If you mean this one, it’s old school Savvy:
I don’t think they make them anymore, but someone else may chime in who knows better than I.
Sab-a-dab-a-doo! The back-country LJ build has officially started
- Thread starter sab
- Start date
I don’t think they make them anymore, but someone else may chime in who knows better than I.
It's been about five long weeks since I worked on the LJ. 
After taking two weeks vacation, I had to go back to work two weeks ago. Like is usually the case when one takes that much time off, the last two weeks have been very busy catching up on everything. Today, I finally found a few hours to dedicate to the build. The last mechanical work I have to do for the Flux Offroad tailgate valance armor and tub rub rails installation is to install nutserts in the tub for them. I elected to do this rather than tap the corner armor. Overkill? Perhaps. But no-weight-added overkill only costs time, so I'm down.
I started by removing the corner armor, with Mrs. sab's help because I spent a fortune having these professionally body-color-matched (that's the epoxy primer on the inside you're seeing):
I had previously drilled the holes for the Flux parts to 3/8" diameter, but for the nutserts, I had to drill them out to 17/32". I did that, and in doing so, I noticed that the outside lower hole for the valance armor on both sides was through multiple layers:
That particular hole presents two issues. First, is the the total thickness. It's right at the limit (.150") of the standard nutsert's grip range. Second, it's very close to the bottom of the panel, so as the nutsert is deformed, it could hit the inside bottom of the lower wall. If that happens, the tapped hole and the O.D. won't be concentric, and because the valance armor is countersunk, I'll have alignment problems. Good thing I keep longer nutserts in my parts bins! The next size covers from .150" to .312" grip range. This picture shows the two options side-by-side:
So, I used the longer nutsert on it:
Oops! One of the issues with using a nutsert at the minimum of it's grip range is that it can easily offset eccentrically because the thinner section is longer. I could feel it offset wrong while installing it. The closeness to the bottom of the cavity pushed it up as I squeezed the levers on the installation tool. I outsmarted myself!
So, I drilled it out and tried the shorter nutsert:
Much better! The mirrored hole on the other side went well, too, with the shorter nutsert. I installed all the other nutserts, and then Mrs. sab helped me place the corner armor back on. I took an acid brush and painted some anti-seize in the countersink cones to (hopefully) make future removal easier before installing the screws.
I ended the day having all the screws in place finger-tight, but I didn't have enough time to line the corner armor up and tighten them.
Stay tuned!
After taking two weeks vacation, I had to go back to work two weeks ago. Like is usually the case when one takes that much time off, the last two weeks have been very busy catching up on everything. Today, I finally found a few hours to dedicate to the build. The last mechanical work I have to do for the Flux Offroad tailgate valance armor and tub rub rails installation is to install nutserts in the tub for them. I elected to do this rather than tap the corner armor. Overkill? Perhaps. But no-weight-added overkill only costs time, so I'm down.I started by removing the corner armor, with Mrs. sab's help because I spent a fortune having these professionally body-color-matched (that's the epoxy primer on the inside you're seeing):
I had previously drilled the holes for the Flux parts to 3/8" diameter, but for the nutserts, I had to drill them out to 17/32". I did that, and in doing so, I noticed that the outside lower hole for the valance armor on both sides was through multiple layers:
That particular hole presents two issues. First, is the the total thickness. It's right at the limit (.150") of the standard nutsert's grip range. Second, it's very close to the bottom of the panel, so as the nutsert is deformed, it could hit the inside bottom of the lower wall. If that happens, the tapped hole and the O.D. won't be concentric, and because the valance armor is countersunk, I'll have alignment problems. Good thing I keep longer nutserts in my parts bins! The next size covers from .150" to .312" grip range. This picture shows the two options side-by-side:
So, I used the longer nutsert on it:
Oops! One of the issues with using a nutsert at the minimum of it's grip range is that it can easily offset eccentrically because the thinner section is longer. I could feel it offset wrong while installing it. The closeness to the bottom of the cavity pushed it up as I squeezed the levers on the installation tool. I outsmarted myself!
So, I drilled it out and tried the shorter nutsert:Much better! The mirrored hole on the other side went well, too, with the shorter nutsert. I installed all the other nutserts, and then Mrs. sab helped me place the corner armor back on. I took an acid brush and painted some anti-seize in the countersink cones to (hopefully) make future removal easier before installing the screws.
I ended the day having all the screws in place finger-tight, but I didn't have enough time to line the corner armor up and tighten them.
Stay tuned!
Last edited:
It's been about five long weeks since I worked on the LJ.
I started by removing the corner armor, with Mrs. sab's help because I spent a fortune having these professionally body-color-matched (that's the epoxy primer on the inside you're seeing):
View attachment 704847
I had previously drilled the holes for the Flux parts to 3/8" diameter, but for the nutserts, I had to drill them out to 17/32". I did that, and in doing so, I noticed that the outside lower hole for the valance armor on both sides was through multiple layers:
View attachment 704850
That particular hole presents two issues. First, is the the total thickness. It's right at the limit (.150") of the standard nutsert's grip range. Second, it's very close to the bottom of the panel, so as the nutsert is deformed, it could hit the inside bottom of the lower wall. If that happens, the tapped hole and the O.D. won't be concentric, and because the valance armor is countersunk, I'll have alignment problems. Good thing I keep longer nutserts in my parts bins! The next size covers from .150" to .312" grip range. This picture shows the two options side-by-side:
View attachment 704852
So, I used the longer nutsert on it:
View attachment 704854
Oops! One of the issues with using a nutsert at the minimum of it's grip range is that it can easily offset eccentrically because the thinner section is longer. I could feel it offset wrong while installing it. The closeness to the bottom of the cavity pushed it up as I squeezed the levers on the installation tool. I outsmarted myself!
View attachment 704855
Much better! The mirrored hole on the other side went well, too, with the shorter nutsert. I installed all the other nutserts, and then Mrs. sab helped me place the corner armor back on. I took an acid brush and painted some anti-seize in the countersink cones to (hopefully) make future removal easier before installing the screws.
I ended the day having all the screws in place finger-tight, but I didn't have enough time to line the corner armor up and tighten them.
Stay tuned!
Making progress!
Appreciate the explanation because when I saw the first pic of the nutsert, I was asking myself: "Where is this guy getting offset nutserts!?!?! And why??"
They say that a picture is worth a thousand words, but what they don't tell you is that sometimes a picture is worthless without words...
I've been thinking about this since I posted the above and rethinking your post. I also have the same Fluxor corners and valance to install.
And your post above has me asking the question of why? You made the same similar consideration of why in your original post. I guess I'm just curious if you would change your position or further with the nutserts.
I've always had a lot of respect from your engineering perspective, so that's why I'm double-checking you before I jump in the same pool.
And your post above has me asking the question of why? You made the same similar consideration of why in your original post. I guess I'm just curious if you would change your position or further with the nutserts.
I've always had a lot of respect from your engineering perspective, so that's why I'm double-checking you before I jump in the same pool.
I was under the same impression when I needed a set and was pleasantly surprised to see them available.
I think this is it @Hackyman
https://savvyoffroad.com/products/bmp-lm-light-mount-for-tj-lj-jk-front-bumper
Last edited:
I’m not sure what you’re asking here, but I think you’re wondering why I didn’t just tap the corner armor instead of using nutserts. I did it because it’s a stronger mounting scheme and didn’t cost me anything but time and a few nutserts.
I can’t predict how an object will contact the tub rub rails or the valance, and at least in my mind, I can foresee the possibility that something pushes on a leading edge, which would result in a shear force on the bolted joint. Using the nutserts will resist that better.
Is that likely? Probably not, but if it does happen, I’m covered. This mentality comes from my racing days. I’ve had the “shoot, I didn’t think that’d ever happen” blues too many times.
To add...A typical 5/16-18 bolt has 18 threads per inch (that what the -18 means in an SAE thread call out), so every ~0.055 of an inch, there is a thread peak. In 3/16" material (.1875) you get 3.3 threads. While that isn't Terrible, you really would like 1.5 x diameter for good, strong engagement (about a half inch). I this instance, I think the nutserts are a good move.
For me, the decision was more based on material strength than thread engagement length. While the 1.5x rule is a good, general rule of thumb, I've successfully violated it in my designs many times over the years. I think it's pretty conservative. Also, in this case, I don't think the nutsert gives any more engagement length than tapping the armor. In fact, it might be less. Here's the McMaster-Carr drawing:
I believe the threaded portion is that below the ribs, so 0.690" - 0.470", or 0.22".
Edited to add: I pulled the STEP file from McMaster-Carr's website and analyzed it in Fusion. It's actually even less thread engagement than I thought - it's 0.16":
Last edited:
I can't answer the material properties question, but I think I can take a stab as to why the nutsert is still a better option despite potentially giving less thread engagement. As he mentioned here...
if something were to hit the edge just right it would create a shear force across the bolt. Because there are multiple thin layers (two it looks like) it would create a very concentrated bearing stress on the threads. Stress loves targeting the smallest points of an object it's working on, which in the case of threads is the root diameter, so the combination of a small diameter and concentrated bearing stress is like a shaped charge on the bolt. By adding the nutsert it turns those two pieces of metal into essentially one solid piece. The shear force now spreads along the whole body of the nutsert which then acts on all the threads that are engaged with it rather than only one or two and dramatically reduces the bearing stress experienced by the bolt.
Feel free to yell at me if I'm wrong Sab, its the only way I learn
.
Those are all good questions, Mike. Steel nutserts are carbon steel (not alloy steel), but it's very hard to find out specifics or material properties. Rather than hardness, which is directly proportional to tensile strength, I would think more in terms of tensile strength and shear strength when comparing materials for this situation, but those are directly proportional to each other, too. For most metals, shear strength is about 2/3 the tensile strength, with a fair amount of variation, even lot to lot. So, I was only considering tensile strength.
The tensile strength of 6061-T6 is typically in the range of 40,000-45,000 psi. I know the steel nutserts are considerably stronger than 6061-T6 just from working with them and with 6061-T6 threaded holes over the years. I'd guess them to be on the order of twice the strength of 6061-T6, but that is only an educated guess, not based on actual data.
As to pull-out strength, there is data available for nutserts, but not so much for 6061-T6 threaded holes, so I'm unable to make a direct comparison. However, I think it's safe to use the tensile strengths to compare, even using my guesstimate for the nutsert material's tensile strength, because all other things being equal, pull-out strength would be directly related to tensile strength.
Finally, there is one other factor I considered in my decision, and that is serviceability. Aluminum's potential for galling exceeds that of steel's considerably, especially under repeated fastening/unfastening cycles. Foreseeing future needs to remove armor for various tasks (I've already removed and replaced my step sliders about five times), the steel nutserts will work better in that capacity.
My decision matrix was much simpler than that - it was based mostly on the relative strengths of the two materials, with an educated guess for one of them. I can't predict the loading that will cause failure (shear or tensile), but with the joint being a steel bolt in an aluminum hole, I know the failure mode will be bolt pull-out with failure of the aluminum threads.if something were to hit the edge just right it would create a shear force across the bolt. Because there are multiple thin layers (two it looks like) it would create a very concentrated bearing stress on the threads. Stress loves targeting the smallest points of an object it's working on, which in the case of threads is the root diameter, so the combination of a small diameter and concentrated bearing stress is like a shaped charge on the bolt. By adding the nutsert it turns those two pieces of metal into essentially one solid piece. The shear force now spreads along the whole body of the nutsert which then acts on all the threads that are engaged with it rather than only one or two and dramatically reduces the bearing stress experienced by the bolt.
Feel free to yell at me if I'm wrong Sab, its the only way I learn
However, with the joint being a steel bolt in a steel nutsert, I can only guess how the bolted joint will fail. It could fail by pulling the threads off the screw, pulling the threads out of the nutsert, or pulling the nutsert out of the body and through the aluminum. I think it would be one of the first two, rather than the last. So, in all failure modes that I believe are likely, the biggest factor is strength of the threads, and I think the steel screw and the steel nutsert are both stronger than tapped aluminum armor.
Cool to learn about. I got so caught up in "shear" and thinking about the pin shear problems I've had to do that I forgot about the very real world (and more common) scenario of just plain stripping the threads in mismatched materials.
Last edited:
Hopefully you had a course in college that covered how failure prediction works in the design world. If not, then the simple explanation is that you have to analyze every potential mode of failure, and then you compare them, and the one that happens at the lowest loading is the likely mode of failure. Additionally, the failure mode can change while the part is failing. For instance, as a column loaded axially starts buckling, that can change the axial force to a bending force very quickly as the load moves off-center, so what started as a buckling failure ultimately resulted in a bending failure. For the case at hand, I was using experience to predict the modes of failure rather than actually analyzing the loads.
Steel bolts have excellent shear strength relative to their size. If you keep reducing the size of a bolt for a joint subject to shear in your head, before you get to a size small enough to fail, you'll look at it and say to yourself, "that bolt is too small." For this reason, we don't see a lot of shear failures. In addition, in my head, the load I was considering that resulted in a shear force on the bolt wouldn't actually result in a shear failure because the bolt is so strong in shear. I didn't explain that thought process very well.
When that load produces shear on the bolt, the bolt will resist, and the failure mode will change as things move. The result is that the load pushes on the bolt in shear, it resists the shear without breaking, things move, and now we have a bending and axial force on that bolt. The weak link is the thread interface, so the axial force causes the failure as the threads fail and the bolt pulls out of the hole.
Edited first paragraph for clarity (I hope).
Last edited:
We definitely learned about design and failure analysis, but it often felt simplified or "ideal." One big thing about my major is that dynamics was not part of our course plan at all, so every single class we took was based around static objects/components. This often meant that in design problems where they would ask you if something would fail and why, it would fail in one very obvious way rather than multiple borderline problem areas and having to choose the weakest link. I also cant recall anytime where we looked at how failure modes would change as things moved outside of buckling and bending. The closest we got to dynamic loads was cyclical loading, and even that became distilled into making it work like a static load and checking safety factor based on that (Goodman and Soderberg). The way I loved describing how our classes worked to underclassmen was "everything is static, nothing moves, but you have to pretend it's moving and it all happens instantly."
Fasteners were touched on in a couple of my classes, but I never felt like they were given the proper attention. Sure we learned about a few different types of fasteners, and we learned how to calculate threaded connections, but they were often used as a more minor part of a problem or not even brought up again at all. Really disappointing considering that it seems like such an interesting topic, but maybe that's just the case of me wanting to know more about a subject than the department deemed necessary.
Unfortunately, that's the difference between academia and the real world. We learn those simplified concepts in college so that we understand the basics, but in the real world, they don't work very well. You just use that knowledge to guide your design methodology, and then you use testing to validate and improve design. Even the fanciest dynamical finite element analysis needs validation in the real world because it's still "simple" compared to real-world. That's easier done in the production-vehicle world. In the racing world, there's never enough testing time, and testing on race weekend is ill advised. Intuition and experience in the race engineering world separates the ho-hums from the superstars. Ho-hums have short careers.
Missing dynamics in your course of study puts you at a disadvantage in the vehicle world, for sure, but in this case, the dynamics classes don't really address the problem. It's more of a case of understanding failure modes and how they can change while the failure is occurring, as I pointed out in my last post. Understanding those two things will help you find root cause for many issues, and finding root cause is required to really solve any problem.
I've never met a mechanical engineer who learned about fasteners in college. I was the bolted joint SME (Subject Matter Expert) at Polaris because I'd spent 15 years studying failures of bolted joints on the vehicles I helped design, along with attending week-long seminars on the subject to help understand what I was seeing. I would say that my skills in that arena have definitely atrophied in the two decades since leaving the profession of mechanical engineering, though. That's where the Internet helps. I still know enough to be able to research and remind myself of the concepts when I encounter an issue.
Difficult, but important to understand because we use this to fill knowledge gaps.
Intuition, I have come to think, is the result of pattern recognition which requires experience. Pattern recognition is something the human brain excels at.
The trouble comes when we get too invested in our theories: The thought that “I want it to be true, so it must be” also seems to be something the human brain gravitates to. So its good to go back to science (well designed experiments with good data) whenever possible.
Sorry, just working out my thoughts on @sab ’s “paper”…
Carry on. Good conversation.
Last edited:
100%
My belief is that intuition is wired into your brain, and you can improve it slightly through experience, but you can't really take a person with low intuitive skills and give them experience to jump up to average or excellent intuition. The Myers-Briggs personality test, first taken during my Introductory Psychology course in college, is what first exposed me to this idea. An example of someone most forum members know with out-of-this world intuitive skills is our own Mr. Blaine. He can quickly look at a problem and intuit a cause and solution. High intuition skills, coupled with experience, can quickly build a knowledgebase that is unrivaled. I think that is sometimes mistaken for improving intuition over time.
Lastly, experience can only be gained through life experience, and this is a lesson for the younger generation. Us old-timer, Boomers and Gen-Xers, were told that work ethic is the most important thing in life. In my racing days, I probably averaged 80 hours per week during the season (we worked seven days a week, and the weekends were races, so Saturday and Sunday were usually 7am-7pm). During the short off-season, I probably averaged 60 hours per week. That means that I built up experience at a rate 1.5 to 2 times that of the 40-hour-a-weekers. So after ten years of that, I was five to ten years ahead of them. I'm grateful for having been through that, but it does take it's toll over time. That's the main reason I left that profession after about 15 years. However, I walked away much better suited for the things that came as my career progressed.
I often compare engineering to playing a board or card game. College is like reading the rule book. You learn all the rules to the game that you're meant to follow, when certain actions should be taken and how to carry them out, and how to start and end the game. When you actually start playing though (aka get a job), you learn that everyone has their own house rules that trump the rule book, some people have shortcuts to make the game faster, and there is always someone who will know the game better than you. I wholeheartedly agree that experience and intuition go miles just gotta keep on my toes and not get complacent not knowing what I don't know. It's unfortunate how easy it is to pick out students, even in my major, who fit into the "ho-hum" category, but I hope they're at least able to connect what they've learned to their jobs now.
Interesting, guess I was taught more than I thought in that case. Learning about fasteners in my degree makes more sense since it largely tried to combine the theoretical with the practical in an even mix. Since manufacturing involves a lot of fasteners, I'm sure that they were compelled to make sure that the average student who graduates from their program could at least read thread designations and know that a torque spec is not some randomly suggested number.
Similar threads
