We sometimes learn the most when things go wrong. In that spirit, this post will revisit the problem that kept 6328 off the field at Granite State in Quals 21. Our analysis revealed an issue that could easily affect other teams.
As we were loading in for the match, our drive coach ran over to the stands in a bit of a panic. The red wire in the main power inlet had pulled free of the SB50! Putting aside for the moment how this could possibly have happened, I
sprinted walked purposefully to the pit for tools and parts while our student electrical lead quickly assessed the situation and got ready to make the repair. Unfortunately, we couldn’t quite make it – the gates closed before we could get a new contact attached. We headed back to the pit to take our time with a full repair, borrowing a hydraulic crimper from our friends on Team 95 (thank you, Grasshoppers!). We also inspected all the similar 6AWG connections.
Anyone on the electrical subteam on 6328 will tell you that we emphasize, from the very first thing in training, that there are hundreds of electrical connections in an FRC robot and just one bad one will cost you a match. Students are trained to prepare wires carefully, crimp properly, inspect their own and each others’ work, and always, always, always tug-test any crimp. And they are generally very good about it! So how could this have happened to one of the most critical connections?
So let’s start with the offending contact. Here it is:
We can see a few things immediately.
- It has the expected “double crimp”; our 6AWG crimp tool has narrow jaws and requires side by side crimping on a connector this size.
- It’s crimped enough that it would have held on to the wire pretty well; probably even well enough to pass a light tug test. But it’s also clear that it’s not a very deep crimp, and it’s very plausible it could have worked loose. After all, robots sometimes see a little vibration…
Here’s the crimp tool that was used:
This no-name tool was bought on Amazon in 2017, our rookie year. It cost ~$50, which for a rookie near the end of build season was a sizeable amount to spend on a tool that we used only a few times. We picked it because it was the most affordable option. While I couldn’t find this specific one for sale any more, there are plenty like it, and I’m sure a number of teams have them. It’s a neat design that supports different wire sizes by spinning the hexagonal jaws to a marked size. It takes a good amount of force to close the jaws, which is why they’re narrow and require side-by-side crimps.
You’ll notice there’s a label on the handle that we put on it ages ago, reminding the user that 6AWG wire has a cross sectional area of 13.3 MM2. Why? Well, it turns out that this crimper is sized for metric wire, so the jaw sizes are based on metric cross sectional area. So no big deal, we pick the closest metric equivalent and off we go, right?
This is where the problem lies. There isn’t a good metric equivalent for 6AWG wire – at least not on this tool, and not in many others either. This tool offers:
- 16 MM2 – what was used on this contact – which clearly is really not tight enough for a good connection, at least if the connector shell is thin.
- 10 MM2 – too small, such that it overcrimps and distorts the contact making a large “wing” of crushed material and often bending the body. In fact to get a “good” 10 MM2 crimp on 6AWG you really need to do a 16 first.
Goldilocks would be disappointed – one crimp that’s too large, one that’s too small, but we’re sadly lacking in one that is just right. The SB50 contacts have slightly thinner walls than the ring terminals we use for batteries, so they’re the most sensitive to this problem.
So that’s it – this crimp failed because we didn’t have the right tool to crimp it properly. And it cost us a match. How do we make sure this is the last time we have this problem?
Immediately after the event I set out to locate a tool with true AWG-sized dies. It is surprisingly difficult. Many tools that advertise AWG sizes turn out to actually be metric when you read the fine print. This is certainly true on Amazon, but even the tools sold by a number of brand-name, reputable vendors are metric and have the same steps as our yellow model. This is true for some of the hydraulic crimpers too – even the very nice ones. This isn’t just an issue among the inexpensive hex jaw crimpers.
What we eventually purchased is a Temco TH0006 which can be purchased direct for $130, is a hydraulic crimper, and comes with extra dies for over/undersized connectors. This produces very nice crimps like these. It even crimps in a size marker so you know which die was used, a nice touch
This was an especially frustrating failure, because it was so easily preventable. But we have a few takeaways from the experience:
- Use the right tool for the job. This is especially true with wire crimpers, and we know that. We should have recognized the risk here sooner. But also…
- Just because you got away with it before doesn’t mean it’s OK. We’ve used our yellow handled crimpers for years. We’d thought about upgrading, but “they work” and there was always something more important to spend team funds on. Well, the cost of the better tool would have been less than the cost of a forfeited match!
- Inspect, inspect, inspect. We’re all human. Mistakes happen. Robots take abuse on the field and stuff comes loose. This loose connection could have been caught at lots of points, from when it was first made (ideally) to a pre-match check. The team does a lot of checks already, but this one slipped through. We can do better.
I hope this has been informative and if your team is using a metric crimper, please double check that you’re getting good solid crimps out of it!