The new wire

Flux core 309L that is . Well , I can run a halfway decent bead but I don't like the way it looks ... I get good penetration if I vee the joint , but if I wander off center I don't get good fusion on both sides . I think it's going to come down to coating my joints with the Solar flux , tacking with the MIG , then finishing the weld on the bench with the TIG . My TIG welds ain't all that pretty , but they look a lot better than my welds with this wire . Some of the welds will need to be made on the truck , those will probably be done with the MIG . Though I haven't tried the OA rig for this yet ... I mean , if ya got it , try it !

Reply to
Snag
Loading thread data ...

"Snag" wrote in message news:spjncp$lbh$ snipped-for-privacy@dont-email.me...

Flux core 309L that is . Well , I can run a halfway decent bead but I don't like the way it looks ... I get good penetration if I vee the joint , but if I wander off center I don't get good fusion on both sides . I think it's going to come down to coating my joints with the Solar flux , tacking with the MIG , then finishing the weld on the bench with the TIG . My TIG welds ain't all that pretty , but they look a lot better than my welds with this wire . Some of the welds will need to be made on the truck , those will probably be done with the MIG . Though I haven't tried the OA rig for this yet ... I mean , if ya got it , try it !

---------------------

formatting link
spent several class nights practicing vertical up on a non-critical project. It was quite a bit easier with the school's Miller than with my little 75A Century.

Either I forgot or no one ever told me there were different types of flux core meant to use with and without shielding gas. Could you have the wrong one?

Reply to
Jim Wilkins

No , this is not one of the dual-shield wires . The problem with the bead is that it is tall and narrow . I'm at max 88 amps (Lincoln Weldpak

100) and varying the wire speed hasn't made much difference . The bead looks nice , minimal oxidation on top , a little on the back side . I'm actually more concerned with the lack of fusion than the bead profile . It may be that I just don't have the amps for SS with this machine ...
Reply to
Snag

No , this is not one of the dual-shield wires . The problem with the bead is that it is tall and narrow . I'm at max 88 amps (Lincoln Weldpak

100) and varying the wire speed hasn't made much difference . The bead looks nice , minimal oxidation on top , a little on the back side . I'm actually more concerned with the lack of fusion than the bead profile . It may be that I just don't have the amps for SS with this machine ...

--------------------------

I get a thinner, taller bead dragging the gun and a wider, flatter one pushing it, with the same settings.

Reply to
Jim Wilkins
<snip>

Maybe try preheating the area some with a torch first?

Reply to
Leon Fisk

Well, just pop on over to the shop, and try it on the Miller 212. I'm not using it today.

Is it possible your wire speed just might be a little fast.

Reply to
Bob La Londe

Stainless shouldn't nees that much heat as it is a TERRIBLE heat sink / conductor - particularly with metal that thin.

Reply to
Clare Snyder

No , if I slow it down it burns back . I think it really needs just a few more amps , which I ain't got . Interesting though , the bead flattens a little after the first half inch or so . It's hard for me to run a bead much longer than about an inch and a half on this round stuff so they all look like crap ... at least my TIG beads look uniformly crappy !

Reply to
Snag

Sounds like preheat might just do the job for you then. Yeah, welding a round piece is a pain. Did some six inch well casing a while back. Fortunately the duty cycle of the cheap stick welder I was using made me stop halfway around anyway. LOL.

Reply to
Bob La Londe

...Fortunately the duty cycle of the cheap stick welder I was using made me stop halfway around anyway. LOL.

-------------------

After buying a Stick/TIG Lincoln I found a better use for my 50A buzz box arc welder as the transformer in a homebrew 12-24V 30A battery charger. It needed a Variac to tame it. If necessary I can still reconfigure it as a

120V stick welder.
Reply to
Jim Wilkins

That is fascinating. I do have a variac on the shelf that was planned for a sheet plastic bending tool, nut since I discovered I could bend poly carbonate on the brake I haven't really played with the original idea much. My buzzbox is AC only so to use it as a charger or power supply I'd need to add a huge bridge rectifier and some serious power caps. Not that I couldn't. That's essentially what I did for the servo power supply on the Hurco mill. Well the transformer wasn't quite that big, but...

Reply to
Bob La Londe

For battery charging unfiltered is every bit as good as filtered - and many "experts" would say better as the "pilsed charge" helps break up suphate deposits, agitates the electrolyte, and dislodges bubbles. Using it as a charger at up to 50 amps doesn't erven take a HUGE rectifier. A full wave bridge that handles 50 amps is only about 1 1/2 inches square and 3/8" thick. AMd only costs a few bucks.

Reply to
Clare Snyder

That is fascinating. I do have a variac on the shelf that was planned for a sheet plastic bending tool, nut since I discovered I could bend poly carbonate on the brake I haven't really played with the original idea much. My buzzbox is AC only so to use it as a charger or power supply I'd need to add a huge bridge rectifier and some serious power caps. Not that I couldn't. That's essentially what I did for the servo power supply on the Hurco mill. Well the transformer wasn't quite that big, but...

-------------------

The rectifier is an MDQ-100A, $12.99 on Amazon. I found the 78000uF cap at a ham radio flea market. A BeesClover 100V/100A DC meter displays the output voltage and current. I buy from Amazon to minimize exposure of my credit card, not because they are a great parts source. Amazon is like Radio Shack, wide variety, uncertain availability, questionable quality.

The welder transformer has substantial uncoupled self inductance that gives it a constant-current output curve, enough that the ripple is only about 1V p-p at 20A DC out. The down side is poor output voltage regulation, it droops from 57V at no load to 30V at 30A, which is still enough to charge

24V batteries but makes it dangerous to leave unattended since the voltage won't stop rising as the battery nears full charge.

If you want lab-quality voltage and current regulation from it, add a DPS5020 digital regulator.

formatting link
Without the regulator the supply can deliver up to 70A long enough to test circuit breakers etc before its 30A output breaker trips. I have a surface mount breaker in my solar power system that's rated for 30A but actually trips below 19. The supply's output current limitation is from overheating, especially the Variac's brush which may be hard to replace. I had to make one by milling and filing a motor brush. The welder transformer I used begins to saturate at a little over 120V in. The no-load input current rose to 5A at 138V in, so you may want to connect the Variac for 0-120V out instead of 0-140V.

-jsw

Reply to
Jim Wilkins

I'm running a 100amp rectifier on the Hurco mill. With a decent size heat sink it was around 20 dollars if I recall. It was enough I had to count it, but not enough I worried about it. The big power caps were a lot more.

Reply to
Bob La Londe

I'm running a 100amp rectifier on the Hurco mill. With a decent size heat sink it was around 20 dollars if I recall. It was enough I had to count it, but not enough I worried about it. The big power caps were a lot more.

----------------------

The limit I couldn't work around was heating of the primary winding. Although it's rated for 50A at 25 arc volts the duty cycle is 20%. I found that an output of 30A for a few minutes followed by ~22A for half an hour would heat it to 70C at the outer windings, measured by jamming in a fine thermocouple after finding the hottest spot with an IR thermometer. AFAICT

20-30A is the max recommended charging current for my 105Ah marine batteries, so I stopped trying to push the output higher. There's a current shunt and an anti-backflow diode that also warm up, plus a reverse diode to trip the breaker instead of blowing up the cap if it's connected to a battery backwards.
Reply to
Jim Wilkins

...

A good way to measure the temperature of the coil is with the change in resistance. Measure the resistance cold, run some low-ish current for a while, measure the resistance again. Keep cycling with higher currents.

Copper's tempco is +0.393 percent per degree C. Which isn't very much, especially with the heavy winding of a stick welder. You'd likely need to use a Wheatstone bridge.

Wouldn't distinguish the hot spots, though. But the hot spots would be internal and indistinguishable anyhow. There's probably a rule of thumb for the temperature gradient in a transformer.

[End of meander]
Reply to
Bob Engelhardt

A good way to measure the temperature of the coil is with the change in resistance. Measure the resistance cold, run some low-ish current for a while, measure the resistance again. Keep cycling with higher currents.

Copper's tempco is +0.393 percent per degree C. Which isn't very much, especially with the heavy winding of a stick welder. You'd likely need to use a Wheatstone bridge.

Wouldn't distinguish the hot spots, though. But the hot spots would be internal and indistinguishable anyhow. There's probably a rule of thumb for the temperature gradient in a transformer.

[End of meander]

-----------------

I have the instruments to measure tenths of milliOhms which would still give me only the average temperature. The primary is 198(?) turns of 10 AWG at 1 milliOhm per foot, rated for 17A, the secondary is 6 AWG and runs cooler. The thermocouple let me graph the temperature rise on a laptop to observe its approach to a steady state, and check by briefly raising the current and then seeing if the temperature peaked and decreased. The temperature rise rating is 115C but I wanted some margin for running safely enclosed instead of open, accessible and hazardous. The fans on my APC inverter and T60 freezer are loud enough that I can't hear whether anything else is running or not, like a fridge or the fan in the power supply.

The datalog spreadsheet shows that at 25V, 25A out the transformer reached

73.4C at 31 minutes, rising 0.5C per minute, and cooled slightly afterwards at 20V, 20A, open and without the fan.

It should be fine at 20A continuously, 30A for ~10 minutes, and all it can deliver for brief experiments. The 30A output breaker will pass 70A for about 5 seconds and the ammeter reads to 100A.

Reply to
Jim Wilkins

I noticed that after swapping Variacs/Powerstats and heatsinks to eliminate lower limits, the square of the max continuous current was approximately the duty cycle (0.2) times the square of the welding current. This makes sense if the limit is I^2*R heating.

Reply to
Jim Wilkins

Cool!! I can see that you're well beyond any "advice" I might have about finding temperature with resistance. I misread your "... measured by jamming in a fine thermocouple ...".

Reply to
Bob Engelhardt

Cool!! I can see that you're well beyond any "advice" I might have about finding temperature with resistance. I misread your "... measured by jamming in a fine thermocouple ...".

-------------------------

For automated production testing the resistance measurement would be fine, because it could be done in a tenth of a second or less without disturbing the rate of rise measurement, and wouldn't require the operator to (mis)place the thermocouple. There should be separate force and sense connections to the winding ends to eliminate variable contact resistance. The measurement will between the inner contacts, whether they are force or sense. Can you guess why?

For bench testing the interruption would be long enough to disturb the rate.

The forward voltage drop of a diode, such as a transistor base-emitter junction, is another good indicator of temperature.

Reply to
Jim Wilkins

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.