Everything You Always Wanted About Lower Ball Joint Bolts

@SpikerEng Very nice and considerable effort, well done.

To all
Agree with Mr GREENBIRD

The surface area and diameter of the "bearing surface or washer face" has got to have a notable effect on the relationship of torque to tension along with resistance to loosening. The term"washer face" has nothing to due with actual washers. It's just bolt nomenclature for the under side of the bolt head.
This might seem obvious, but I did not notice much/any mention as far as a bolt recommendation. If I missed it, apologies.
Couple of pics comparing the differences.

Non flange


Flange


Worth a mention, die cut washers may have some degree of a crown further reducing the contact area (maybe). Top and bottom shown.


Almost last but not least, for the record.
These bolts came from the factory. Apparently very rare so no sweat with the no show in the tests. They were paired with the regular/longer/no shank flange bolts. They were used on at the inner two (thru) holes in non "protector" applications (earlier). They were on my 99 (bought new), several customers and several Toyota service replacement kits. I always replaced them with the common type.

Both take offs and new from the misc bolt bin.


Comparison from when I first noticed around 00.


Here is one used as a non cutting thread chaser/cleaner followed by a DIY mystery tool.



Carry on.

 

According to my ancient HP15C, the tensile stress area of these capscrews is .0963 sq inches. Division of that into the clamp load at any given point on your charted test values gives a value for the tensile stress in the fastener. If the Red and Green screws were targeted at 7500 - 7750 pounds of clamp, then the resulting stress in the little screws would be 77,900 - 80,500 psi. Using the numbers I grew up with (for a factor of safety), that would mean the design "ultimate strength" of the fastener material would be in the range of 115,000 to 120,000 psi. Due to the failure history showing that the clamp load isn't high enough to prevent failure - an old practice, when "dancing on the edge" (especially when thousands of units are involved) would be to shift the grip upward by 10% and try that.

The Black fasteners seem to be going that way - but the scatter you got is troubling.

 

I used these instead of any of the bolts from Toyota. Tensile strength of 150,000psi. Recommended by another poster here.


https://www.mcmaster.com/95735A635/

 

what torque value did you use with these bolts? 59 ft lbs

 

If you use the same torque with 10.9 screws, you may not get the same friction result as the Red/Green coated Toyota capscrews. If you want to try to get slightly more grip than the 7500 pounds - up 10% is usually enough to get a positive, noticeable fatigue result - so shooting for 8250 is reasonable (and similar to the result shown of using the Black Toyota screws). If you look back at the Fastenal torque chart shown earlier, it says that an M10 x 1.50 has 8115 pounds of grip and shows torques for various friction conditions. The stress area of a M10 x 1.25 is going to be slightly greater - so the grip will be slightly higher and so will the recommended torque. I haven't done that math.

I hate to ask - but is there a way to simply "re-fit" these with M12 x 1.75 screws? Trying to "science out" a problem with material strength changes, fancy coatings and so on is usually not as good as "brute strength and ignorance".

 

I used 59ftlbs. I was working on the truck today and all of my paint marks have not moved. I think I have about 15k+ miles on them now. Did some rock crawling and drove on some pretty punishing washboard roads for 30-45 minutes straight.

 

I promise no more math in this post - just the winners and losers!

Taking the data collected in the previous tests, I ranked the bolts to find the winners and the losers. I first defined what I consider to be the important criteria for LBJ bolts, then assigned different weighting factors to each criteria. For example, criteria that improved confidence in knowing the actual preload generated at FSM torque was weighted more heavily than secondary criteria such as preload degradation with reuse. I then scored the bolts against these criteria using their measured performance, and added up the scores to get the total combined score for each bolt.

Using this admittedly subjective criteria, the combined scores for all bolts are in the figure below. The winners are the Black Bolt (90119-10933) and the ARP Bolt (673-1004). The Red and Green bolts were also good, but not as good the Black and ARP bolts. The 10.9 Zinc bolt was far back in last place. (Additional details on the scoring is at the end of the post).



Based on these results, my top recommendation for LBJ bolts would be the Toyota Black Bolt w/Washer (90119-10933) or the ARP Bolt (673-1004). The Black bolt excels in its incredibly low friction coefficient, which leads to a silk-smooth installation, and allows the bolt to reach 10,000 lb of preload at only 37 ft-lbs, significantly higher than many other bolts could achieve at 59 ft-lbs. It also provides the highest confidence in the preload generated at the applied torque, and does not see much degradation with reuse. The ARP bolt is the strongest of the group, giving it very high margin between installed torque and yield torque. It also does not degrade much with reuse.

The Red and Green bolts are certainly adequate for LBJ use, although they don’t have as much margin between installed torque and yield torque as some others. They also have the highest degradation with reuse, so I would not recommend reusing these bolts more than once. By the third or fourth installation, you would likely be seeing only half the preload at the FSM torque as you would during the first installation.

The 10.9 Zinc bolt fared far worse than all others, primarily due to its very high friction variability from unit to unit and batch to batch. Without testing each batch like I did, you would have no idea how much preload is being generated at the FSM torque. For this reason, I would recommend avoiding any generic bolts for the LBJs.

There is one caveat about the Black bolt that must be mentioned. This bolt was designed to be torqued to 37 ft-lbs, due to its very low friction coefficient. The downside of this is that if these bolts are torqued to the more commonly quoted FSM torque of 59 ft-lbs, they will likely yield. I’ve seen many forum posts where folks have torqued the Black bolts to 59 ft-lbs and had them break, then concluded (wrongly) that the Black bolts must be weaker than others. So the caveat is that if these Black bolts are installed in trucks that originally came with Red or Green bolts, anyone working on these trucks in the future must be told to torque the bolts to the lower torque values. This is particularly important if you let a mechanic work on your truck, who may not be aware that you installed non-factory bolts with lower torque requirements.

A Remaining Concern for LBJ Bolt Failures.

An important element that these tests did not address is the possibility of LBJ bolts losing preload and backing out, especially during offroad use. I believe that this is the most common failure cause for LBJ bolts, because properly preloaded bolts will rarely fail or shear off (the friction generated by four preloaded LBJ bolts is quite high, and there are two large shear cones on the LBJ flange that engage the steering knuckle to resist the shear loads).

However, if the bolts lose preload, they are more likely to loosen and eventually fall out. I’ve seen a number of posts with folks finding one or more missing bolts in the LBJs, suggesting that torque alone may not adequate to keep these bolts safely in place during offroad use. Even if the bolts don’t fall out, loss of preload will subject the joint to gapping and slipping, and the bolts will become more likely to fail from bending loads.

A common practice to address this risk is to use Loctite or a similar threadlocking compound. While these are effective if used properly, there are two concerns – first, there is no way to know if the Loctite was applied properly and has formed an effective bond. In addition, it is not possible to look at an installed bolt and know if Loctite was applied to it during installation. The biggest issue with Loctite is that it affects the preload that is generated during installation. While it does not change the preload very much during the first installation, my testing found that on reuse, even if the bolt threads were carefully cleaned with a wire wheel, the preload ended up being much lower than on initial installation. The most likely explanation for this is that cured Loctite remained in the female threads of the knuckle, increasing the friction coefficient. So unless the threads are chased with a tap after each use, there is likely to be much less preload in the joint than intended or desired.

Even the “best” bolts will be of no value if they loosen in use. So I believe that properly “locking” the LBJ bolts to keep them from loosening is as critical as selecting the “best” bolts, if not more so. I'm currently looking at some options for the highest strength bolts with bulletproof locking features - I'll certainly update you if and when that comes to fruition.

Additional detail on scoring criteria:

1. Preload Uncertainty – This criteria ranks the uncertainty in preload generated by a bolt when it is torqued to FSM torque (or any other value). The narrower the range, the less uncertainty in the preload, and the higher the score.



2. Bolt Strength – It stands to reason that all else being equal, the stronger the bolt, the better. Stronger bolts can be preloaded to higher clamping loads, providing higher margins against slipping and gapping.



3. Preload at FSM Torque – Some bolts reached much higher preloads than others at FSM torque values. Higher preload is better, as it reduces the chance of gapping or slipping.



4. Margin between FSM torque and Yield Torque – This criteria evaluates how much safety margin exists between the FSM torque and the torque at which the bolts fail (yield). The higher the margin, the more room for error exists.



5. Preload Retention with Reuse – Some bolts saw a significant reduction in preload between first use and subsequent installations; others did not. Since the LBJ bolts are commonly removed to separate the knuckle from the lower control arm, the ability to reuse the bolts with high confidence would favor those bolts that perform consistently during repeated installations.



6. Friction Coefficient – This criteria has little practical value other than “feel”. The bolts with low friction coefficient felt very smooth during installation, and reached the FSM torque values easily. The ones with higher friction, notably the Zinc bolts, were jerky and had significant stick-slip during installation, making the torquing process more difficult.

 

Thanks Mike! I hate threadlockers, they work but have many issues, as I tried to touch on in my last post. I would really like to come up with something more robust and bulletproof for locking than "glue".

 

Great work and very informative, thanks for sharing all of this Leon.

------------------------------------------------------------

EDIT 04/07/23 - I'm placing my post from another LBJ thread here - for now. Some of the post content was deleted when the thread was nuked by it's creator. Rebuilt from cached copies.

If you have the dust protectors, then the correct bolt is the black w/washer -
90119-10933. They are torqued to 37 ft lbs ONLY. That bolt specifically. 37 ft lbs ONLY.

If you do not have the dust protector on the LBJ, then I would order the ARP bolts and torque em to 59 ft lbs. If you want to stick with OEM then the RED (90105-10505) over the Green would be my choice.

The dust protector was used on 2001-2004 Tacomas, 2001-2002 4Runners, 2001-2004 Sequoias and 2001-2004 Tundras.

**NOTE** Some have opted to use the black w/washer bolts (90119-10933) WITHOUT the dust protector. I have no experience with this and have not done it. If you do, do so at your own risk. Be mindful that the bolt seats fully when snugging up by hand and once snug, torque to 37 ft lbs.

Part numbers for reference:

4333039556 (Passenger Lower Ball Joint)
4334039436 (Driver Lower Ball Joint)

9017116050 (Castle Nut for LBJ) - can be reused but I just replace it.
9538103230 (Cotter Pin) - ALWAYS replace with a new one, never reuse.
*A note: If ordering the LBJ, I can't recall if it comes with a cotter pin and new castle nut*

4334660011 (Dust Protector) - replace if ripped/torn otherwise reuse them.

90119-10933 (Black bolts w/washer for use with Dust Protector ONLY - 37ft lbs torque). Always replace, don't reuse them - not worth the risk for such an inexpensive bolt.

EDIT: There is a part# for the LBJ Service Recall Kit (included 2 LBJs, bolts and cotter pins) that toyota was using during the recall campaign. I have no idea what bolts it comes with, which part# it is or if it's still available to order.

-------------------------------------------

If you're interested in testing your existing lower ball joints, take a look here:

https://youtu.be/sp-D0EtX2gU?feature=shared

 

I've pondered using studs, but I think there are some downsides. For one, if you just want to separate the knuckle from the LCA, to swing it out to replace the axle, you will have to separate them more to clear the studs. Not sure if there are any real downsides, I frankly haven't investigated it much. But yes, we mechanical engineers always prefer studs over screws.

 

I used all fasteners "dry", as delivered. Adding lubes would have opened a new can worms that I didn't want to deal with. The Black bolts actually had way more preload at 37 ft-lbs than Red and Green did at 59 ft-lbs. It is just a fantastic bolt, IMO.

As for washers, both the Black and the ARP use them (the others have flanged heads). It was interesting to see how much galling and wear happened under the flanged heads, compared to almost no wear under the hardened washers.

I agree that the Zinc plating appears to have behaved very poorly in these tests. I would not use these bolts without lubricant, but then someone would have to have to redo all these tests to see the effect of the lube - I'm not doing it .

 

I agree, Black bolts at 59 ft-lbs are very likely yielded - good idea to replace them!

 

Thanks for the additional info, I've never seen those bolts you had form the factory. I suppose there is no way to find the PN of them?

I completely agree with you about the importance of washer (and the effective bearing area/diameter). Out of curiosity, I looked at the wear patterns on the bolts flanges and washers after repeated use, and measured those "effective" diameters - I then used those numbers in my calculations to try to estimate the friction coefficients under the bolt heads and to estimate how much shear stress the bolts saw in addition to tensile stress.

 

That's the problem though - these bolts have a high (and known) strength, but without testing, you will have no idea how much preload you are getting at 59 ft-lbs (or any other value). I ordered the 10.9 bolts from the same source, and the preload at 59 ft-lbs varied by a factor of 2.

 

No, I never saw a pt#. Best guess is they were some kind of temporary substitution in a supply issue.

 

That would make sense, they look nothing like anything else I've seen in that location.

 

In a word no. Speaking for the inner 2 of the 4 bolts, there just isn't enough meat on the spindle. Heck, with M10 it's a little sketchy.
Little hard to see but the proximity of the bore to the edge of the spindle (wall thickness) is minimal (right arrow). Plus it tapers to zero, maybe 5/6 threads total.

 

I think most LBJ bolt failures occur not so much due to weak bolts, but to under-preloaded bolts which back out with time and allow gapping and slipping. So I think finding a robust locking feature (with a properly preloaded bolt) is more beneficial than looking for the strongest bolt.

 

Best locking feature I can think of for now is the good ol' paint pen and routine visual checking method.

 

I'm working on something that I think would be better, still need to check a few things out before being sure.