Prepare yourself for a lot of different opinions on this one.

Regarding minimum radius, 16" is below minimum standards for HO scale. 18" is widely considered to be the bare minimum. 22" to 24" radius works well for short equipment. 26" to 28" radius is better. 30" to 32" is best for a minimum radius. If you want maximum flexibility and the space to pull it off, 34" to 40" radius would be highly desirable. The longer the train car the bigger the turn radius. I purposefully do not buy or run trains greater than 40' on my layout for that reason and my radius's are 22" and 24" respectively.
The radius is generally recommended to be 3x the length of the rolling stock. Let's take a 85' passenger car. 85/87.1 (scale) = .97 or 11.72". Take that and multiply it by ~3.
That's for smooth reliable operation that looks appropriate. And most will tell you 32" is about right for passenger service. But Bachmann will make passenger cars that run down to 18" thanks to their innovate coupler design. But the curve between cars, and track overhang isn't prototypical looking. There's large gaps.
Also when running long cars, you have to increase distance around parrallel curved tracks. Most parallel tracks are 2" apart (HO scale). Around curves you'll have to increase this to almost 2.5" for passenger trains around tight curves. This is to keep "clothes lining" rolling stock from bumping into each other around curved tracks.
If you're going to run long cars, make your radius's 35" or bigger and all your problems with turns and long cars will dissappear!
You could sell all your long cars and run short cars like I do.

 

I’ve run full size hi-cubes, multiples in tandem, on 22” radius & on #5 TOs. Pushing required a speed limit so the flanges didn’t jump the rail, but truck and coupler swing were not an issue.

The 26” radius should be fine. The #4 TO is one suspect to me. Since you’ve already verified the problems persist on #6 TOs and 30” curves, I believe you are correct that the issue is truck/coupler swing… or more specifically, something inhibiting proper operation.
How far do the trucks freely turn by hand? What diameter wheels are installed?

I can’t imagine anyone knowingly making cars that cannot negotiate a 30” radius or #6 TO. Hence I suspect human error during assembly, incorrect wheel dia or something like that. If the set came with an “exploded view” of the parts, I’d closely compare that to the actual cars. Just my deductive reasoning. Like maybe weights are in backwards or something silly.

 

I’ve run full size hi-cubes, multiples in tandem, on 22” radius & on #5 TOs. Pushing required a speed limit so the flanges didn’t jump the rail, but truck and coupler swing were not an issue.

The 26” radius should be fine. The #4 TO is one suspect to me. Since you’ve already verified the problems persist on #6 TOs and 30” curves, I believe you are correct that the issue is truck/coupler swing… or more specifically, something inhibiting proper operation.
How far do the trucks freely turn by hand? What diameter wheels are installed?

I can’t imagine anyone knowingly making cars that cannot negotiate a 30” radius or #6 TO. Hence I suspect human error during assembly, incorrect wheel dia or something like that. If the set came with an “exploded view” of the parts, I’d closely compare that to the actual cars. Just my deductive reasoning. Like maybe weights are in backwards or something silly.

I’ve run full size hi-cubes, multiples in tandem, on 22” radius & on #5 TOs. Pushing required a speed limit so the flanges didn’t jump the rail, but truck and coupler swing were not an issue.

The 26” radius should be fine. The #4 TO is one suspect to me. Since you’ve already verified the problems persist on #6 TOs and 30” curves, I believe you are correct that the issue is truck/coupler swing… or more specifically, something inhibiting proper operation.
How far do the trucks freely turn by hand? What diameter wheels are installed?

I can’t imagine anyone knowingly making cars that cannot negotiate a 30” radius or #6 TO. Hence I suspect human error during assembly, incorrect wheel dia or something like that. If the set came with an “exploded view” of the parts, I’d closely compare that to the actual cars. Just my deductive reasoning. Like maybe weights are in backwards or something silly.

Thank you for your comments we think the same. Upon further examination,it seems the trucks only swing so far and then they rub against the coupler box. The backs of the trucks also rub whenturned at an angle on the side of the 1st well. I filed this down,but reasoned the real hang up is the trucks hitting the coupler boxes. On a complete 30" radius with no turnouts cut in,the cars do negotiate these "fairly" decently,but again sometimes the trucks rub the coupler box if everything doesn't line up just right. It looks like a mess up on Walthers part as my 89ft autoracks and 72ft centerbeams negotiate the curves fine even in #4 t/os. I'm only using the #4s as I have so many from my past layout and don't have the funds to replace them at present,at my age and limited income,maybe replacing will never be an option. I checked out the various locations the #6 t/os will need to go and they will be a nice fit. Worked hours last night on the problems and am going to try to replace the 2 turnouts and use a 30" curve. At present this looks like it's going to be complicated to get tracks back into alignment but I will see,and let you folks know. The problem shouldn't exist on that generous of radius as you stated. Thanks again to all. Al

 

That’s what I expected to hear about the coupler boxes.
If they are hitting by more than 1 mm or so, that’s likely a design flaw. But if it’s just barely touching, check the wheel diameter. They may have put 33 inch wheels on it. 28 inch wheels may clear the coupler boxes. 28s about 1-2 mm smaller than 33s so how much they hit should suggest if that will solve the issue or not.

 

That’s what I expected to hear about the coupler boxes.
If they are hitting by more than 1 mm or so, that’s likely a design flaw. But if it’s just barely touching, check the wheel diameter. They may have put 33 inch wheels on it. 28 inch wheels may clear the coupler boxes. 28s about 1-2 mm smaller than 33s so how much they hit should suggest if that will solve the issue or not.

Will give that a try. Also,these cars were bought in sets of 3 cars. Each car has separate draft gear boxes on both ends of each car. I figured the center car would be connected with a drawbar to the 2 end units but no drawbar is included. This still shouldn't have anything to do with the wheels though I wouldn't think. Thank you,Al

 

As a general rule, the closer one gets to engineered, or even to 'real', minimums, the higher the probability of problems. Introduce other variables, such as slight variations in curvature along curves, which the typical and careful modeler will be almost certain to generate, variations in assembly of cars, and even variations in rolling resistance from car-to-car due to axle conditions and flange spacing in the axles, you run a substantial risk of a derailment on curves that are close to the claimed minimum radius for the equipment being used.

Yeah, rules are meant to be tested, even frequently, but it doesn't follow that they won't impede your progress due to the variances I mentioned earlier. The fact is that the hobby didn't have to contend with many cars longer than about 60' until, and unless, you wanted to run 80' passenger cars. If you wanted body-mounted couplers and diaphragms, you had to broaden your curves, even in 1940. Now, with autoracks, well cars, and many of us still run heavyweights in the 86' range, what used to pass for 'minimum' curves 30 years ago just doesn't work any more. I run only transition rolling stock, nothing more modern than 1957. I won't have a curve or a turnout affording me proven radii of 33" or less. This is both for looks and for reliability.