Announcement

Collapse
No announcement yet.

Econotig: Partial Flame Control (50%-100% only)

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • jjohn76
    replied
    Glad it worked out! It almost sounds like part of the shunt windings are shorted together, which limits the effective windings by about 17% (370VAC instead of 430VAC). That could be the reason why you're not seeing the entire amperage adjustment. Either that or the gap has somehow adjusted (alignment or spacing), but this is far outside my understanding...
    Last edited by jjohn76; 11-20-2019, 07:27 AM.

    Leave a comment:


  • osmu0011
    replied
    Good catch on the unbalanced output waveform statement in the patent, it caught my eye as well, but then I wrongly dismissed it thinking it wouldn't apply since I was dealing with DC. Obviously it does apply since it is the only power output adjustment for both AC and DC on this machine.

    Also, I follow your logic and I believe your assessment of the shunt control circuit is correct. To answer your question about the shunt coil voltage, the highest I measured with the DMM was 370VAC with the leads disconnected.

    After fine tuning R57 a bit more tonight, I have much more useable pedal travel. It isn't 100% utilized, but it is much more than it was, I'd say close to 80%. By adjusting R57, you can move that 80% "band" earlier or later in the pedal travel, so I set it a bit biased towards the high end as it made it a little more friendly for thin materials. So, now 0-15% of pedal travel is minimum output, 15-95% is variable output, and 95-100% is maximum output. Good enough for me... Think we can close the books on that!

    I still feel like the minimum amperage is a bit higher than what it should be, but I'll have to measure it someday to know for sure. I tried some 20ga (0.036") steel test parts tonight and was able to work with them, it was not pleasant, but it did work out ok as long as I went quickly enough. I know 20ga is listed as "difficult" for this machine, so maybe I've just maxed out the capabilities on the low end, but using a different Econotig years ago (the machine I learned on) it did not seem to be as hard as this one...

    Leave a comment:


  • jjohn76
    replied
    Edit, looking at the tech manual, page 23, the shunt winding seems reverse biased, so it switches on the negative AC cycle instead of the positive AC cycle like mentioned above. Either way, I think the timing comes down to the time constant set by capacitor C22 and the variable resistance from R57-panel amperage potentiometer-pedal amperage potentiometer-R41. If the voltage measured at RC9-4 doesn't vary over the entire range set by the panel and pedal, then either that potentiometer is off, or one of the components in that chain is out of spec.

    Leave a comment:


  • jjohn76
    replied
    Reading page 10, lines 16-20, it seems like the shunt coil only conducts in one direction (unbalanced for gtaw), which is why Q13 is a thyristor (one direction conduction) instead of a triac (two direction like in figure 4). Q13 is used to short across the shunt coil. I would guess R52-R53-C35 is the Snubber for Q13, with D22 and D20 the voltage clamp. When the shunt voltage AC cycle goes negative, D21 is forward biased, which activates Octocoupler OC1, that energizes Q11, which energizes Q10, that discharges C22 through R46. The voltage across C22 drops to the voltage across R46, which is based on the amperage control settings (long resistor divider network). When the shunt voltage again goes positive, Q10 is turned off, which allows C22 to charge through the amperage control settings resistors (R57, panel potentiometer, pedal potentiometer, R41), until it's voltage is high enough to forward bias Q12 and optocoupler OC2. OC2 forward biased triggers Q13, which shorts the shunt coil. The amperage control settings and the R57 potentiometer control how long it takes for C22 to charge high enough to trigger OC2 and Q13.

    RC9-4 voltage should vary over panel and pedal settings as the duty cycle of the optocoupler OC1 LED varies with shunt driver timing. RC9-3 should be a constant voltage at about 7.5V.

    What did you measure the shunt voltage when the coil was disconnected? Does it get to 430VAC?
    Last edited by jjohn76; 11-18-2019, 09:04 PM.

    Leave a comment:


  • Noel
    replied
    I've read it three times today. I'm still wrapping my mind around it.
    To make more sense of it I'd have to print pages, blow up diagrams to enhance visuals, get out the high lighters and a note pad with pencil then spend some serious quiet time studying.

    I agree that patent held a wealth of information and does offer a much better explanation then I've ever received on the subject of how they make welder magic happen.

    I've seen and wondered about welding issues that appeared electrical function related but without knowledge of such things could only speculate on the reason behind them. Blame the board and wash hands of it is rapidly becoming there's a reason and I can figure it out maybe?

    Reading that patent took some of the mystery away and replaced it with a greater knowledge, understanding and appreciation, to what goes on inside the box, things on the board, and appearing on the schematic with why they do.

    While not able to offer up much past moral support, encouragement, I apologize for any distraction from the path to that solution. Following along and paying attention just the same.









    Leave a comment:


  • osmu0011
    replied
    Yeah, I read thru the patent too, a lot more technical details from a lot I've come across which was really nice in this case.

    If you were to tweak the air gap distance (not likely on this behemoth), or insert a material that has a different permeability, I would imagine you could possibly change the reluctance on the shunt leg and so change the gain of the main transformer? I'm assuming that this was all tuned at the factory and nothing has changed since as it is quite a solid structure, so what else might have changed? Would something cause a change in impedance of the primary, secondary, stabilizing (Z1), or blocking (Z2) coils that would have impact on current output? In the service manual, it does say that Z1 "smooths current from main rectifier in DC mode and limits output in low AC setting"...

    Leave a comment:


  • Noel
    replied
    https://patents.google.com/patent/US5187428A/en

    Thanks for the mention to the patent number. I googled it. Interesting read.

    "As will be explained below, air gap 208, in conjunction with the physical dimensions of magnetic core 102, determines the gain of the amplifier effect, thereby determining the minimum and maximum output current of transformer 100."

    "Any ideas why the low end power would be so high?"

    Your question was running around in my head and when I read this, my head already spinning from all the mad science explanation to things, I asked my self if that statement was a clue to the answer?





    Leave a comment:


  • osmu0011
    replied
    Yeah, I'm trying to sort out how it shunts both sides of the wave as well. Looking at the waveform (chart H), it is a little different than a "normal" phase control which may provide more of a clue. Perhaps triggered by the negative transition with a delay to close, than remains closed after the positive transition with the same delay...

    I knew you really meant R57 I did try this quickly last night and it did appear to make an effect, so I will have to fine tune it a bit more when time allows. We'll see how far that gets me. I will also check the voltage across the shunt with leads disconnected as you suggested and get back to you.

    What puzzles me is why the arc is so hot when the shunt coil is disconnected as the power should be minimized. I may have to set up a shunt and try to measure the amperage to see what my range really is right now (should be 30 to 180 amps). Any ideas why the low end power would be so high? Geometry of the transformer, short in a coil, etc.?

    Leave a comment:


  • jjohn76
    replied
    Sorry, I mixed up R56 and R57. What voltage are you getting across the shunt winding when it's disconnected from the board (idle and under arc)? You may be able to get some indication of the shunt driver timing if you measure across C22 (RC9-4 to board common). The average voltage across that capacitor should change with panel and pedal amperage adjustment.

    Leave a comment:


  • jjohn76
    replied
    Todd, I still haven't quite figured out how it shunts both the positive and negative AC cycle, but it seems that at every negative transition, the LED in OC1 is forward biased, which energizes Q11 and Q10, bringing Q12's anode to ground (C22 is discharged). The current through the amperage control circuit (limited by R56, the panel amperage knob, and the pedal resistance) controls how quickly C22 recharges and turns on OC2, which shorts the shunt and increases the output current. If you turn R56 clockwise, it should delay C22's charging, which keep the shunt winding open longer and reducing output current.

    Edit: I mixed up R56 and R57.
    Last edited by jjohn76; 11-18-2019, 06:59 AM.

    Leave a comment:


  • osmu0011
    replied
    Ok, so new bottle of 100% Argon hooked up and new 2% Lanthanated tungsten electrodes installed and the same behavior is still happening... which was expected.

    I put the DMM on the shunt coil leads to see what the voltage was doing and this is what I saw:
    - before triggering the arc (no load), voltage was around 60vac
    - with front control knob set to 100%, when arc is triggered using pedal, voltage climbs to 370vac (spec says should be 430vac)
    - as pedal is pressed from 0-60% voltage remains constant at 370vac
    - as pedal is pressed from 60-100% voltage decreases to 80vac
    I repeated this using the pedal to trigger the arc (pushed pedal to 100% with knob at 0%) and varied the amperage using the front control knob. The same behavior was seen where 0-60% of the knob did nothing, and 60-100% of the knob dropped voltage.

    As a different test, I then disconnected the leads of the shunt coil just to see if output amperage dropped to minimum with no variable control. The control was completely gone as expected and the amperage was low, but not any lower than when the shunt coil was connected. So, now I think I may have 2 issues as the minimum current is way too hot for the thin materials it should be able to handle and I only have half the control span that I should.

    I just happened across the patent for this which may contain some further info, will have to read thru that later tonight. If anyone is interested, it is US Patent 5187428.

    Leave a comment:


  • jjohn76
    replied
    I think about as good as you can get without an oscilloscope is the "RMS" reading from your multimeter. There were some AC specs listed for the shunt driver circuit. RC9 has a few test points that may provide info, but those are probably more informative for timing. I still don't quite understand how that circuit works. OC1 seems to check when voltage is reverse biased across the shunt winding and resets Q12's driver through Q10. The amperage control potentiometers all seem to limit the current to the driver circuit, which controls the rise time in Q12's gate circuit. Hopefully that made sense...

    One word of caution about the oscilloscope is to make sure what you're measuring is isolated from mains and your leads have enough attenuation. The AC voltages through the shunt winding seem to peak around +/- 600V.
    Last edited by jjohn76; 11-10-2019, 10:31 AM.

    Leave a comment:


  • osmu0011
    replied
    Ok, so here's an update from last night. I made it thru the list of test points and most checked out fine while using the board ground, so that is good news. There were a couple of oddities from the listed values, for example pin RC1-12 was the same as RC1-9 (as you mentioned in post #22, how could they not be). Another was RC1-10 which was out of the listed range for both modes, but I was able to adjust with R57 for tig and R56 for stick to bring it back into spec. The next was RC1-11 and I'm questioning the listed values again as I was getting ranges from 12.0-5.0vdc with pedal travel which seem to make more sense than the 3.7-5.0 listed in the manual.

    With all that adjusted, I went to run more arc tests and then immediately ran out of gas... so I'll have to get some supplies this week sometime and see if anything is better. Assuming that it isn't any better, I would think the next logical place to troubleshoot is the shunt driver circuit and coil. Any suggestions on how to start picking that apart without an oscilloscope, or should I just see if I find myself a cheap little USB scope and save some hassle?

    Leave a comment:


  • jjohn76
    replied
    Staring at the schematic some more, it sounds like back to those potentiometers... R57 looks like it controls the GTAW amperage range. At least now you have the prescribed voltage test points so you can get the duty cycle/trigger timing right on Q12.

    Leave a comment:


  • osmu0011
    replied

    You are correct, that is how the pedal is wired. C-E resistance varies from 1k to 0 and D-E varies from 0 to 1k. But as the Miller tech said to me “the Econotig was designed just a bit differently than all the others...” so it has no connection for D, they just use C-E as a rheostat. K pin on machine is for use with a hand control.


    Click image for larger version

Name:	4292341C-518A-4D53-AC8F-BA104EEDA238.png
Views:	108
Size:	210.5 KB
ID:	604373
    .

    Leave a comment:

Working...
X