmer, connect the programmer to your computer, load the Hex file and click on "write". It will only take a couple of seconds.

NAVIGATION

HERE IS AN EASIER WAY TO NAVIGATE THIS BLOG, click on the links below and read the pages from the bottom, up:

 THE BLOG STARTS HERE: PAGE 1 
 PAGE 2
 PAGE 3
 PAGE 4    
 PAGE 5 
 PAGE 6 
 THIS PAGE: PAGE 7

Pulse Arc Update

My function generator which came a long way with me went up in a plume of smoke. I had to replace it and bought a Rigol DG1022. It's a nice generator but pricey. I am going to use this generator to play with different waves to see if I can get the correct waveforms for the pulse arc welder. I am thinking of designing a new board from scratch, its already halfway drawn on a piece of paper :). The Budget is also a little low on funds and I will start to buy components as soon as possible. This welder will use much smaller and easy to get capacitors.
These things take time and lots of work so don't expect to see it before the end of this year.

ROGER CARR LET ME KNOW THAT FUTURE ELECTRONICS SELLS THE IRFP2907PBF MOSFETS FOR $3.60, THANKS ROGER.UPDATE: FUTURE ELECTRONICS CHANGED THEIR PRICE TO $6.00.....DO NOT BUY FROM THEM!!!!

HAPPY NEW 2011 EVERYONE

PCB"S ON ORDER

PCB's ARE SOLD OUT 02/05/2011

PULSE ARC WELDS

The top three photos was taken when I just started playing with the pulses on the pulse arc welder. The top left photo is two washers welded together,  the middle is two strands from a 12 gauge copper wire welded to a washer and the photo on the right is two strands of a 12 gauge wire that were held against the tungsten electrode while firing a 45ms pulse. The bottom two photos were taken after I managed to manipulate and change the pulses to give better welds. The plasma flame produced here is extremely hot.

I have made thousands of welds and adjustments to get this far and will keep on going until I get it to work perfectly.  I  accidentally made a plasma cutter that cuts through metal when I made the pulses continuous and also managed to weld  a copper wire to aluminum.
I'm limited to the metals I have to do further tests however and will appreciate if someone with knowledge of these welders can help me with some advice. Email me at sprbok@gmail.com

PROCESSOR CONTROLLED PWM RELAY DRIVER SCHEMATIC

 

When a magnetic solenoid/relay is energized, a magnetic field has to be formed strong enough to pull the plunger in. When this plunger is pulled in, the power can be reduced to a much lower level to keep it in.

 This really cool circuit can do the above. If you pull a line, straight down, just before U5 and then build the circuit on the left, you will have a pulse width controlled relay driver circuit that can handle two relays with a total current of 1.5 amps. There are DRV103 IC's that can handle 3 Amps and by changing your power supply to 3 amps, can pull in really big relays/solenoids. 

These PWM relay drivers can be activated by supplying 3 to 5V on pin  8 of U2 and U3. The two red LED's  will turn on when an over current or over temperature is detected and will shut the IC's down. 

This circuit can be changed to drive 12V relays by changing R1 and R2 resistors on the power supply. 

When 3 to 5V is supplied to pin 8 of one of the IC's It will turn on fully, supplying 24V to the relay/solenoid and  22milliseconds later (C4 andC5) will start to pulse the voltage at 5000Hz (R3 and R4) and the duty cycle can be changed by adjusting R5 and R6. 

R3 and R4 can be changed to get a different frequency, C4 and C5 can be changed if the relay/solenoid needs more time to pull in before a pulsed signal is supplied, just check out the data sheets.

How to adjust:

Put a voltage of 3 to 5V on pin 8 of the DRV103 IC. When the relay pull in, adjust R5 or R6 until the relay fall out. Set it back a little and energize it again to see if it stays in. If the relay doesn't want to pull in at all then replace C4 and/or C5 with bigger capacitors.

THE GERBER FILES FOR THIS BOARD CAN BE DOWNLOADED HERE

 

THE REAL THING

I have tested this board and it works as designed

PULSE ARC WELDER PCB BOARD

This is the prototype board, 4 X 4 inches, that will be driving the Argon and electrode solenoids of the pulse arc stylus.
The white block, printed on the board,  in the bottom left corner is the 24VDC power supply and the block at the top left corner is the PWM relay drivers. The area on the rights is the microprocessor circuit.
The microprocessor is going to do a lot of things. It will determine if the electrode is touching the work piece, activate the argon solenoid, calculate the time when to give a pulse to the PWM relay driver circuit to retract the electrode, produce a couple of bleeps and give a pulse to start a pulse arc weld.
This processor will also talk to the other processor in the capacitor discharge welder to make energy pulses possible and much much more.
The PWM relay driver will supply 24V to the relay/solenoid and 22Ms (adjustable) later will give a pulsed adjustable duty cycle. This will reduce the current, preventing the solenoids and electronic driving circuits from overheating.
I'm waiting for the components and boards at the moment and will start doing some test welds soon.