CNC Build: The Bed and Breakfast at 4 years and 40,000 miles

[KCStudly]

With tubing you are only going to be able to weld from the outside, so welding the back side is a bit misleading.

That said, a 100/ct full penetration weld is not strictly required (i.e. penetration from one side of the material to the other with no voids), especially on a fillet that has the leverage from the weld on the far side of the tubing to counterbalance it (i.e. stabilize the joint). You do, however want to have full fusion between the filler material and the base metal.

I would be less concerned with a little void at the root, than I would be with the apparent lack of fusion in some areas that you show (the fine lines that extend toward the edges of the fillets where they do look like you had good penetration).

To me this looks like an issue of torch pattern and travel speed, and perhaps torch angle. Another thing to be aware of with MIG is that it lays so much filler all the time that it is hard to get a hot start. In other words, if you pause travel at the start of the bead to get things up to temp you may end up with a pile of filler over a cold spot so the natural tendency is to start right off traveling, so for a short test weld you might not have the best heat range setting (i.e. the recommended heat settings for a given thickness material will vary depending somewhat on the length of the weld; tacks and short beads would need to be set hotter, longer beads build heat and can be set colder.

On weld prep, for relatively light materials, I usually grind the bevel so that all but 1/16 inch of the perpendicular face remains. In other words, if the material is 1/8 inch thk, I will grind half away. If it is 1/4 inch thk, I will grind 3/16 inch away. Depending on access, if both sides of a weld are accessible I prefer to bevel both sides so that there is an equal amount of weld on each side for more even heat/less warping. The angle of the bevel is not as critical; less angle can restrict you from getting down into the bottom while wider angles just leaves more gap to have to fill.

[capnTelescope]

Thanks, KC, for taking the time to 'splain some things.

With tubing you are only going to be able to weld from the outside, so welding the back side is a bit misleading.

Agreed. This was slowly dawning on me while I was looking at my specimens. Sometimes it takes me a while.

The angle of the bevel is not as critical...

Cool. While I hadn't welded at all in the past, I did have to deal with producing weld preps for various pipe schedules. IIRC 37-1/2 degrees was a common angle for the prep on a pipe end (75deg included angle between pipes, which I only just now realized). The 1/2deg sounded awfully precise and picky. Maybe not so much. The 1/16 edge width sounds familiar. I'll give it another try with better prep and one side only.

To me this looks like an issue of torch pattern and travel speed, and perhaps torch angle.

That includes just about everything but volts and wire speed. I'll play around with these and report back.

Thanks again! The beer's on me!

[tony.latham]

Brad:

Okay, enough is enough. I'm hearby certifying you as an amateur part-time welder capable of gluing up steel with globbed on molten steel. With one caveat: KC has nailed the critque of your welds, but make sure you understand the warping problem with welding and how to mitigate it. Your trailer chassis is not going to fall apart, but it might warp!

Tony

[capnTelescope]

I'm hearby certifying you as an amateur part-time welder capable of gluing up steel with globbed on molten steel.

Thanks, Tony. I feel all ... certified. Some would say I've been certifiable all along.

make sure you understand the warping problem with welding and how to mitigate it.

You've nailed my next concern: making the frame "flat" and "square."

I'll research that a bit before I get started. Probably whole books written on the subject. I just happened to have a copy of The Procedure Handbook of Arc Welding from Lincoln Electric. It's a big book, so maybe it's in there somewhere. Also, How to Weld is on order.

[tony.latham]

I brought that up 'cause I ..... 'cause i.... ew, this hurts... warped mine when I welded the Dexter axle brackets on. I think welding up a chasis frame, on flat concrete is a matter of spotting it together square and flat and then juicing it together. Not a big deal. (Anybody wanna weigh in here?)

Somewhere on this forum there's a post or two about welding a Dexter bracket. If I recall, you're not using a Dexter. But that's where you need to be careful, I think, and get some advice. And not from me.

TL

[KCStudly]

Certified and warped. Yup, you're in the right place here! (Sorry, couldn't resist! )

I show some good methods for keeping the frame square during assembly in the beginning of my thread. Things like clamping perch plates on to align the faces and starting by tacking the middle of the outside corners first, using the diagonal check method, then adding temporary diagonals to hold square.

It will warp, especially when welding the suspension components and tongue on (because the weld from those are all on one side of the frame). Mine moved about 5/16 of an inch, IIRC, despite the use of temporary strong backs.

Some build the cabin separate and just pull the frame to it (I think Aggie79 Tom did this). Others have used shims. My friend Karl and I used a hydraulic ram and jig table to form mine back true.

[capnTelescope]

... and get some advice. And not from me.

I'll take your advice and get some advice.

My friend Karl and I used a hydraulic ram and jig table to form mine back true.

Now there's a thought. Just glob on the molten steel willy-nilly and take it to one of those high tech body shops with a frame straightener. They'd be able to straighten that puppy right out.

I think I'll go back and review your thread.

[capnTelescope]

That long cold winter sure ended abruptly. It's in the 90's out there.

For the past week or so, I've been walking into my shop to be greeted by the piece of 1/8 ply that will be the underlayment for the aluminum skin on the hatch:

As you can see, I've been teaching it to relax and bend. The problem is that it cuts off the main route to the rest of the shop. I've been letting it do this while I did some finishing on the hatch spars and ribs. Well, I've had enough of going around the back of the router to get to anything, so today was devoted to putting the hatch frame together and moving all this to the assembly bay... er, garage. While the dry assembly went together smoothly, this time it didn't. Things got out of order and had to be taken apart, stuff like that. In the end, I won, but it was a tough fight. Here's how it looks:



Then I laid the 1/8 ply on top:

Looks pretty good so far, no? No. I was perfectly OK with the picture above, with the ends being up in the air. After I got all the stuff I needed into the assembly bay, I clamped up and found out that the skin doesn't really want to follow the hatch curve very well at all. I've got gaposis in the middle of the curve when I try to get it to submit to my will..


That's the best I got after an hour of messing with the clamping:


When I go to glue it up, I think I'll have to start in the middle of the arch and work towards the ends, the opposite of Plan A. Add some filler wood at the bottom edge, and do some hand work. I need to get some more clamps with deeper throats, too. There's that name again.

[RandyG]

When riveting big pieces of skin, we were taught to start in the middle and work out, I guess it makes sense. I like how you trained that piece to bent like that, very simple.

[KCStudly]

Um... aren't you just a little bit worried that the hatch won't fit the cabin when you go to pair them up?

The people that seem to have the best success building hatches do it right in position with the parts clamped or temp screwed into place in the galley opening; so that's the way I plan on doing mine, at least until I get one of the two hatch inner skins into place to lock everything into relative position. Thoughts?

[capnTelescope]

When riveting big pieces of skin, we were taught to start in the middle and work out, I guess it makes sense.

Once again I have made a great discovery of something that was already known. I'm still waiting to have an original thought about anything.

Um... aren't you just a little bit worried that the hatch won't fit the cabin when you go to pair them up?

No. I expect them not to. That's what all that R&D and 4 design versions was all about. The hatch skin will be flush with the outer profile and between the two outer walls. I will trim the hatch skin to the opening, so it should look perfect when done. If the router doesn't slip...

OTOH, You are correct that it is not yet time to skin the hatch. It will be easier to mount the hatch frame (to the aft bulkhead) without the skin. Glad we had this little talk. I was about to jump the gun. At least I got that obstruction in the shop cleaned up.

[lincolnlerner]

i am afraid that what he means is that the hatch will change shape and no longer match the td.

[tony.latham]

When you do attache, how about screwing and gluing, and remove the screws prior to skinning?

[KCStudly]

Aggie Tom built his hatch frame in place shimmed and clamped/screwed directly to the walls, then he attached strips of ply diagonally across the outside of the frame work to hold everything true so that he could pull the assembly off and do the inside skin on the bench. then he installed the hatch and completed the outside skin (at least that's how I remember it ). Seemed to be a good approach to me.

Mine will be a little different because my inner skin goes on top of the hatch ribs (which will be exposed on the underside). So if I put both pieces of skin on my hatch with it rigged in place I won't have access to the inside of my galley to undo the temporary fasteners. I plan to screw and glue one of the sections of skins on first, to stabilize everything, then with access thru the remainder remove the partial assembly and finish it by attaching the second skin section on the bench.

You may want to consider something like either of these methods where the frame work is built in place to fit, the geometry is stabilized by partial or temporary skin like elements, and then the final skin is installed.

[capnTelescope]

i am afraid that what he means is that the hatch will change shape and no longer match the td.

Fear not, Grasshopper, all will be revealed.

When you do attache, how about screwing and gluing, and remove the screws prior to skinning?

I'll probably just shoot some staples. Into oak?

and KC said a whole bunch of stuff. (BTW, I reviewed your trailer frame build and much became clearer. Good job of explaining. )

Since I need to clear out the shop so I can do some table saw work, I'll set up in the garage again. I need to do that anyways. Hopefully I can 'splain the plan better when I can take some pix.

8,000 views! Thanks everyone!

Stay tuned!