Pete sat down for a chat with Specialized’s EMTB Product Manager Joe Buckley to find out what was involved in updating the Levo.
Specialized’s Turbo Levo got an overhaul recently, and we took a closer look at how a brand like the Big S develops the new bike and the new e-systems with their eMTB Product Manager, Mr Joe Buckley.
What was the impetus for updating the Levo?
I would say that just like with any bike, we want to keep making it better, right? I always joke that if we, as product managers, made the perfect bike we’d be done and out of a job. The impetus is always to improve. In this category, we’ve always strived to make it ride as much like a traditional bike, we’re trying to make it a cycling experience rather than something totally different.
Right after the last launch we did in Croatia which was in August/September 2018, a group of us got together and we went to Verbier, we got hold of a bunch of competitors’ bikes, we spent a week in Verbier riding a heap of different ebikes. We talked about what we liked and disliked, where we thought that there was room for improvement, both in terms of the bike itself, as well as the e-system.
We learned a lot from that, and we kicked it off like every other project which is to create a sort of virtual bucket of what you want to improve on the bike, then you really start getting into it, then you start to figure out what you can do, what’s actually possible. Sometimes you run into roadblocks with the technology available, and you then have to make compromises.
What came out of all this was that we knew there were improvements to be made on the ride quality, and we knew that there was some e-system stuff that was lacking. Geometry was due for an update as well as kinematics, and for that stuff, we worked with our Ride Dynamics team. Brad Benedict leads that team, a small team that sort of sets the menu for our different ride experiences, whether it’s cross country, trail or enduro. They develop the target kinematic and geometry, for say a trail bike, and we work with that going forward.
Our Swiss team then go back and try to solve all the e-system updates, like the new TCU display for the battery. The basic requirement is to show the crucial information but the realisation was that there was a whole load of information that we could add to it. We’ve always been fans of clean cockpits, so we though why not make the TCU have all the features a good cycling computer would have? We also wanted to make the system more robust. We’d had issues with the wiring harnesses getting contaminated with water, then corroding, so we wanted that better sealed.
Once you’d decided to update the bike, what happened next?
Next step is to start laying it all out. That is when you start to work out what is and isn’t possible. Target geometry might not be possible if you want to have a certain tyre clearance, for example. One of the big challenges for us is the motor. The motor takes up way more space than a traditional bottom bracket so when the Ride Dynamics team sets the table but it’s not always that straightforward. With something like the Stumpy EVO, you just crack on, but on an ebike, the main pivot location is often a problem.
We have to go higher in the main pivot to accommodate for the motor, then we have to start adjusting other pivot points to compensate for that. With the outgoing Levo, we didn’t have a reservoir shock on the frame so that a water bottle could fit. Trying to get the standover as low as we wanted with a water bottle and a reservoir shock this time around presented some challenges.
Shorter chainstays was something we wanted to work on, which was a big deal for us and probably one of the hardest decisions. As the motor takes up so much space, ebike chainstays are always artificially long, because you have to get the rear wheel far enough away to clear the motor and give the travel you want. The last Levo had 455mm chainstays, and by playing a few tricks we could get a few millimetres out of it, but ultimately we knew that it wasn’t going to cut it.
It was a big decision as a 29er brand, with the contact patch and the increased rollover but it’s a much better riding bike with the 27.5” rear wheel, with the better turn-in, and is a lot more playful as a result.
How many people are involved in that process and what do they do?
It’s a huge group. The Ride Dynamics team have a couple of people working to set the geometry and kinematic targets, we have three different CAD engineers working on this… It’s been two years since I have been to the Swiss office because of COVID so I can’t know how many people are working over there, but considering the hardware and software work, there could easily be thirty people working on this. It’s a huge investment to try and pull this stuff off.
What’s involved in updating the motor and other systems for an ebike?
All our software and hardware team are, for the most part, in Switzerland. When we launched the Turbo bikes, we didn’t have any back end support. We’d be replacing parts or individual wires when things went wrong. The team over there has designed the Turbo Studio system which is a system that allows shops to diagnose and get some analytical data from your system. So you can now plug your bike with a USB cable, and the bike will give you and the shop feedback on the motor errors, or any other issues that will help fix problems.
Let’s say you get a bike, you go for a ride and there’s a problem. If you have a system problem and the app is connected to the bike, the bike itself records the error, if you’re connected then it goes to the cloud, you send me your serial number. I could then put your serial number into the Turbo Studio, see what the issue is and get you to head to the shop where the parts would be waiting for you. It hasn’t solved every issue, but it has made problem solving and working on the bikes a whole lot easier.
The team in Switzerland has worked tirelessly to put all this in place. We’re developing these systems on our own, we’re not Bosch that is a much bigger company, or another bike company that buys in a third party motor that has all the software and hardware with it.
There’s pros and cons to this, obviously. We had an integrated battery on our first e-bike, rather than other ebikes at the time that came with bolt-on batteries. If you’re not making your own system, you’ll have to buy a battery and motor from a company that wants to sell motors and batteries, not bikes.
Does it take long to get a working prototype in-hand?
Mules will almost always be in aluminium, because it’s a lot faster to produce them. In some cases we’ll do a carbon mould modification if the change is minimal. Depending on what you’re prototyping, it can take a couple weeks to a couple months. By the time we’ve done our final CAD and done the testing, we open it all up and try to get bikes to market. That’s not the case with this bike, the alloy model will come later but that’s generally how we like to do it.
What made the carbon bike come first?
For this one, we focused on carbon because we had limited resources and we wanted it done. It was a case of do one bike quicker or two bikes slower.
How many prototypes did you make before settling on what would be the production model?
Generally, if we’re talking a carbon frame, from when we hand over the CAD, it could be as quick as a couple of months. You’re not necessarily really riding something that’s safe at that point, but you can ride it. We used to be a bit more care-free and ride a lot more questionable stuff, now that we’re a bit wiser, we’ll try and get something on the lab machines first to make sure it can take a minimum amount of abuse.
We’ll then build a couple up just to make sure that the components actually fit, for one, and make sure the tyre doesn’t hit the frame at full bottom out. The goofy stuff. We then try to see if there’s anything that pops out at us while we ride them. At that point there’s various iterations of carbon layup to get the strength and weight where it needs to be.
Once we get something more durable and ready to be ridden in the field, we’ll do a full run of test bikes. Those actual test bikes might be six months from handing over the CAD drawings.
We were pretty happy with the first run bikes. Then it comes down to shock tuning, the charge port door designs, cable routing… The little details that take a while. The new bike was a revelation. Even if you’re making a cake, you can taste the cake straight out of the oven, but you’re still going to put the frosting on, the sprinkles, the candles, right?
How important are athletes to testing new product?
They’re helpful in stages way before the prototypes. Ride Dynamics works with the athletes to develop geometry and kinematics. The feedback on race bikes gets translated down to the new bikes. That drives a lot of the stuff at the start of the project but less so for prototypes.
This might be a little different if was an Epic or a Demo, but ebike racing isn’t mature enough to have full time athletes to go to.
Beyond prototypes, what form did your testing take?
We have a big lab in Taiwan and a big lab in Morgan Hill, California. We try and split our testing between the two depending on what projects are being done where. The load testing for the initial frames that we get out of the mould is done in Morgan Hill.
Once we start proof testing for production, there’s a massive amount of testing we have to do, especially for this bike as we have six sizes, each one of those front triangle sizes has to go through a litany of tests, the you have the rear triangles… There’s different ultimate tests and fatigue tests on every different frame component in every size. There’s a lot of testing that has to be done.
Even after that, we get a pilot run done, then have to redo all the testing on the pilots to make sure they’re good to go.
People will see us riding a field test bike and assume that we’re good to press the button on production, not realising that we’ve a full year of testing still to do before we even start production.
Favourite moment(s)?
Everyone on the team will have a different favourite moment, but for me, we didn’t have a normal press launch, as we usually would. I was lucky enough to head up to Washington State where a lot of the US journalists are based and do press with them one-on-one. It was great to hear that they all had similar feedback completely separate from each other, like the bike was way lighter than it felt, and way more playful. It was good to hear as it validates how we felt about the bike.
Any disasters?
The global pandemic not withstanding, yeah, there’s always something.
We did have a few issues in testing, things were sailing along nicely, we’d then get to a new test and it fails horribly… That’s when the shit hits the fan. It turns into a giant problem solving mission to not have to add years onto going back to the drawing board.
