Straightening a Model T Exhaust Manifold

Sometimes, because of repeated heating and cooling over the years, the exhaust manifold becomes warped. Each of the four exhaust ports in the block should have a gland ring inserted in it. Each gland ring (Part #3064) fits into the corresponding hole in the manifold to keep the hole in line with the port. When these rings are missing, the weight of the exhaust pipe on the end of the manifold can pull it down and warp it so the holes don't line up with the ports, sooner or later causing exhaust leaks. At a swap meet I paid $10 for a manifold that looked brand new but wasn't straight. Having read about manifold straightening on the MTFCA forum and the Tulsa and Long Beach Model T clubs' websites, I decided to make a jig and try it myself. I went to the local welding supply and overpaid for eighteen inches of eight-inch I-beam. I found no suitable scraps in the salvage yard, so I spent almost $40 for a new piece. The yardman had just cut off the piece for me and I was carrying it into the office to pay for it when a truck drove in with a load of scrap that included exactly what I needed. If I'd been ten minutes later I'd have paid much less than half of what I spent for a new piece. Ah well, c'est la vie.  

So here are some pictures showing the project and its results. 

The raw material.

Cutting and grinding.

Filing and tapping.

The finished jig includes a 6"piece of 1/2" coarse threaded rod to exert force on the heated manifold, and a pusher with a curved end that fits against the manifold. The pusher is made of a piece of square tubing with a bit of pipe to provide the curve on the end.  A large opening in the base allows for heating the manifold from below.


Gland rings are inserted in the manifold holes to facilitate checking for straightness.

Pieces of angle iron that fit between the rings are adjusted just high enough to let the manifold slide under them. They are to hold it down flat on the jig.  

A straight edge set against the rings on #2 and #3 shows how much #1 and #4 are bent out of line.

In all my reading about this process, many have stressed that it's best to rely on only the weight of the wrench to turn the rod and force it against the manifold. As the heat softens the manifold the wrench is supposed to slowly descend solely by gravity. But I have to confess that occasionally the turning rod stuck, and I had to help it along.  That said,  I used minimum force applied very slowly.

I was very pleased with myself when I checked the manifold for straightness. 

But I found during straightening that removing the bend in the middle also moved #1 and #2 farther from #3 and #4, and I had to pull the ring out of #2 to let the manifold move. I think my answer to this will be longer pieces of angle iron that will go over the rings instead of between them.  

My pleasure at the results died when I turned the manifold over and saw what happened where I heated it from below. Obviously, I got the heat source too close.  Fortunately I have several junk manifolds I can use for practice until I really learn how to do this job. This is definitely something to avoid. I consulted a shop that does cast iron repair and was advised that having this manifold fixed would cost about the same as buying a new one.


After this first attempt ruined a 26-27 manifold almost five years ago, I didn't have occasion to attempt another straightening until recently. The exhaust manifold I had to install on my touring car had excellent threads and looked good, but was warped enough to prevent it from fitting properly. A straight edge across the first three gland rings showed that the fourth was about 3/32" out of line. That wasn't very much, so I decided to try straightening it. This time I was very careful to watch what I was doing when I applied the heat, both from above and from below. When I finished I was delighted to find that the straight edge showed it straight, and I was even more delighted to find that it was a good fit on the engine. I think perhaps this works best when you straighten a slightly warped manifold rather than one that's way out of alignment. Having done it successfully this time, I wouldn't hesitate to try it again. Some people claim that it uses up too much acetylene to be cost effective,  but I don't think what I used on this was enough to cost anywhere near the price of a new manifold.  

Updated September 26, 2017

Here's a link to the  Tulsa Model T Club manifold jig page.

Tulsa manifold jig