1997 Honda Accord - Install Upper Control Arm, ball joint won't seat

I have a 1997 Honda Accord,SE, 2.2L Sedan with 137,000 miles.

I am in the process of replacing the upper control are on the passenger’s side.

I am having a problem with the installation. The ball joint stud is not pressing far enough into the steering knuckle. The hole for cotter pin is not yet exposed.

I tried tightening the castle nut to pull the stud through and that ended up stripping the nut. I am now using the original castle nut and don’t want to risk pulling the stud through by tightening. I tried banging the top of the control arm w/a hammer, but there is not enough clearance to get a good hit.

Is there a special tool that I need (ball joint press?) or are their any tricks to doing this?

Pull the stud out and compare the diameter and taper of the new to the old.

I will do that. Should the stud fit into place without too much difficulty and without needing a special tool?


The FSM says to put the tires on the car, lower the car, and let the car weight push the ball joint into the knuckle.

Thanks! That is exactly what I needed to know. I did rent a ball joint press this morning but it made no difference (and was very awkward to use based on the size of the press). This is my first ball joint experience so I am learning as I go.

The diameter of the taper on the new vs the old “looks” the same. I used an adjustable wrench to roughly measure it.

However, when I put the old one back on, it fit into place w/out too much difficulty (I did use the press). So, this tells me that there is a size difference with the tapers on the stud. I will return it and try another one.

Thank you for your responses!

I’ve gotten a couple of suspension components over the years that appeared to be the same on the surface but in actuality were a bit larger by some thousandths of an inch.
If the part number matches up then it’s due to lousy manufacturing and disregard for standards.

Although rare the problem of parts not being to spec is one that does occur. Some that have cropped up here were apparently miss-matched components, i.e., the wrong ball joint installed in the control arm while others were poor quality control in machining. Some discrepensies were so minor as to not show up until many miles of driving caused play in the tapered seat. It might be lucky for Mr Campbell that the problem was recognized and corrected without a catastrophic failure from installing a “seemingly” good part. I hope Mr Campbell gets back to us with the results of his efforts to properly repair his car.