PowerPro 256 makes HF but wont start arc

[acannell]

Still very happy with my PowerPro 256 and I have made alot of money with it. Finishing up some production welding today (TIG, 115A, AC) and the arc suddenly went out. Hf working, still seeing the non-stop lightning storm, just the arc wont start.

If I leave the machine off for a few minutes, I get another few seconds of arc, then it stops again.

I tried sanding the points and it didnt help.

I also tried replacing the MC14053BCP on one of the daughterboards that has previously fixed similar issues for me..no change.

Any ideas? I have a full suite of electronics test equipment so I can get in there and do whatever needs to be done. I'll make sure to report back what does an does not work so others get help from it as well.

EDIT: whoops..welding mode is AC not DC. Also, note that I have not changed welding settings in the past year besides main current level.

[acannell]

also I think the LED with the exclamation point inside the triangle is turning GREEN....does that mean duty cycle exceeded? is there a duty cycle timer or something which might be malfunctioning and turning off the arc on purpose? can I check that somehow? the welder has been idle for 10+ minutes and that LED was green..so I dont think its a real duty cycle alarm..but maybe that circuit got damaged?

EDIT: okay updated description of symptoms...

I can get 1 to maybe 30 seconds of arc now by powering the machine off then back on. The arc goes out suddenly during welding. The HF continues to work. The green LED with the symbol of the triangle with exclamation point inside always seems to be on once the arc goes out (it is not on when I first power on).

So I'm guessing something died in the overcurrent/duty cycle sub circuit? Seems to weld fine when I get that few seconds of arc.

[Rambozo]

Duty cycle warning is typically tripped by a thermo sensor on the main heat sink with the IGBT modules. There might also be current sensors that look for imbalance between the switching elements. There may be a true problem and the protection circuit is doing it's job to protect your IGBTs. It should be relatively simple to trace out how those circuits operate, and test everything before making a conclusion. Just be aware that the output usually isn't ground referenced, so if you use a grounded scope or bench meter, you can let the smoke out in an exciting way. Battery powered test gear is the way to go for this. Beware of a possible IGBT latchup condition. I would hope that part of the protection circuit looks for that.

[acannell]

Duty cycle warning is typically tripped by a thermo sensor on the main heat sink with the IGBT modules. There might also be current sensors that look for imbalance between the switching elements. There may be a true problem and the protection circuit is doing it's job to protect your IGBTs. It should be relatively simple to trace out how those circuits operate, and test everything before making a conclusion. Just be aware that the output usually isn't ground referenced, so if you use a grounded scope or bench meter, you can let the smoke out in an exciting way. Battery powered test gear is the way to go for this. Beware of a possible IGBT latchup condition. I would hope that part of the protection circuit looks for that.

thanks for the help! so the warning light has two colors right, orange and green? the manual doesnt say which color means what. I would think orange would mean overheat and green would mean something else, any ideas?

looks like what I could do right away is trace the led harness to where it goes on the main pcb and see if I can figure out how the circuit there operates

[acannell]

okay..using a schematic everlast sent me that may or may not match my machine, it would appear that if the triangle/exclamation point symbol LED is green, that its a overcurrent condition.

im basing this on finding that the harness for that led goes to the X16 location on the main pcb, via a small daughterboard

that connector has 7 pins, and one of them seems to be shared between the bi-color led and the main power led, which is connected to pin 7 (black), and thats what x16-7 is shown as on the schematic..so thats consistent

also the other pins on that connector seem to match up with the schematic, x16-5 appears to be the "power" led , and the two next to it, x16-4 and x16-6, are the overcurrent and "heat" leds. the overcurrent detection circuit seems to be at the top of the page, with its sensor "L1", a 2 wire inductor, connected to 1X1. I cant find 1X1 on the main pcb, but I do see a two wire toroid inductor around a big wire, that is connected to the main pcb via a screw terminal. the terminal is potted so I cant see if there are any indicators around it. also that toroid is just free floating around that big wire, nothing holding it in place.

anyway, it seems what triggers the fake (real?) overcurrent green led is when I drop current down very low on the pedal, almost to off...then the arc shuts off and wont come back on until I power cycle the machine, but then it works fine immediately for what may be an indefinite amount of time until I drop current down too low again

given how much is going on in the welder, I'm not sure this really indicates a problem with the overcurrent circuit..who knows whats going on..maybe the overcurrent is real and being caused by something else, that happens when I drop current down very low

[Rambozo]

okay..using a schematic everlast sent me that may or may not match my machine, it would appear that if the triangle/exclamation point symbol LED is green, that its a overcurrent condition.

im basing this on finding that the harness for that led goes to the X16 location on the main pcb, via a small daughterboard

that connector has 7 pins, and one of them seems to be shared between the bi-color led and the main power led, which is connected to pin 7 (black), and thats what x16-7 is shown as on the schematic..so thats consistent

also the other pins on that connector seem to match up with the schematic, x16-5 appears to be the "power" led , and the two next to it, x16-4 and x16-6, are the overcurrent and "heat" leds. the overcurrent detection circuit seems to be at the top of the page, with its sensor "L1", a 2 wire inductor, connected to 1X1. I cant find 1X1 on the main pcb, but I do see a two wire toroid inductor around a big wire, that is connected to the main pcb via a screw terminal. the terminal is potted so I cant see if there are any indicators around it. also that toroid is just free floating around that big wire, nothing holding it in place.

anyway, it seems what triggers the fake (real?) overcurrent green led is when I drop current down very low on the pedal, almost to off...then the arc shuts off and wont come back on until I power cycle the machine, but then it works fine immediately for what may be an indefinite amount of time until I drop current down too low again

given how much is going on in the welder, I'm not sure this really indicates a problem with the overcurrent circuit..who knows whats going on..maybe the overcurrent is real and being caused by something else, that happens when I drop current down very low

Sounds like you found the current sensor. Remember the first rule in troubleshooting, Thou shall always check voltages. While you may not know what the proper voltage and current levels are, because that machine is multi process, you might just be able to compare between the different modes to see what the differences are. The current pickup may get divided down or amplified, and then run into a comparator or opamp to set the threshold. Be sure to check the reference voltage and how it is being derived. If that is low, the overcurrent circuit will be too sensitive, even though everything in it is working fine.