Ribbit Rivet...

Billfred & I have traded messages on the use of 1/4" rivets to assemble the kitbot frame instead of the 1/4-20 hardware. Estimated savings is a pound or three. This post to to share what we came up with, and solicit further input.

First, a riveter: Harbor Freight sells an inexpensive one here.

Then, rivets. McMaster has them (search item 97519A652 for example) but Grainger seems to be slightly more cost-effective. (Note that you’ll need a variety of grip ranges)

Lastly, we thought that steel rivets with standard shear strength (1240 lbs) should be sufficient, but no testing has been done. While this is far less than the shear strength of 1/4-20 hardware, that same hardware appears to be overkill.

Upside is weight savings, downside is difficulty in dismantling the frame (never had a need to do that) and repairs if a fastener loosens. Perhaps an adhesive to reinforce the joints to further absorb shear.

Sources, ideas, thoughts, experience from teams that did it?

Don

Good news! that may not be a downside. Some teams (I think 148/217) have reported that drilling out a rivet and reriveting is faster than loosening and tightening a bolt.

I am skeptical, since tightening an accessible bolt only requires positioning the tools and making a turn or two. This could only be faster if you have a lot of nice sharp 1/4" bits around and the nut or bolt head is positioned such that it’s highly difficult or impossible to reach. Rivets are definitely faster for blind assemblies.

… Upon further examination of the kitbot frame (57 has yet to use it) just about every nut on there looks difficult to reach even on first assembly with nothing in the way. So I can actually easily believe that rivets are more maintainable for the kitbot. Plus, it’s not like there’s not a ton of places to put in extra rivets for good measure. I’ll be considering this now…

Have you tried this? My first impression is that the diameters aren’t going to be right. Normally holes for .250 fasteners are not .250 in diameter, they are a standard clearance hole (in aerospace we use .286 plus or minus .005 for structure) to accomodate machining and position tolerances for parts that are drilled in detail. I don’t know what they are on the kitbot - I would guess they are punched so probably even larger size and position tolerance. For rivets, the holes are usually line drilled at assembly, not in detail, close tolerance (pretty close to .250) and immediately assembled either with the actual rivet or temporary fasteners (“Cleco’s”) to keep all the holes aligned. This is because a rivet depends on interference with the hole when it is bucked (swaged). If the hole is too large, you will not get the shear capability, and tensile capability will be next to nothing.

From what I’ve seen, we practically rivet everything. We can drill out a rivet in a very short amount of time, and riveting a new one in place is almost as simple. Using steel rivets is another story. Many a broken bits from those, but we only use them in places that REALLY need them. Plus, I think it’s easier to just drill in from one side instead of having to fit a wrench or something up into the inside of a possibly cramped frame. Just my opinion though.

Agreed, but remember we’re talking about pop-rivets, not real rivets. I have yet to measure the holes in a kitbot, but I suspect they are only about 0.012 oversize, which is right where the rivets - er, pop-rivets, want them.

Fascinating - here I am thinking that the 600 or so lbs of shear strength of aluminum rivets versus the over 1200 lbs for steel rivets is not enough… But, reconsidering, perhaps it is plenty, especially if we use a good adhesive that will fill in any gaps in the hole (point taken, Gary) and also add to the joint strength - some kind of epoxy paste I suppose. High compressive strength, good adhesion to aliminum, can manage some shock without cracking.

Although steel is less expensive, at $13 per 100 for ‘expensive’ aluminum I can’t see it as being much of an issue.

Don

Our team used the rivets mentioned last year on the kit bot frame. They held up well and my subjective feeling is that it was a little stiffer. We borrowed the rivet gun from a local automotive shop. However, I believe that we have about had it with the kit bot frame. We are resource strapped and need an easy frame solution. This year we are planning on using 8020 Quick frame. By using the bumpers, Poly covering, and electronic board as structural elements, the total unit should be good. I made a proto type base and have been abusing it. It’s holding up very well. However, I have concerns with the corner connectors. They may not hold up for teams that go to several regionals and off season events. The should be fine for our team’s one regional.

I’ll ask, since nobody else has; what kind of rivets were you using in the frame?

I shopped around Grainger (since there’s a warehouse here in town) and found a few potential options:

5LE49 - 1270 shear/830 tensile, $8.86/25. (Aluminum w/ aluminum mandrel)
5LE56 - 2400 shear/1850 tensile, $10.40/10. (stainless/stainless)
5LE53 - 2400 shear/1850 tensile, $7.45/25. (steel/steel)
4AR31 - 1270 shear/1550 tensile, $7.01/100. (steel/steel)

The hitch with 4AR31 is that the material thickness only goes up to 0.250". While that’s on the money for the kit frame (in theory, since you’re joining two 1/8" pieces of metal next to each other), it would seem like going to ones with 3/8" grip range (which all the rest are) would give us a bit of breathing room.

No small part of my motivation to try rivets stems from what Kevin noted: it’s a pain in the butt to get to some of those nuts, particularly once you’ve got the rest of a robot on board. It became the job of one of the kids at each event to check each nut and bolt and tighten as necessary, and Ron Karpinski still handed me a fistful of nuts and bolts at BE.

Greg Needel was talking with me the other night on the subject, and he was a little wary of going all rivets. He threw in the idea of leaving one bolt in on each end of the connections to soak up some of the shear loads. While I’d still take zero bolts over non-zero bolts if it does the job, checking 12 or so bolts sure seems a lot easier than checking the (if I counted right) 72 on a standard kitbot.

1618 is getting the riveter Don linked to, though the team is currently on a Thanksgiving hiatus until the 28th. I’m hoping we can get some knowledge before the holidays, or at least before Kickoff.

Correct me if I’m mistaken, but going all rivets on the kitbot like this isn’t a permanent solution even for a season, is it? Just bring the screws and nuts along, and barring a catastrophic failure in a match, you can always drill them out and put bolts back in fairly easily within a match or two, correct?

If you can’t replace the rivets in an hour or two… well isn’t that the point?

We used rivets from Mcmaster-carr. Part # 97519A653. I would caution riveting right from the beginning. If you make a mistake and need to drill out the rivets, it’s a pain. We put the frame together with bolts and riveted after we were sure of the dimensions. There are higher strength rivets available, but they are very very difficult to remove. For the last couple years we have used the kit frame and riveted 1 x 1/8" 6061 angle on to the KOP base. It has been very effective. Our team lis low on skills, money and machining ability. It worked for us. Last year it took the students forever to build the frame. This year we are going to use the 8020 quick frame. It realy is quick to build a frame.

One more detail on the 1/4" rivets. We did not use back up washers. It is a real pain to get them on inside the frame. If they are not held perfectly flat against the inner wall you will have a bad rivet joint. On the 1" angle 1/8" rivets we always use backup washers. They are critical to a good lasting riveted joint. We have not had a steel backup riveted joint fail in 4 years of using rivets. However, a couple joints were accidentally done with AL rivets and those joints did loosen up.

Alright, a bit of an update on the matter.

We tried the Harbor Freight riveter. It failed; the riveter could not reliably hold onto the mandrel when we needed it to and release when we needed it to. After struggling through two riveters’ worth of jams and stripped threads and frustrations, we switched to the Marson Big Daddy Riveter (the name of which has already become a running joke within the team) today, and knocked it out in about a half-hour. It’s so smooth and easy to work with, it made me genuinely giddy. That riveter runs about $100, but it’s some of the best money the team has spent all season. The Grainger part number is 4Y128; on McMaster-Carr, it’s 90239A510.

We started with steel rivets, but a bit of consultation with some folks (along with worries as to whether it was the rivets that were causing our riveter issues with the Harbor Freight unit) led us to switch to Grainger part 4Y031, aluminum rivets with an aluminum mandrel and grip range right up to .250". With all the rivets in place, the frame feels fine–maybe even better than last year’s with the absence of all the little wiggling that happens with bolts.

We’ll be building our drivetrain over the next week (maybe this Saturday if the AndyMark order gets in, which we’re doubting); we’ll know more about how they hold up as build season rolls on.

Ahhh, my favorite subject: Rivets.

We use rivets (pop rivets) on every joint on our robot that connects to sheet metal. Even items that need to be disassembled quickly are held together with rivets. I swear by them. We use 5/32" and 3/16" aluminum rivets with aluminum mandrels. We also use 3/16" alimunim rivets with steel mandrels in certain high load areas. We have used a total of 20 of these “high load” rivets in the last 3 years (2005, 2006, and 2007). We are not using any steel rivets this year. We have approximately 2,500 rivets on our robot this year.

I challenge anyone using a nut and bolt to disassemble and reassemble your bolted joint to beat my students in a timed race with us using rivets. We use a pneumatic rivet gun from Harbor freight. They are a whopping $24 each. They last about 3 seasons and then you have to get new ones.

Attached is our gun in action on a previous robot.

I do not think you need steel rivets, but I have no test data to back that up. I can tell you this: we have never failed even one rivet on our 2005, 2006, or 2007 robots.

Riveting the Shuttle Assembly.JPG


Riveting the Shuttle Assembly.JPG

We’re going with rivets on the kit frame as well and wherever I can convince students it’s simpler. Oddly, we’re using the harbor freight riveter and we aren’t having any issues aside from and operator error tonight. I think it’s important to make sure you’re using the right tip and that the internal spacing is correct since that affects the jaws clamping, releasing, etc.

Our frame feels pretty good, but I was a little concerned in a few places. We’ve drilled a few extra holes to fill with rivets to add a little strength. I’m mostly concerned with the lack of precision in the kit frame. There’s a distressing amount of slop in all the joints, such that only two out of the three sides of joint are being properly clamped. The third side ends up with about a 1/16th gap that the rivet won’t clamp up, so rivets on the side aren’t going to do much against shear loads. Or much of anything, really.

Also, to Paul, do y’all use the rivet gun in the Pits? If so, are you bring in a compressor/etc. or do you have some other solution?

We use the riveter in the pits. We bring a pancake style compresor that is sold at Home Depot or Lowes. It is the same compressor finish carpenters use for their finish nailers.

Glad you guys are having good luck with the rivets. They have been a very successful method for our team since 2002. One more tip. When drilling for the 1/8" rivets we use double end short drill bits that are typically used by sheet metal duct fabricator. They drill aluminum well and when the bit dulls just flip it. Once again we have never had a riveted joint fail. About 4 joints over all those years loosened a little.

Billfred, we also had similar issues with the Harbor Freight riveter - it wouldn’t let go of the mandrel after popping the rivet.

We fixed it with a liberal application of grease to the gripping jaws. After removing the outer tube, we also unscrewed the inner tube. Some grease on the jaw outsides and they were reinserted, along with the spring, and tightened. Then we added a healthy dollop of grease to the inside of outer tube (and got it all the way in to the bottom, maybe 1/2" deep, around the screw in nose-piece) and screwed it all together - now it works like a charm.

You definitely get what you pay for.

Don

I add a little twist to that old phrase:

You usually don’t get what you don’t pay for.

Anyways, thanks for the rivet gun lube instructions, that is something I never though of and sounds very helpful!

We’ve had lots of small rivets fail, but we use them where we shouldn’t, and we also use them as sort of a safety valve…the little rivets break off instead of letting the mechanism get bent up.

Lubrication could well be an issue with the HF riveters, now that I think about it–over the multiple times I disassembled it, it appeared that there was only just enough lubrication to get your hands messy.

Of course, to fix that would involve getting the kids away from the Big Daddy Riveter. That might be a task. :wink:

I challenge any team to build their robot with 95%+ rivets :smiley:
Team 1501’s Scorpion had 1900+ rivets :ahh:
In the three years that we have been around we have probably used almost 5000 rivets.
Beat that, if you can :smiley: