Technical Drag link design questions

Greetings, I need to fabricate a new drag link for a 1929 Franklin custom roadster. The old link is bent but the end caps are salvageable. The link itself is thick wall pipe with threaded ends. Because of the car's original design, there needed to be some bends in the link and this is where the link failed. Since none of the original body is on the car, I can make it a straight shot from the pitman arm to the linkage on the left wheel. I have some questions:

1. ***uming there is enough clearance and sufficient articulation at the end caps, is there any reason I should not use a straight drag link?

2. Other than weight and cost, is there any reason I should not use round stock steel instead of pipe? I'm thinking the round stock would be significantly stronger.

Anything else I might be overlooking here? I'm not terribly knowledgeable in this area.

 

I would use solid bar stock. But, the bends must have been there for some reason. If you haven't changed the geometry (lowered it, moved the steering box, etc.) I would seriously consider bending it to match the original.

What was in the way that they were bending it around originally?

 

Many cars with side steer around 1930 had S-bent drag links. They did that to get full turn to the left without hitting the edge of the tire, as well as missing the top of the king pin boss. The ones like early Mopars I had, were solid. I suppose they could have been forged because both ends where the socket cups and springs seemed to be made all as a one-piece type draglink.

 

Here's a photo of the steering linkage intact, prior to failure.

 

Steering link post failure

 

DOM tube is the standard on drag links and tie rods.
The wall thickness is pretty stout on the DOM stuff and that kink looks like a thin wall conduit kind of kink

 

Straight line as shown by green stick. The photo flattens the depth but be advised that there is plenty of room between the stick and the spring.

 

A straight shaft would work for sure on the pitman arm end because the pivot is horizontal. But, I think the spindle end with the vertical pivot might get close to binding. I bet that's why the original is curved. Maybe install the front joint and see how far it can rotate and not hit the end.