Fuselage jig - how flat?

[wbpace]

Not wanting to sound like a weasel (what a way to start a thread!), but just how flat should I make the surface of my fuselage jig? I've assembled it using engineered joists, so it is quite straight and incredibly stiff, but the 3/4" plywood I bought took on just a bit of a warp before I put it altogether. I swear it was flat when I bought it!

So when all assembled, along the short dimension the outside long edges come up the barest fraction of an inch towards one end. I can slip some paper under my level, so it is not absolutely flat. If you stare carefully at the picture below, you can see how the top corner curls up just the barest amount. Fortunately, this is just on a fairly small area, but it is where I will jigging up station 0 (the firewall).

In length, there is probably around 1/8" or so difference in 16 feet. I can't get it better than that because I have to position it diagonally across the (mildly) sloping garage floor. That makes the leveling geometry quite complex.

I was able to take out most of the problems, but I'm left with what I think is below my ability to shim out.

It's my fault. I had a hint that the plywood may have been a bit wet, but I got in a hurry and got it anyway. It's ACX from a quality lumberyard (not Home Depot junk). Nevertheless, I should have gone to a specialty plywood store I use for major woodworking projects that has the really good stuff. So now I am wondering if I should go get the "right stuff" from them, or stick with what I got. Perhaps just an overlay of uber-flat particle board.

I only ask because I don't want to be an unnecessary perfectionist. There are places to be particularly precise, and in other cases it is a waste of time. I'm not sure which this is.

I'll post the story on how I assembled it another time as I think it may be interesting. If I had not had this problem, it would have taken me less than a day to build a solid, perfectly straight table.

O'Bill

 

[generalgrant]

it's really had to see an issue in this photo; it could be off by a mm or a foot. If you shot it with a laser or strait edge, how off is it?

 

[wbpace]

Yeah, it's hard to see. What I am asking is what are my tolerances for flatness and twist in the table. I fussed with it a bit more and got it a bit better. I just seem to have about 1/16th inches of twist in the first two feet of the passenger cabin, making the top left of the picture just a tad too high.

Normally I would call that just fine and move on, but given how small errors can have significant impact on such a large structure (I.e. fuselage), I decided to ask.

I'm not usually this "retentive", but in this case I just don't know if this is good enough.

O'Bill

 

[generalgrant]

Hard to say what the impact will be. Only advice I can provide is this. Knowing the total investment you will make on this journey what is the real cost in dollars and time for lumber and a table now if you can't rig the aircraft sufficiently later for strait and level? How much personal frustration will here be after you spent several thousand dollars and years of time and have the result not be what you expected and hoped for? What problems could you face when your fuselage is no longer to plan?

 

[Pttim]

Hard to say what the impact will be. Only advice I can provide is this. Knowing the total investment you will make on this journey what is the real cost in dollars and time for lumber and a table now if you can't rig the aircraft sufficiently later for strait and level? How much personal frustration will here be after you spent several thousand dollars and years of time and have the result not be what you expected and hoped for? What problems could you face when your fuselage is no longer to plan?

Welcome to homebuilding! This wont be the last time you'll remake something.

I have to agree with the General on this one!

 

[oahupilot]

The heat from welding the clusters together will add significant twist and warping in the parts. Chances are you will have to straighten the frame in the end any how, by re heating the clusters and checking the frame for final fit. A lot of people approach frame building with a micrometer when the welding operation just does not hold that level of tolerance unless you have a seriously heavy jig made of metal.

You can always use a hand plane to flatten your table back to level, if it bothers you that much. Also you could just as well buy some MDF, which is supper flat and not to pricey.

The video on tinmantech website on building a 4130 tube frame really hammers home the point of not wasting to much time trying to get the perfect cope, or spend to much time building fixtures, because ultimately your goanna end up with some degree of twist and warp in the frame that needs attention. Its a good video for watching just how smooth, simple, and quick the workflow can be.

 

[Raceair]

I agree with Oahupilot. Putting heat to the clusters will make a more significant change than a table thats 1/16" out. Get the table as level as you can, shim a leg or two if necessary, and go from there…..Ed

 

[subnoize1]

My sad two bits with this topic comes from watching the tube I have practice welded on curl and walk all over the table. The best way I could explain this to somebody who has never welded anything is the metal comes alive at about 600 degrees.

No mater how straight you get it in your jig, once you take it out and weld it you will be straightening it back out for weeks afterwards.

Get it close is what the old timers have told me, fix it after welding. You can always weld on more metal and grind off what you don't like.

In fact an old guy showed me his Cub and the engine mount and we measured it and went to his next door neighbors and measured that one and they were very different but both were factory build. Probably even in the same jig. It wasn't the engine mount that was the problem, it was the fuselage had warped and was straighten AND THEN the engine mount had to fit the imperfect fuselage on one side and near perfect engine on the other.

John Monnett said once that these planes aren't designed to be built by professionals. They are some what tolerant of mistakes. Like the placement of the tail post on the Sonerai by true center line measurement or by counting stations (and losing a half inch).

Anyways, my plane is going to be safe but ugly and I am fine with that. It will be straight flying but probably kind of wavy looking.

If you fret over being perfect you will never finish. Do the best and have fun is my goals. Maybe I will end up with something that will take the bite out of the financial sting of flying while I build my next, more perfect plane!


JB

 

[wbpace]

First, I agree with everything said above. Given how much the I've been told that the fuselage will move around due to heating, as mentioned by Oahupilot and Raceair, my inclination was to leave well enough alone. But I also agree with Generalgrant about the cost/benefit of changing my table top, and I am very familiar with having to remake things (sometimes several times) as pointed out by Tim.

In the end, I decided that it would probably gnaw on me forever, so I replaced the top with 3/4" ultra-flat MDF for only 40 bucks. So now my error is down to just 1/32" or so, and I have my peace of mind. And I challenge anyone to do better than that on their total flatness!

So for those interested in using this method of building a table, if you do it right, you can make this in a single day - in fact, just a few hours. The only reason not to is because you want "EAA tables" to eventually convert to permanent workbenches. The relatively tall joists may make this not the best option for that purpose.

Would I do it this way again? You betcha! I don't need them for long term, and it was very fast, even with my rework. If I had to do it again, it would probably take a grand total of no more than 2 hours to assemble plus 2 hours to fuss with overall leveling (only because I have to level on a diagonal to my slightly sloped floor garage floor, which complicates it). You can't beat that with a stick! It took far more time to think about it, find a supplier for the joists, and purchase the materials.

Total cost was around $271 including the extra $40 for the replacement top and renting a truck to transport the 16 foot joists. You can do it cheaper with dimensional lumber, but I would question your sanity !

The four keys to the whole thing is to:
[list type=decimal]
[*]use pre-made 24" high sawhorses from Home Depot (~$35 for two); That way they will be of uniform height and will place the deck at a very comfortable 35" off the floor
[*]use the smallest engineered joists available (usually 9-1/2") ($1.79 / ft or ~$60) because they are laser straight to begin with - absolutely no shimming of the top - and you get them in one 16' long (or longer) piece
[*]use 3/4" "fine grain" MDF (the kind without open cells like you get with melamine) (two 4x8 pieces for $80); You need to create a box - top, sides AND bottom - for rigidity and stability
[*]use drywall screws for everything ($7) - you won't even need to countersink for them, just drive 'em in!
[/list]

In other words, around 200 bucks for the whole works (if you don't need to rent a truck).

This video tells you all you need to know. It was my muse, my guide, my inspiration: https://www.youtube.com/watch?v=BVz4cbpICks&noredirect=1. I will echo what he says in the video that any other way of building your table "just doesn't pay!"

Thanks to everyone for their thoughts!
O'Bill