Looking to do a front axle flip on a 64 g10

You may have the right idea regarding the attack angles, so here is a couple more measurements that I hadn't pursued.....
The drag-link hole in the pittman arm at direct vertical is three inches above the spring hanger bolt and three inches AHEAD (your diagram shows it behind).....
Now, I need to stop right here and state that straight vertical pittman arm placement is the halfway point, lock2lock, but isn't necessarily the factory clocking.....I was using vertical clocking to describe relative steering box location because it's easy to describe and is how mine is set up....
Again..... Bump steer is only really an issue closest to the center of about 60 degrees of steering wheel travel while you are trying to go straight at speed.....
I'm intrigued by the idea that the arcs of the two drag link arms would be set to cross over 90 degrees at different points in their travels to diminish the bump.....
I still can't think of a way to mathematically prove the concept short of making physical models using the measurement parameters that will work with what you are doing...
Maybe make a fixture that can hold the pittman arm end of the drag link in fixed locations at different heights and distances and, using an adjustable drag link, try the different positions out by bolting the drag-link into different positions on the fixture and bouncing the truck to see if bouncing turns the wheels...This would establish where the drag-link hole in the pittman arm needs to be...What I'm visualizing as this "fixture" is a steel plate fastened vertically and longitudinally in basic alignment with the pittman arm hole that can be drilled with various experimental drag link actuation locations to find the placement that most reduces growth and shrinkage in drag-link length (bump steer) as the suspension moves vertically....

Well here's an update and a mea culpa. I had to cut up my floor and the gear box housing, start over, and finally find the sweet spot where there's no bump steer...and it ain't where I thought it would be.

But first a return to the steering issues in that last diagram, unrelated to bumpsteer: I had thought that the pitman arm should be straight down, which then led to a total baloney explanation for the steering arm being angled slightly ahead of the axle. The more likely explanation is that in the stock design, the drag link mounts on the outside of the pitman arm, and then angles inward as it descends to the tip of the steering arm. The slight inward alignment puts the drag link at a 90 degree relation with the stock steering arm when viewed from above. I had already heated my pitman arm, bent it outward, and made room for the drag link to fit from the back side to run parallel with the frame, thereby permitting slightly tighter right hand turns before the tire rubs on the drag link. I was missing the obvious need to set the steering arm in a 90 degree relationship with the drag link. All my reasoning about equal X and Y angles for more balanced steering feel was thereby a bunch of hooey...I apologize if this has led anyone astray, and at this point I can't delete that last diagram from 5/15 or do an addemdum.

So what gets the steering centered?: 90 degree relationships at both ends of the drag link...the pitman arm is not vertical as in all my diagrams...it's at 7 or 8 o'clock and at 90 degrees with the drag link, just like the steering arm needs to be at the other end. Lo and behold the steering becomes balanced with the steering wheel centered at the gear box mid point, equal swings right and left, equal adjustments of the pinion stops. Since my draglink is parallel to the frame, my steering arm had to be rebent, putting the draglink mount point directly above the axle centerline, establishing a 90 degree relationship, and doing away with the "Y" angle in that last diagram. All's well...seemingly.

So what fixed the bump steer?: keep in mind the drag link rotation point is at the end of the pitman arm, and the pitman and drag link should be in a 90 degree relationship with the van aimed straight, as though driving down the road. The drag link then rotates from the pitman pivot point to follow the upward and rearward motion to the spring / axle / steering arm. By making a larger gear box side plate I could move the whole works around by trial and error, clamping it to the frame in different spots, and when I raised the gearbox another 2 inches,  moved it an inch rearward and then adjusted the drag link to fit, everything finally came into agreement. Finally, no rightward tweak of the steering wheel and no leftward turn of the wheel hubs during suspension travel. Given that my axle is an inch rearward on the spring, probably adding a touch more to the horizontal motion, and despite moving the gear box an inch rearward, my drag link still ended up an inch longer probably because the gearbox needed the higher location and thereby a longer reach to get to the steering arm. I'll add one last diagram at some point in the future to show why this is, as well as a couple of pictures when I get the floor and housing all tidied up again.

The bottom line is that you can get all of this to work in a side steer arrangement, but the gearbox needs to be raised any time you decrease the vertical distance between the draglink attachment points and thereby flatten the downward angle of the draglink. e.g. flipping the stock axle and/or installing a 4x4 axle above the spring. It should be a simple calculation: by whatever amount the van is lowered with a flipped axle (or the steering arm raised), the gear box needs to be raised by at least an equal amount. With an above-the-spring 4x4 straight axle the steering arm is above the axle and pinion centerline compared to a flipped stock axle, which really flattens the drag link and makes for some interesting fab work. My gear box is now close to 6 inches above stock.  

Again, my apologies for the errors in my diagrams / thinking and any confusion this has caused. Should there be a next time, I'll complete the project, and then mention the trials and tribulations. Some times the best we can do is to keep on living and learning...I've noticed that it generally goes better when I keep my mouth shut.

Cool that you have solved the puzzle!
Anxiously awaiting the results of your first test drive over a wavy surface!.....
Once real world testing has verified your solution to bump steer, maybe we can come up with a math/geometry explanation so that future builders can have it a bit easier....

Raising the steering box to attain a downward angled trajectory for the drag link defies my sense of logic, yet it does copy the factory model..... I'm just not sure WHY that tactic would work (?)
Be Good!

Well...I'm fairly confident it's a done deal, and have made a new mount and cover to house the steering gear box 6 inches above the stock location...and it's all welded in. There's no steering wheel movement in the cycle when lowering it from the front bumper to the floor, and then jumping up and down on it, which gives a good 3 inches of travel as it sits, though still shy of the stops by an inch. My scrawny 155 pounds isn't good for more. No shocks on it either.

So from the bumpsteer point of view I'm happy, but it took my clamping the gear box side mounting plate all over the place before the sweet spot was found. The working end of the pitman arm, where the drag link attaches, is exactly 6 inches higher and one inch rearward than stock, with the drag link center to center length ending up exactly 31 inches long (stock is 28). This might sound unusual but results from a more rearward location of the steering arm end, just past the axle centerline, but that relates more to steering issues...see below. As for the bumpsteer that started all this, and in response to AzDon, to me it's just a matter of geometry...I'm not sharp enough to put any math to it...the circular swing of the draglink from it fixed position on the end of the pitman, when driving straight down the road, has to very closely match the steering arm movement as it travels upward and rearward with spring travel (green arc). The green arc might even be closer to a straight line, but there's only one spot where the working end of the pitman produces a drag link arc that best matches it across the 6 inches or so of suspension travel...otherwise you get bumpsteer.

And once more there's egg on my face for abandoning the X and Y business when setting up the actual steering. The Borgeson box has a 24:1 ratio compared to the stock box 22:1, with 6 1/2 turns lock to lock, or 3 1/4 turns in each direction from center. (Gear boxes are designed to be tighter in the center portion of the worm gear, so that's where you really want to set them up.) That's a bunch of turning, but with the pinion stops set equally so that the left front tire stops short of hitting the drag link in right turns, it worked out to 2 1/2 turns in either direction, for a total of 5 turns lock to lock...I don't know what stock is, since I took everything apart 4 years ago, but I  hope it won't feel too slow, or take too much effort. The problem to solve since my last post was getting the balanced 2 1/2 turns, as this directly relates to the angles that the drag link has on its ends to the pitman and steering arms. I welded up my pitman, clocked at what I thought was 90 degrees with the intention of heating and bending the steering arm to match. Trouble was it would take 2 1/4 turns to hit the stop in one direction, but 2 3/4 turns in the other: one direction had a shorter distance to cover, which meant I had to bend the steering arm off of 90 degrees, the working end where the drag link attaches ending up a couple of degrees rear of the axle centerline. Closer scrutiny revealed the pitman was a couple of degrees shy of 90. This puts the X and Y stuff back into play, and rather than the actual angles, it's the angular distance that they determine that has to be made the same. The angles have to be slightly different as the pitman is 5 inches center to center, while my steering arm looks to be 6 1/4 (stock is 6 1/2). This was purely a trial and error exercise when it came to bending the steering arm, making slight adjustments in the drag link length, to get equal turns right and left, and you got to have an adjustable drag link to dial in suspension design changes like this; the factory can then mass produce everything at the proper lengths once determined. The other egg on my face was in thinking that the stock drag link must have mounted on the outside of the pitman, but in welding the stock arm to a new center the taper faced inward, so I assume the stock drag link runs parallel to the frame and mounts inside of the pitman, in the same manner that I set mine up. This leaves unexplained the less-than-90 degree position of the stock steering arm in front of the axle centerline, unless the pitman has a greater-than-90 to balance it out. This will remain a mystery to me as I don't have a stock suspension to look at; I will gladly be "disabused" of all such notions in exchange for a better explanation. Nonetheless, it ended up all balanced, and I hope never again to light up the torches to bend that steering arm.

Lastly, as can be seen in the pictures of the gear box housing, that's a big mess of metal in the middle of the floor. There is room for my left foot, and I'll cover the side hump with a rubber pad to act as a foot rest. It will have to do. So the moral to this story, at least as it relates to flipped axles, or in any change that decreases the vertical height between the ball joint centers on the drag link, you have to do something to restore the related angle of the drag link as it drops to the steering arm. The only way I could think of was to lift the steering box itself, and thereby getting back to the stock geometry.

I can fill in some of those blanks for you......The factory axle handles only about 5.5 inches of lock2lock travel and the factory steering box is capable of about seven inches of travel at the end of the pittman arm....
You want the spindles to hit the axle stops well before your box runs out of travel so you can push against those stops....
It's also okay if you can turn a bit farther in one direction as long as the steering box is centered....
The drag link on second gen DOES attach on the inside of the pittman arm...


That's a lot of stuff in the cab with you, but you did a clean job of it!....
For anybody else considering your solution, they will be giving up their clutch pedal hardware to accommodate the drag link....

After re-reading your post, I have another comment (or Two).....

Your slower-ratio steering will result in more turns of the steering wheel to perform the same steering movement, which will act as power steering.... If you don't like it, use a smaller steering wheel....

Bump steer is  only an issue near the steering center.....When you are going straight down the road at speed, you don't want every vertical movement of the suspension attempting to change the position of the steering wheel in your hands....It's less of an issue while making turns...
So, Yes, You basically want to consider the drag link attachment point on the pittman arm as "stationary" for the purposes of determining and eliminating bump steer.....

Another feature of the downward trajectory of the drag link that I hadn't considered is that "positive caster" tilts the arc of travel of the steering arm from higher when forward (left turn) to lower when rearward (right turn) which may or may not affect bump steer and may be part of the reason for the downward trajectory of the drag link.....

I'm super-stoked that moving my engine back 18 inches allowed me the space to convert to cross-steer, which side-stepped all of these issues and resulted in zero bump-steer....

Lately, I've been considering full-hydraulic as a possible cool way to go....

Well, It's been a year since this thread paused, so I thought I'd bump it so that anybody considering an axle flip can see what a complex subject can be with the factory side-steer drag link setup.......
As stated in the thread, I solved my bump-steer completely by creating a cross-steer conversion, but only had the space for it because I had moved the engine back 18 inches....
My success has been verified on the streets and highway...
Other solutions may be available by applying some serious rocket science, but I'll wait for the test drive results before endorsing them....