Flipping terminals help

Discussion in 'The Projects Forum' started by DC_Kid, Apr 24, 2014.

  1. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    i have an odd project and need some advice on how to tackle this problem.

    this is for a welding device, a inverter type TIG machine. mine is a tad older and does not have a pulsing feature. but given the controls built-in i can make a controller interface that can give me pulsing capability.

    the foot pedal controller is a switch with a 3-wire pot. the pot is 10k-2watt with center wipe. i dont know how these 3 wires terminate in the welding machine, but i do know how the pot works in the pedal.

    the switch is a on/off switch. when you press foot pedal the switch closes and the unit starts at lowest current setting, the wiper on pot is on one side of the pot, and when pedal is fully depressed the wiper is on other side of pot.

    to make my pulsing work i need a way to flip the wiper from one side to the other using PWM. this wont work using physical pot, etc, so whats the best way?

    example, pot connections A-B-C where B is wiper. in lowest pedal setting A-B is zero ohms and B-C is 10k ohms. at full pedal A-B is 10k ohms and B-C is zero ohms, thus a flip of the wiper from one side to the other.

    currently, the unknowns are current vector and voltages seen on the pot.

    for PWM i am figuring it will be adjustable 1-5Hz and 10-90% duty cycle

    i would like to use dial pots for the PWM control for fast adjustments as needed. no display needed, etc. for flipping the wiper i would like it to be 100% non-mechanical.


    I was thinking maybe two small solid state relays with one side of switched side tied to one side of other relay (the wiper), and then a 10k tied between the other ends of relay, then flip on/off the relays (one on while other off, then flip it to move the wiper)...?

    any suggestions?

    the attached pic is for example only, I would need to choose the right ssr, etc.
    I can flip the opto's with 555, but 555 is not easy to make pwm the way I want it...... but if you have a good schematic to share using 555 I would use it as I have plenty of 555's on hand.
    noticed I placed the 100ohm resistors in there to protect the A-C leads from connecting together if both opto's went on at same time. the 100's only affect min(start) and max(pulse) current by just a smidge..... but still 10k between A-C, etc.
     
    Last edited: Apr 24, 2014
  2. inwo

    Well-Known Member

    Nov 7, 2013
    2,433
    315
    Although controlled by a pot, most machines (not welders) I've run into, have a pull-down resistor. If pot gets disconnected, it goes to lowest setting.

    You may not have to simulate a pot. Just input a zero or XXvolt (PWM) on the wiper lead.

    If it is designed for user pot control, input may be conditioned to reject fast changes.

    I understood the on-off switch function to be separate issue from this modification.

    As high side voltage is unknown. Use the lead to high side of pot to supply the PWM output.
     
  3. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    let me add some clarity

    the on/off is just a switch using its own pair of wires in the 14pin amphenol connector. it boggles my mind why they use a huge 14pin amph connector when only 6 pins are used. i have a on/off button on my torch, and this pulse controller would simply replace the foot pedal pot so i can achieve a pulsing affect, etc.

    if the pot is not connected the unit will start at max current setting when the switch closes. however, i have no idea how the pot is being used inside the machine. i suspect the 10k pot (A-C) completes some circuit in the unit and the wiper is used to bias something using A-B or B-C (likely moving towards pull-down as pedal is pressed). due to the way the wiring diagram is for the foot pedal (the connector uses all 3 connections to the pot) i do not believe it functions simply by wiping using A-B or B-C, etc.

    so you are saying all i really need is the 10k between A-C and then do a pull-down on B-C (0 ohm) to pulse? seems possible, but as noted, i have no idea if the machine needs to see that A-B (10k ohm) connection to stay running on the low setting

    i could try to dig and find a schematic, but from what i can see already i am only finding replacement board stuff and not schematics of any one board.
     
    Last edited: Apr 24, 2014
  4. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    What is the power supply voltage of the 555?
    The led voltage is about 1.2 Volts.
    In the datasheet of the opto relay is given that the led current should be 1 mA min.
    5 mA would be better for correct functioning.

    Bertus
     
  5. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    oh, havent gotten there yet. the supply to run the controller will be built as needed, etc. i am at the concept stage of this controller, have not chosen parts yet, etc.

    1) will using the opto's like i have shown work? i dont know the current vectors through the pot, so the opto's need to be the "ac" type to allow bi-directional flow.
    2) i dont know yet if using 555 is best way for PWM for changing freq and duty cycle.

    your thoughts?
     
  6. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,509
    2,369
    There is a multitude of 555 PWM controllers out there, you would need to tailor the PWM frequency to a very low value, also I don't think you would need an opto, In fact it may be possible to supply the 555 circuit from the outer terminals of the pot. check it for voltage just out of interest, if you have a schematic of the unit, check where the pot is fed from.
    Max.
     
  7. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
  8. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    those 555 pwm's dont have dual adjustments, just duty cycle, but i might see about using a rotary switch to switch in different caps where C1 is, this would allow me to change the freq using a single pot R1 for duty cycle, etc.

    @MaxHead - well, i am trying to avoid relying on anything from the unit itself, the goal here is just to switch the wiper in a controlled manor, etc.
     
  9. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    In the given post there is also figure 5.3, wich uses a 555 as oscillator and a 393 as comparator:

    [​IMG]

    The 555 sets the freaquency and the 393 the PWM value.

    Bertus
     
  10. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    i didnt scroll down.
    its an extra IC i would need to use.... but an option.

    and to note, i think the description of that 5.3 PWM circuit needs fixin, the text probably should say "10k pull-up R8" and not R7.

    if i go with rotary switch that has 1 pole and 5 throws, i could switch in 10/30/50/70/90uf to give me rough freq settings of 7/2.33/1.4/1/0.77Hz
     
    Last edited: Apr 24, 2014
  11. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,509
    2,369
  12. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    thats duty cycle only, which is listed on the AAC page that bertus gave. to change freq you either need to change the pot or change the cap, or use 5.3 on the AAC page, etc.
     
  13. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,004
    1,523
    I'd be very surprised if the foot pedal pot is working like you think it is. More than likely it is set up like a volume control. If not that it's used as a voltage divider. Never saw a TIG pedal set up any other way.

    What brand and model TIG machine is it? Miller and Lincoln both have their manuals and wiring diagrams on the web. Most of the pulse machines just vary the duty cycle not frequency, on pulse. Frequency is usually varied on AC only.
     
  14. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    a voltage divider, probably so, but i dont need the in-between, the wiper moves completely to the other side at full current (full pedal press). thus to pulse i just need to move the wipe from the min side to the max side.... you see it now?

    its a esab heliarc 161 unit and it has "start" and "max" controls on the unit itself, so i could set my max (pulse) to some #, and then my non-pulse "run" current using the start setting (a % of max), etc.

    and no, even on DC the pulse freq can be varied (1Hz, or 2Hz, or 3Hz, etc etc). on AC the current freq varies, mine is something like 60-240Hz when using the pedal, but this is independent of current and pulsing current settings, etc.

    by definition, "pwm" has freq and duty cycle, etc.
     
  15. wayneh

    Expert

    Sep 9, 2010
    12,093
    3,031
    Just to niggle, I disagree that you need control over frequency in order to implement PWM. It's all about duty cycle only, IMHO.
     
    shortbus likes this.
  16. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    the definition of PWM is a "freq + duty cycle". you can fix both of them, or make one adjustable, or make both adjustable. so by definition, you need to "control" the freq, whether that is fixed or not, etc.

    the std US 60Hz is in fact a type of PWM, its 120Hz sinusoidal @ 100% duty cycle..... why 120Hz you say, because there are two "pulses" per "cycle" and one of them is a mirror negative pulse. we typically think of a pulse as square wave, but its not restricted to square wave, and sinusoidal doesnt really look so much like a "pulse", but then look at the AC after it passes though a single diode, now it looks like pulses @ 50% duty cycle, etc.

    i am looking to make both freq & duty-cycle adjustable, etc.

    heading back to my concept question, can i flip the wiper using opto's? are there any electrical characteristics i am adding to the connections, or will i get what seems like a hard physical flip of the wiper from one side to the other? what type of opto's do i need to allow that bi-directional flow?
     
    Last edited: Apr 25, 2014
  17. wayneh

    Expert

    Sep 9, 2010
    12,093
    3,031
    You have an odd view of what is and is not considered "PWM". I've never heard the term applied to anything other than square waves, with the control of time-average output achieved by modulating the duty cycle. The clock is fixed, more often than not, but that is not relevant.

    If you can link to another definition, I'd like to see it.
     
  18. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,004
    1,523
    I must have over the years used 20 different Tig's most with pulse. Never yet saw one that pulsed minimum to maximum. Or I'm not understanding your meanings.

    Here's the manual if you don't have it - http://www.google.com/url?sa=t&rct=...MEtJ8zZmiw-i925skHfOfYw&bvm=bv.65397613,d.b2I
    It shows the pin out of the pedal connector.

    Don't think your going to do what you want with just a simple 555 timer. Off the top of my head it will probably take a flip flop and some other logic. Two pots would be needed to set the minimum and maximum currents, then the PWM would control that.

    And in the end you have to keep in mind there's the chance you will blow a board in the unit. Not trying to discourage you, just stating the obvious.
     
  19. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    the pin-out is useless info for knowing how the attached circuit inside actually works with the pedal pot, would need a schematic, not a pin-out, etc.

    a pulse is a pulse of higher current.
    the pedal works by starting at the min current (% of max as set on the panel), then as you press the pedal the current will follow all the way to full pedal which is max current set by another knob on the panel.

    i have the schematic for my pedal, what i dont have, as i mentioned, is the schematic for whats inside the machine, thus its kinda a blackhole as to how the pedal pot is used inside the machine. this does not mean i cant make a controller independent of the machine, etc. i dont think i will blow a board in the machine as all i am doing is presenting 10k to the terminals (just what the pot does) and then moving the wiper from one side to the other using opto's. can do the same using a physical relay, but a physical relay cannot keep up with my PWM needs and would likely fail on the mechanical side, etc.

    and yes, a 555 provides sink to ground and source to Vcc via a single pin, so in essence it is a "flip-flop", and when connected to opto's in the right way the opto's become "flip-flops", etc. its what i attached to my op.

    i thought about just using mosfet's for the switching, but the opto's give me better isolation, etc.

    the effect will be this
    http://www.weldingtipsandtricks.com/tig-welding-equipment-pulse.html


    you are boxing yourself in to the clocking/digital/switching world where clean/tight square wave is desired. PWM can be applied to any signal of any shape, and can even be imposed atop of a sinusoidal carrier, etc. so if you are a digital/switching guy, sinusoidal AC after passing through a diode would be considered a real crappy square wave, etc.
     
    Last edited: Apr 25, 2014
  20. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,004
    1,523
    I thought the pin out would be helpful. With it a second pedal could be made with the pulse built into it. Still think it could be done with two pots, voltage dividers. The high one set by the pedal position, the low one manually.

    Quote,^ "a pulse is a pulse of higher current." Think that is backward. The pulse is the low current. The high current is the welding current, it pulses to a lower current to control the heat.

    I never liked the pulse myself. A lot of guy's use it as a crutch, instead of learning how to weld correctly. As a way to try and get the "stack of dimes" weld bead.
     
Loading...