I was test fitting my doors before painting them and noticed this:

Driver side.

Click to view attachment

Passenger side.

The gaps were ok before I welded in my Engman kit. I must have gone too
fast and ended up shrinking the passenger side??? I also did a lot of work
to the passenger side door jamb.

I measured the distance from the front suspension point to the rear inside
mounting point and got 207 cm on each side. So maybe the door jamb
sheet metal is screwed up and not the whole car (fingers crossed).

The front gaps on both sides look ok, about 3/16. The doors are the exact
same length. The drivers gap is 5/16 and the passenger is 1/16.

Now WTF do I do???

John

Wow... Know any really big guys? I have heard of the Engman kit doing that. McMark was telling me recently that he only welds part of it.. Not sure which part though. I have also heard of putting bracing in the door openings to hold the integrity. Not sure what to do once the shrink has happened.

Ayupp ...

Note the bolt in braces in the pic below.
Andy

Click to view attachment

Put the car on a lift, open the doors and use a torch to heat up the long on the side where it shrunk.

You might be able to get some of the shrinkage back ...
Andy

I did use a brace bolted to the upper seat belt mounting point and the lower
hinge plate. I think the brace I used must have been too weak. It was made
from 1 inch square tubing welded to a (pretty small) turn buckle.

It is also possible I just screwed up the sheet metal at the rear of the door
openings because I did a lot of welding in these areas.

I should have re-fitted the doors while working on the door jambs.

Should I reassemble it and see if it aligns correctly?

Any measurements I could do using simple tools to see if the whole car
is screwed up?

Arrrghhh!!!

John

> Put the car on a lift, open the doors and use a torch to heat up the long on the >side where it shrunk.

> You might be able to get some of the shrinkage back ...

Yeah Andy, I was wondering about that. If I evenly heated the side that was
shrunken and then braced it with a big jack screw or something.

Anyone ever try that?

I really need to know first if this is a cosmetic problem or a chassis warping
problem...

I measured the distance between front and rear suspension pickups diagonally and got
221 cm one way and 220.5 the other. This is less than .22 % difference, so the car is
pretty nearly square. If one side of the car was longer than the other, this would not
be so close. I am thinking this is a sheet metal problem not a chassis problem, or is
this just wishful thinking???

http://www.914world.com/specs/bodydims.php

I feel your pain. I repaired my 75 when I got it in the early 80's. I had no experience welding on cars. I learned to MIG weld on my car. My gaps are just like yours. I replaced metal around the passenger long at the jack points. I also replaced the rear quarter panel. I didn't use any bracing. I would really like to repair my car but I am not sure the best way to go about it. One suggestion I received was to stretch the passenger long. Using clamps at the pinch point at the bottom. Maybe weld some brackets top and bottom and use either jacks or screws to stretch the long while using heat. I have tried nothing to date. I am just living with uneven gaps.

Thanks man...did you have any trouble getting your car to align?

After lots of careful measurements, it looks like the rear fenders are
exactly the same length. The problem is that the passenger long is .375 inches
shorter than the drivers long. This checks out with the diagonal measurements
I took. 221 vs 220.5 allows for a difference of .21 inches in length. I am sure
there is some error in my measurements as well.

I suppose I could go Cellete bench or other frame machine to fix it but only
if I can't get a good alignment done.

Anyone know roughly what this costs? I think Andy had some of his cars
fixed once...

Maybe a 20 ton hydrallic ram like they use for straightening crash damage
would help? That might just cause lots of other cracking though...



John

It will be very difficult or impossible to find any metal on the outside of the car to push on that will be strong enough to not crush under the pressure of a jack. A Cellette (sp?) bench or live with the gaps adjusted as well as possible look like the only real options now IMHO.

According to wikipedia, steel expands .00000645 inches per inch per degree
F increase in temperature. The longs are 45 inches in length, so if I heat
them 1200 degrees F they should expand just right to remove the shrinkage.
Acetlyne burns at 5700 degrees F in oxygen. In theory, I could carefully
heat the long up, probably hotter than 1200, brace it, and let it slowly cool.

I may try it if someone else has done this and had it work...

It might stain my new paint though.

Maybe it would buff out with Mothers.

John

I've got a question for the author of this string and also others that have had the same problem while welding in the Engman.

Is it a matter that the longtudinal is shrinking or that it's taking a curve??

In other words, is it just the door gap at the top (which would be curvature) or the gap shrunk on both the top and bottom equally (which would be shrinkage in length of the longtudinal.

I've always assumed that it's curvature that people are talking about. I also assumed that it's due to the fact that the added bracket is an L shape and covers the top, which is welded. (the bottom of the long is partially prevented by shrinking due it's attachment to the floor pan.


I'd really like to know due to wanting to install one of the kits.

Wes

Wes,

This is a really good question. I went out with a level and measured both of my
longs, and guess what, both are arched. The middle of the longs are higher than
the edges on both sides. I estimate there is 1 mm of arch. The bottom edges
of the longs are also arched, but not as much, maybe half as much. I can
measure more carefully tommorrow.

You may be right that the top of the long shrinks more than the bottom because
of the floor pan and also because the Engman kit does not connect to the bottom
of the long. Maybe that means I need to apply heat only where the Engman
kit is to restore the correct shape, even though I did a lot of other welding
to this long.

Keep in mind that I am not knocking the Engman kit at all, it is a piece of art.
Mine fit perfectly, I just got too aggressive about welding it in, even though he
tells you not to. I will put them in my next car, but I will weld a lot more slowly.

My car was in really bad shape when I got it and this is my first resto project
so it is not surprising I F-ed it up a little. Don't make the same mistake!

John

I had that happen as well when I installed some steel. I talked to my bud who painted the car and hes got a dozen pull-pots awaiting me in Washington. He said I could heat things up but his way would work better. My car has a little extra steel, we installed 1/4" inner liners. I think it would be hard to get those anvils hot enough to move with out pulling a bit.
You need to just take your time and get it right, looks like a nice car !

The A pillar, at least the part that the door hinge attaches to can be moved. If you are careful and make two vertical blocks that are custom scribed to the front and rear door jambs, you should be able to use those and a hyraulic jack to push the door open.
That is how I would fix the side that is tight...
While you are pushing, pull out on the rear fender lip at the wheel well.

Rich

Rich,

Thanks for the suggestion. Have you tried this before?

Maybe what you suggested plus some gentle heating?

John

Yes I have done it.
Push on the outer rear b-pillar while pulling out the rear fender well.

If you doubt me...go hang the door...pull on the fender right at the wheel...pull it out gently 90 degrees from the center line of the car...as if you are adding a fender flare. you will see the gap change...you just need to use a jack to do it slowly and keep it there.

RIch

OK, cool, I will try it tomorrow.

Thanks....John

I would think the passenger side will be easier to adjust than the driver side. Rich's way sounds doable. I don't know if the paint will not get screwed up at the rear jam or the top of the long before you get through. You might try using a 2x4 and lay it on the top of the long and use a bfh and lower the center of the long (make it longer) maybe put jack stands under the jack points so the impact will work better, but 3/8 is never going to happen. You might as a last resort try grinding some of the top rosettes out of the long. Another last resort would be to weld some anchors to the top of the long and use a big spreader while heating the long. It seems to me that a added support at the roof would also help as it's a lot farther up in the door opening to keep the roof straight as you are doing all of this. I will weld some braces to mine when I try this as door openings changing are a real time consuming problem. Have you measured the roof opening? I would do that and see where it's at before you start a game plan. I hate to say it but the rest of us are learning a great lesson at your expense.

The passenger side seems square so maybe you can move the door forward enough to get the door to clear and live with it. The driver door is a good challenge if the top opening is still square but maybe it's also wide and you can just pull the top of the door opening together. I would not use the windshield frame to do any pulling. You could use the targa seat belt mount and pull to the top hinge for more leverage to close the gap. Be careful that you don't just pull the jamb hinge to the rear and screw the front gap up.

Charlie,

Thanks for the suggestions. I was hoping to help other people to avoid this
problem by show casing my own error.

I also thought some World member would know what to do to fix it,
and I still do.

I am pretty sure I read on some forum that a guy had this problem and
some old time body men fixed it by heating it up but I can't find the post
anymore.

OK, just for reference, I spent about four hours on welding in all three pieces
of my Engman kit. I used a Hobart Handler 140 amp machine set on 3 for heat
and 50 for wire speed (this is pretty hot for sheet metal on this machine).

I welded all the rosettes, and also seam welded a lot of the openings, edges,
and joints where the pieces meet. I did most of it in one session, but some of
the final seam welding I did the next day. I welded 4-5 rosettes and then waited
5 minutes or so to cool off, then welded some more. It was probably a 75
degree day when I did it.

I tried Rich's suggestion of using a jack to spread the door opening. It maybe
got me another 1/16 or so but was starting to buckle the rear fender near
the back of the door. I was using 2x4s on the door pillars. I probably need
to make carefully shaped cauls (pieces of wood) that match the door pillar
shape to spread the load more and have a second person pull out on the
rear fender.

I can live with this if I can't fix it...first time I hit a tire wall at Sebring I hope
it is on the passenger side.

Thanks again for all the suggestions...

John

About thirty years ago I torch welded the floors in a early military jeep. I did a lot of welding without stopping at first. I welded a seam about a foot long and steped back and admired my progress and it seemed ok. After moving to the other side to do another I got about 1/2 way through and heard a loud pop like someone hit it with a hammer and the first side split about 6 inches beside the weld. Being a stubborn sob I immediately rewelded it and went back to the side I had moved to. When I finished the second side 12 inches the first side poped again so I stopped and had a dr pepper and cussed a little. From then on if I couldn't put my hand on it it didn't get welded. You can take a 2 inch piece of pipe 20 ft long and weld a 12 bead along the top side and stand back and watch it bow down.

I'm trying to get my mind around this whole "shrinkage" thing, so I'm going to post this to try to make it clearer. (in my simple mind)

Let's pretend that we take a 3 foot long length of 4" square tube and support it on it's ends (just sitting on saw horses). To keep it simple, let's assume that it has no weight.

A torch is used to heat the top of the tube till red.

While hot, the top of the tube expands and the tube bends with the center moving upward.

It's allowed to cool down slowly and no force is applied to it.

As it cools, the metal contracts and the center "bump" goes down.

I was under the impression that it would not go all the way back down to level and most assuredly wouldn't go lower. Am I wrong here?? (the closing of the door gap at it's top would indicate I'm wrong)


To further this;

Let's assume that we start with a new tube.

Heat the top with the torch until red and then while still hot, throw cold water on it.

Prior to cooling it off fast, the center would have moved upward. Cooling it off fast would result in it contracting more, and would result in a downward bend. Once again, am I wrong here??



What I'm trying to get clear in my mind is what is happening while welding on the top surface of the Engman angle. My guess is that depending on how the car is supported, the gap at the top of the door would open while the welding is hot, and then closes up while the weld cools off.

Wes

Sorry that you are having this issue, but you are really scaring me now. I have pretty much the same brace in place right now and I just finished welding in place the the new outer longitudinal on my passenger side.

Right now everything if fine (measurement wise), but I guess I need to remove the braces to see what the measurement are without them in place.

In my experience, this is the problem, John. I rosette weld, and that's it. No outer seam welding. Just a few quick tack-welds.

See if your top fits okay. The greatest deflection will be the top of the window frame vs. the top of the roll bar/targa bar.

I've used a Porta-Power to restretch a shrunk body. It worked well.

Might have been me. Though it wasn't nearly as bad as what you're dealing with.

http://www.914world.com/bbs2/index.php?sho...80724&st=80

See post #85

pete

When I welded my Engman kit, I measured my door gaps (using calipers) both before and after welding. As slow and metodical as I was, the gaps still closed up a little (they were perfect before I started). What I did was to weld a few rosettes one then side, then move to the other side, make a few welds, then back to the first side and so no. When the panel would start to get hot, I'd stop and let everything cool down before resuming. I used a TIG to weld in my kit which I'd never do again; Way too slow and tedious but it does keep the heat input to a minimum.

Wes got a chance to look at the end result of my Engman install, but to help answer his question "heat shrinks." Anything I've ever welded has moved or shrunk as a result of welding. Scab plate installations (such as the Engman kit) are always prone to create shrinkage in my experience.

Andys

Maybe you could do some milling and shimming of the door hinges ?????I dont know just guessing or think`n outloud....good luck

Mark: I did a fair amount of seam welding on mine. Maybe 25% of the tops
and less of the bottoms. I also did the seam welding in a short time. Part of
the problem is that I had finally learned to make a really pretty seam weld
so I think I just went overboard.

The top still fits pretty well. It takes a little effort to lock the front latches.
I am going to check it again to see if the windshield frame is bending. Man
worrying about that is going to keep me from sleeping at all at work today...

How do you use a porta power to stretch a car body?

Pete: I am sure that is the thread I was reading. How did your guy know
where to jack up your car? I wonder if your car naturally stretches when
on jack stands because of the weight overhanging the front and rear jack
stand points? Maybe you need that stretch to counter act the shrinkage?

It would be nice to get this down to a repeatable science, for the sake of
future 914s...

John

Mr. C: I was wondering about that too. If I needed to add some shim plates it
would be easy, but I need to remove some metal. I need negative shims....

Andys: Makes you wonder if all those GT stiffening kits are doing to peoples
rear wheel areas. They are not as big, but I bet some shrinkage is going on
that is just not very noticeable. Maybe not enough to matter???

John

I was think`n you could mill the hing on the P-side some and then shim as needed?the other side looks like you could just shim ....but then hows the glass and the top gonna fit ??....

Experimentation + years and years of working with metal.

He welded it cold and was super mindful of the potential for problems others had experienced -- and still got shrinkage. Fortunately, he knew what he was doing and was able to fix the problem. Both doors fit up nicely and close perfectly. The final test will be the windows...

pete

Slim line hinges may work....hmm..

Great, windows. Sheesh, I was not even thinking to worry about those...

Maybe I will need some lexan windows to save weight...

Good thing it never rains here in FL.

John

I think something to think about would be to do one side at a time.

Put jack stands at the front and rear of the door opening.

Put a floor jack at the front so it just barely supports the front weight.

Put a floor jack at the rear so that it also barely supports the weight. Then lower it slightly.

Remove the rubber seal on the door, close the door and measure the gaps. That way you could close the door while doing the welding to see how stuff is moving around.

My guess is that while welding and the top of the long is hot, the rear of the car would sag downward and touch the floor jack you have there. Don't raise the jack!!!!

As the long cools, the rear would raise again, off of the jack.

The weight of the cantilevered rear section should limit how much the top of the long shrinks.

What I'm getting at is that you should maybe "preload" the longitudinal.

Adding a brace, although helpfull wouldn't preload the section (unless you screwed it outward real tight) and as a result, when you remove the brace, it would spring back and close the top gap.

Think about this; when a frame shop is trying to get a frame straight, they pull (or push) it past the point where they want it to be.

Of course, this is just me talking out the side of my mouth. I've yet to weld in an Engman.

Wes

By doing what?

Wes

John,

Your post was freaking me out all weekend. I had to come in to my shop and put the door on my factory 6 #41 to see if I had done what you had done. My car had perfect longs but we put in new floors pans and I also put in the Engleman Inner long kit. It it a great kit.

I welded a short time and rested a long time and them went on to a different area and came back to an area near it in 1 hour. It took a couple of hours to weld one side but I had work to do in the shop. I fired up the welder 4 times for one side.

This is for other people to know what not to do.

“Andys: Makes you wonder if all those GT stiffening kits are doing to peoples
Rear wheel areas. They are not as big, but I bet some shrinkage is going on
That is just not very noticeable. Maybe not enough to matter???”

I have found a lithe shrinking in the inner fender well near the batter box but it will not matter.

Bob

Click to view attachment

Basically, I think he heated up his previous welds as he applied pressure with a jack. IIRC, I think he did this OFF the lift...but it's been a long time since we did this.

I'll try to go over to the shop this week and ask him about it!

pete

(edited!)

Bob: Your door gap looks pretty good. Sounds like you took your time.

I think my big problem was being much too aggressive on the seam welding.

Wes: I think you are on to something. I wonder if Pete's body guys did not
do exactly what you are saying. Jack up the middle of the long so the weight
of the front and rear of the car are trying to bend the ends of the car down.
Then heat a little so it sags the right amount then let it cool.

My car was pretty much a throw away, but I still want to make it as good as
I can. I also want to follow the medical rule of "first, do no harm". I may
need to think about it for a while before doing anything else to it.

Many people have installed this kit without any problems, it would be interesting
to see how many did it on a lift, on the car's wheels, how long they spent,
did they seam weld at all.

Don't mean to scare ya out there, I think this is pretty much my fault for going
too fast and doing too much.

John

So,
To use a hydraulic unit to spread the door, you need to make a perfectly matching cawl, scribed to the door jamb.
The front cawl should apply pressure to the exact spot that the hinges are mounted.
I have done this with the door mounted and the door stay removed.
This allows you to open the door wide enough to apply the pressure, and once the pressure has been applied, you gently pull the rear fender.

Then remove the jack and cawls and test the door.
Rinse and repeat.

On the side with a large door gap, if it was me, I would weld on some welding rod to the back of the door and grind that down and call it a day.
This is a standard old school bodyman trick...both of them.
Rich

Rich,

Which part of the rear door jamb do you rest the caul against, the inner
part where the door striker is or the outer stepped edge? (If that makes
any sense).

Thanks,

John

outer edge so you are pushing against the fender (which will move) and not against the inner b-pillar (which will not move)

What you are doing on the tight door is opening up the b-pillar and hinging it open a bit.
If you place the cawls so that you are pushing directly against the front hinges, that area of the a-pillar also moves.

The KEY is to cut the cawls so you apply all the pressure to the front hinge line, and the outer section of the b-pillar/door jamb.

You can also grind a bit off the back of the door, but you will need to weld up the door skin afterwards.

Rich

Adding the material will fix the small door but I'm not sure the window will go back far enough to seal at the targa seal.

It sure would be nice if somebody stepped up with a "saws-all" car that could be played with and documented.

Wes

Sorry for the continued hijack John, but I figured I would reply to my post above. It sounds like you are getting some good advice on what to do.

So I removed both of my braces this morning and measured before and after. Before was fine and oddly enough when I removed the passenger brace it seemed to come off relatively easily. But... I seem to have lost about 2mm on the passenger side. I put the brace back on (no change to the turnbuckle and it goes on relatively easily) and it still measures -2mm. But with just maybe a single turnof the turnbuckle it's back to the correct measurement. It doesn't make sense to me about how easy to brace is going on/off, but regardless it shrunk a small bit.

Bottom line is that I think I didn't have enough pre-load on that brace and it allowed it to tighten up enough so that in a relaxed position it lost 2 mm. But I am not worried as I am putting a full cage in the car and right now it seems to be easy to put just a bit of pressure to get it in the right position. Which I think I can accomplish with the cage. When we put the cage in, we are going to have to be very careful about how we do it so that we correct some of this as well as not make it worse.

This is where Pete's old post that he linked above was interesting. I think it shows just how much this car will flex and gaps change just by loading the body in a different way.

Richard,

I think I made the same mistake when I put my brace on. I think I just made
it a little snug in the expanding direction, when maybe I should have pre-loaded
it a good amount. I think using two braces like Andy's picture is a good idea
as well. I think using heim joints or serious turn buckles is also important.

I remember in Jeff Hail's thread he broke a forged turnbuckle due to shrinkage
and put on a heim joint instead. Where is Jeff? I think he has one of the most
valuable threads for welding repairs.

John

John,

I think part of the problem is that if I knew it would shrink in that direction I would have pre-loaded it. But that is 20/20 hindsight. Depending upon the welding order, location and technique IMHO, it probably could cause the door gap to shrink or expand. So pre-loaded may or may not be the right answer. Maybe with slightly different technique it may have been just fine with the existing brace setup?

I think most people are just concerned that they car doesn't fold in half when you have too much of the structure removed. And that is true, but there is also the second issue of having to make sure it holds that position when done. I suspect that what has to be done is to be very careful and check multiple times as you go to make sure you are not causing it to move too much either way as you put metal back. Maybe once it's structurally OK (repairs are tacked in place, but no final welds have been made), you remove braces and check measurements? If the measurements are off, you put the braces back and pre-load in the right directly (pull or push) and then weld in a strategic and slow order and correct gap problems as you go?

And as Pete's post indicates, how you suspend the car also will affect that measurement. Right now I have the car on the rotisserie suspended from the bumper mounts. Ideally I guess I would want the passenger side (that has had surgery) match whatever measurement I get on the unmodified driver side (assuming the driver side is healthy).

Good luck on getting this resolved on your car!

Richard

Welding never spreads any structure. It will always shrink on cooling. Prestressing will probably be the best help with proper support. There should be no doubt if the car is supported only at the very ends it will sag. The longer it sets the more it will sag. Even roof trusses sag over many years and the 914 is not a truss design unless it has a cage.
When it sets on it's suspension points they are much closer than the bumper mounts so some of the weight of the body is not involved in the gravitational pull. If a rotisserie is used along with the door gap supports maybe a support along the bottom of the long to the floor of the shop would help but the welding process is the real answer to the shrinking.

I know fg is not the same but my friend that does early vettes has several ladder supports that bolt in all the body mounts to keep their bodies from sagging when they are on a rotisserie.

A lot of the problem is there is no roof on the 914.

My other friend a fabricator for his entire life knows how much to open a 90 on just two legs of tubing, on different materials, to get them close to 90 when he's through welding the inside of the corner. SS pulls the worst. Usually he gets really close but sometimes he will need a spreader. If he can use a jig he will tack all the inside sides of a square then weld the sides and outside then last weld the inside. Working on all four corners equally. All while it's clamped solidly.

I guess along with speed it is more obvious why the mfg. used spot welds instead of seam welding.

Technically correct, but IMHO not accurate for the situation I am trying to describe. For example take a pipe or box tubing and bend it into a semi circle. Now weld on the outside of the bend. It will shrink on cooling, which will pull on the outside of the bend (try to straighten the pipe/tube) which will increase the gap between the ends of the pipe. Effectively spreading the structure.

So in our case the door opening is the structure. Weld too much on the top of the longitudinal and it pulls the gap together. Weld too much on the bottom of the longitudinal and I can see that it could shrink the bottom enough to increase the door gap. Granted I think it's much more likely on the 914 to cause the door gap to shrink than grow.

The fact that the engman kit covers the top and inside means all the heat went there. There is not the same thickness on the bottom to get the same amount of shrinkage as the top, if the bottom were heated. He did say there is 3/8 difference in the longs length between the sides now also so bowing the long down will not make the top of the long 3/8 longer I don't think. The pass. side appears straight but short. Heating the driver side bottom might help close just the top of the gap though, it might be worth a try. My experience with heating to cause shrinkage is it really needs to be red hot to work and I don't know what that will do to the heater tubes or any coating on the inside of the longs that is there now.