Mike BR

I guess roughly speaking, the bypass would set the pressure and the choke would set the volume. A low pressure high volume blower could stall before you could restrict flow enough with only a choke. And if you used only a bypass, there might not be enough pressure available at the tuyere to overcome a tighter-packed fire. The better option is probably to use a more suitable blower. But then I make do with what I have too much to criticize someone else.

Either way, I wouldn’t want it stuck in my head.

Have you tried peening the welds? As I understand it, warping happens mostly because the weld puddle freezes at a relatively high temperature, then contracts as it cools, pulling the surrounding metal with it. If you can peen the weld and spread it back out the the size it was when hot, you can often reverse the warp. It looks like you might be able to support the brim joint from the outside of the hat (inside of the angle) against something like the edge of an anvil, or better, a swage block, and then work the weld from inside the hat with a hammer. You're probably way ahead of me, but I've had good luck using electrical wire (stripped down Romex, for example) as TIG filler for copper.

It seems to me that there are at least three ways an anvil can absorb energy. One is through the surface deflecting (and not rebounding strongly or quickly enough to put the energy back into the work). A second is through the anvil structure flexing (think the 3/8" plate example). And the third is through the entire anvil moving downward under the blow and flexing the stand underneath. The first can be minimized with a hard anvil face, the second through using a compact mass as the anvil, and third through using a heavy anvil. No material is perfectly rigid, so if an anvil face deflects and absorbs energy under a ball bearing, it will deflect and absorb energy with a piece of hot steel under the hammer. And any energy absorbed by the anvil doesn't go into shaping your work. But I would guess that the energy lost that way is quite small compared to the second and third paths above. A ball bearing theoretically contacts a flat face at a single point, creating (initially) infinite pressure even though the total force is small. Hot steel spreads the force over a much larger area, so the pressure at any point (and therefore the deflection) is lower, at least relatively. And common sense says that the even a mild steel face won't actually deflect that far under a hammer blow against hot steel. So it seems very likely that a lot more energy is lost to the anvil structure and the stand underneath flexing than to deformation of the surface. I do expect that the edges of a soft steel anvil would roll pretty quickly, even though it might hold up fairly well if you only worked over the sweet spot. (You could always make a hardened block to use when you need an edge.)

I like to multiply metric lengths by 4 and then move the decimal point as appropriate. 1mm = .040, 1M = 40”, etc.

Here's one other possible approach: Take a length of 1/2" square and bend it into the big "U" you described. Now take a piece of 3/8" square and form it around the outside of the first piece. Probably easiest to do this with a torch and bending wrenches (and perhaps a clamp or two). Grind the outside corners of the 1/2" square to round them off, and sand (and maybe polish) the inside of the 3/8" piece. Now cut a strip of the material you're using, and sandwich it between the two curved pieces. Hold the assembly together with C-clamps. Then you can hammer the sides of your strip down against the 1/2" square form, heating with a torch to anneal as needed. If you're using steel, you could work it hot, also using the torch. For non-ferrous (especially if you're just making one piece), you could start with the 3/8" square, and trace it on 1/2" plywood so you can cut out the inner form. You'd need to cut a series of holes so you could clamp the steel to the plywood. The disadvantage is that you'd need to take everything apart before you could anneal. Keep in mind that you'll be shrinking the sides of the piece quite a bit, so you'll probably need a number of anneals if you aren't working hot.

Just flagging for anyone who doesn't know that tempering in a gas forge (or dragon's breath) can be tricky. The atmosphere can be reducing enough to stop the oxide that creates the colors from forming, even though the metallurgical changes inside the piece are going on as usual. You can help compensate for that by frequently pulling the piece out into fresh air (or by learning through experience what to expect). I guess it's less critical if you're just coloring the piece. But a beautiful pigeon's throat piece going full blue the second you pull it out of the forge could be frustrating in itself.

If anyone's interested in a 9 minute Swedish video on Soderfors, there's one here: https://www.svtplay.se/video/8PvyB4m/arkitekturens-parlor/soderfors-bruk-tierp?info=visa It's more about the worker's housing than the works itself, but it's kind of a neat company town. Most of it should be reasonably obvious, but the masonry they examine near the end (on the "widow's house") is "slag stone" -- blast furnace slag formed into blocks.

With a piece of 1" round inside the the section to be bent, you could do it hot over the anvil. I did that with 3/4" pipe (and 3/4" round) to make a 90 degree bend for a forge burner. Even with bending around a form to get a constant radius and -- of course -- cutting off the straight sections at the ends, it took a fair amount of tweaking and tapping to thread round bar back out of the curve. It worked, though. For blower legs, you could just leave the 1" round inside. You might not want to do that with sch 80, though. . .

If it were me, I'd drill holes in the feet and bolt them to a wall. But I think I once heard someone say I was cheap . . .

I tweaked my taper attachment and needed a project to try it out yesterday, so I turned an adapter that is hopefully good enough to work (with a 3JT chuck). If you message me with your address I'll send it. You'll probably need to buy a 1/2-24 tap and chase the threads. Not having anything 1/2-24, I measured at 3/8-24 bolt with safety wire and a caliper, then added .125 and used the dimension to cut a gauge of sorts. I then cut the internal threads with piece of 3/8 hot rolled with a little HSS drill bit puddled on. So it will likely need to be chased -- and if it fits too loose, there's always locktite, I guess.

I think I see variations in thickness in the corner bends that might be consistent with welds. And we know that has to be at least one forge weld in each piece. You could pierce a billet and forge out an endless loop of bar. But it would take more than clever assembly to feed that through a slit.

Back to the drill press, I noticed that the catalog Irondragon posted said you could request two different chuck sizes or a Morse taper (if not a Morris one). Conceivably Champion would have threaded an arbor if you asked them to.

Of course, there's no way to tell when it was modified, or why. In general, standardization has meant moving from more threads to fewer, so you shouldn't assume anything on old equipment. Actually you should measure the diameter as accurately as you can, too. A few months ago I cut a set of 14mm X 32tpi threads, so you never really know until you check.

If you find a machinist, one option would be an adapter that's 1/2-28 on the inside and JT3 on the outside. That way you wouldn't lose working height and wouldn't put extra stress on the quill (though you might have to shorten the threads to take full advantage).

I have to admit I can't be bothered to watch the YT videos. But has anyone explained why it makes sense to hang the power supply off the stinger when a bench could support it perfectly well? Regardless of the configuration, though, I bought my shop welder before the inversion revolution. $60 doesn't seem like much to pay for the ability to run a few quick beads anywhere there's a 120V outlet. Even if a few quick beads is all the machine has in it.

Funny — more than once, I’ve found the old tool when I opened the drawer to put the new one away.

My rule of thumb is always to take advice when I get it. Even if I'm doing something exactly the way I was shown, if someone suggests doing it differently, I try it (unless they're an obvious idiot, of course). I figure it never hurts to know another way to do something, and who knows -- I might like the new one better. The flip side is that if I'm bound and determined to do something differently from everyone else, I go off and do it in private, then tell about it later . . . maybe.

I’ve noticed that the newer axial fan hair driers stall quickly when the outlet is restricted. I have my doubts that one would work well with a forge. An old noisy one with centrifugal fan and snail-shaped housing might be better.

My 2X72 use a 5" polyurethane caster as a drive wheel. Based on advice I read somewhere, I popped the wheel in the freezer for a while, then mounted it on the grinder and crowned it with a coarse file. Seems to work pretty well. (The grinder is based on an old Harbor Freight 8" buffer -- think Grizzly knife grinder, but not nearly as "nice" )

That’s a good one to remember. Motor poles should only affect how the VFD converts an RPM setting to frequency. And I always set mine directly by frequency. But your VFD was probably trying to run the motor at twice the rated speed . . .

As promised, I checked my manual. Assuming your interface is the same as mine, the obvious settings that could cause low voltage/low torque are: Base Frequency, PD004. Should be 60 -- a higher setting could cause low torque. (I never bothered to change mine from the factory setting of 50, in part due to the speeds I normally use. But that's probably not the best choice.) Max Voltage, PD008. Should be 220 -- a lower setting could cause low voltage/torque (obviously).

I have a VFD of the same brand, but smaller. Mine didn't come with a gauge or external switch, but does have the two rows of terminals where you can connect various control wires. There are also settings you can program that determine what the terminals do (to some extent). It's possible that if you run it without hooking that ribbon cable back up correctly (if you even can run it like that) or without resetting the program, there would be trouble. More broadly, you can program the VFD to control how fast the voltage increases with the frequency. If that's set too low, it could cause it to run fast with no torque. (I can look at my manual tonight and see what the voltage settings are called.)

Like Steve says, double check the wiring. If only two of three leads from the motor are connected properly, it might run like you describe. Then, I'd check the settings on the VFD. I have two that require quite different steps to program, but there are a lot of setting that could be wrong. If you don't have a manual, find one online. And if there's a "restore factory settings" option, that might be worth trying.

If you google it, the NEC has a special table for calculating the required wire gauge for dedicated welder circuits. You can get away with a lighter gauge since the limited duty cycle at high settings prevents heating the wires as much as a constant load would. Of course, there's no problem with using heavier wire than required, but copper isn't cheap these days . . .