Next I loosened every mounting point on the major offender's side (driver's side) and checked the play in every mounting point, put the suspension through it's motions and just generally thought about the problem. I noticed that even now the long bolts looked bent. I was pretty sure that wasn't the case, and it must have been an optical illusion by how cοck-eyed the parallel arm was compared to the bushing and the bolt.
I found what I believe is the source of the bent bolts. First off, when pushing the long bolt first through the front parallel link, then through the first hole in the cast iron knuckle the bolt heads straight through to the second hole in the knuckle but misses the center of the hole completely. I know this isn't scientific, but it seems that the holes in the knuckle once were aligned and now they are off some. This however isn't the biggest problem. The bolt can be coerced into the second hole without too much difficulty and then it passes through the rear parallel link. This is where the major problem lays I believe. The rear mounting hole in the knuckle is not flush. What I mean when I say this is that as the bolt enters the hole it crosses the first plane which is flush with (parallel to) the two planes of the first hole. As you exit the rear hole with the bolt the plane you just broke is not flush with (parallel to) the other three planes. It looks like this:
Both knuckles had this abnormality. From an engineering point of view this makes no sense at all and I'm attributing it either to casting imperfection and poor quality control on Nissan's part or there's a tiny possibility that when I took the cutting wheel to the car the first time to get everything apart I managed to cut into the cast iron at this less than ideal angle. Either way, when the nut on this long bolt is tightened down the metal center of the bushing ends up mating up with this stupid angled plane and if the rubber isn't flexible enough, enough force could be applied to the bolt itself to bend it. It doesn't help that the first time I assembled everything I torqued all the 22mm nuts/bolts to 80 ft. lbs. or more. This is way above the 55-65 ft. lbs. in the FSM. It wouldn't normally be a problem to torque that tight but in my case it made things worse as it pressed the metal sleeve in the bushing up against the plane of the knuckle with enough force to compress the gap completely:
This shows what happens when you tighten the nut down any considerable amount:
This is a bit exaggerated, but you get the point. The reason the parallel link doesn't completely match what the bent bolt does is because the bushing allows for some play.
ES bushings in this situation were a lot worse. Here are pictures of the real thing:
If this were the case the entire time with my car, the correct torque on the nuts and the flexibility of the rubber could have made all the difference. You will notice that this also forces the rear portion of the knuckle outboard some, which exacerbates the toe-in problem I have. After looking at the cast iron surface carefully and determining that I hadn't caused this situation with the cutting wheel (sorry no pictures, I was frustrated and filthy at this point) I decided to correct what I felt was an engineering error and/or oversight by Nissan. I took the cutting/grinding wheel to the cast iron knuckle and fixed that bastard (sorry no pictures, I was super frustrated and super filthy).
I know this was a ballsy move that not many would have attempted but I'll be damned if I couldn't make a better rear suspension than the sad fools who cut corners on our cars.
Cutting down the cast iron knuckle helped to orient the link with the bolt properly whose misalignment I believe accounts for the bending bolts, and removes the possibility that it will ever happen again, ES bushings or not. The parallel link was not perpendicular with the bolt yet even after all this, as at full droop the suspension still looks like this:
After doing this and getting every ounce of toe-out I could from the entire assembly and then tightening everything up to keep the toe-out I dropped the car again and gave the alignment a good look. Nothing had changed and I was no better off than before. It was late, my back was killing me, I was terribly frustrated, dehydrated, hungry, my head was spinning, I was dirty and sweaty, I had work early the next morning and this weekend wasn't the relaxation I needed from the previous week. I decided to give up and see what I might come up with after a night's sleep. Before cleaning up for the night I put the rear toe adjustment bolts at their most extreme toe-out setting (way past the adjustment marks on the washers) and never took any pictures because, well, I was not up for it. I'll maybe get pictures of the final alignment soon.
This was familiar territory, as this was the situation I drove off with the first time I put the ES bushings in. The toe adjustment bolts were at their extreme and things looked sort of okay. The passenger's side had almost neutral toe or maybe a tad bit of toe in, camber was a tad on the negative side and the tire mated up well with the fender being flush as it used to be. All in all the passenger side is livable.
The driver's side is suffering from a tiny bit too much toe-in, a tiny bit too much camber and the tire ends up being a tad bit inboard all together. You've just read a novel (this entire thread) on how to spend a bunch of money and countless hours just to end up worse off than when you started. I can think of a couple things I'd rather have done with that time/money, couldn't you?
The only consolation I feel is that I now know everything there is to know about the B13 rear suspension and there's this thread to shed more light on its mystical workings.
Before all of this started I had incredible rear tire wear characteristics. After the second installation of the parallel links (1 week after the first in hopes of fixing the problem in preparation of a road trip) and the 2,000 mile road trip I'd reduced my brand new rear Falken AZENIS RT-615 tires to treadless carcasses. The fronts on the other hand are wearing extremely well. With this final installation and adjustment I'm confident that my tire wear will improve quite a bit (should have left it this way before the road trip but I was stubborn) so all was not lost. I could have gotten this out of the ES bushings though, and something is still very wrong as the maximum adjustment is needed to even approach a good alignment. Before all of this started the toe adjustment bolts were almost at the middle of their adjustment.
It seems almost as if the rear-most mounting points on the chassis have shifted outward and/or the forward mounting points have shifted inward. Either that or the old parallel links grew and shrank respectively when I took them out and the new ones from Greg are just as fυcked. I just can't seem to get stock components to give me a stock alignment. The only non-stock pieces are the springs and struts, but I had a perfectly fine alignment after installing them and riding with them for a while so I've ruled them out. There's a small possibility I could sort of help the situation by finding the longest and shortest links and using that to my advantage but that wouldn't get me nearly back to where I was when things were perfect. I know this is a stretch (and I hope it's a stretch) but here's some pictures of the parallel link mounting points on my car:
The side that the metal seems to have separated from the frame rail is the side that's causing the most problems, but the other side has massive issues as well and doesn't look nearly as bad. I don't think this happened recently, I think the car came like this from the factory. Still, the idea of putting the car in a frame straightener of sorts has crossed my mind.
Last night I decided to put this project on the back burner as the alignment is halfway decent and the bolts shouldn't bend. After sleeping on it, I'm thinking the real/correct answer is to move the mounting points of the parallel links on the chassis around with some cutting and welding. This should get everything back to normal. Modified parallel links would get the job done too but is the less elegant solution. Either stock units could be cut and welded to be shorter/longer or once again something like the
Ingalls Engineering's adjustable parallel links could be used. The GA16 powered cars share a parallel link part number with the SE-R and the other link is a different part number altogether. Maybe the part from the non SE-R chassis is the answer...