2026-02-16

Last modified: 2026-02-16 04:45:07

Car engine mounts

Let's actually work out if mounting the engine only by the footrest holes is adequate.

So I need to know how much torque is on the gearbox output shaft at peak torque in 1st gear.

And therefore how much tension is the bolt under, and how much margin does it have until it strips the thread?

If we have 16 hp, that's 12 kW.

I think peak torque is at about 8000 rpm, and at 8000 rpm my gear ratio spreadsheet puts the gearbox output speed in 1st gear at 973 rpm.

So if you have 12 kW at 973 rpm, how many Nm is that? About 118 Nm.

I have a 16T 420 sprocket which has 12.7 mm pitch, so 32 mm radius, so it is pulling on the chain with 3700 N. (As a sanity check, this is well below the breaking strength of a 420 chain, which is apparently more like 16000 N).

And then let's imagine that this force is pulling horizontally, and the engine is going to want to pivot around the rear-most bolt locations, putting tension on the front-most bolt locations.

From looking at pictures of the engine and measuring in CAD, I think the rear-most bolt holes are approximately directly below the gearbox output shaft and 94 mm away, so the torque on that "pivot" is 350 Nm.

And then it looks to me that the front-most bolts are about 44 mm forward of the rear-most ones, so the force on those bolts is about 8000 N. And there are 2 of them, so 4 kN per bolt.

Grade 8.8 tensile strengths found online ( https://www.engineeringtoolbox.com/metric-bolts-minimum-ultimate-tensile-proof-loads-d_2026.html ):

• M6: 16 kN
• M8: 29 kN
• M10: 46 kN
• M12: 67 kN

So 4 kN per bolt is not a big deal even if the bolts were only M6. I know they're bigger than that but it's not convenient for me to measure right now.

The 4 kN is (probably?) on top of the preload in the bolt, but in conclusion I think it is OK to mount the engine purely with the footrest mounts.

ChatGPT review is:

• chain snatch etc. can multiply your loads 2x-4x
• assuming equal distribution per bolt is not sound, in worst case one bolt will take 100% of the load
• bolted metal joints usually have more stiffness than the bolt, so taking a 4 kN load through a bolted joint will probably only add like 0.8 kN to the bolt tension

So taking the worst case from there (chain snatch gives 4x load, and we don't get to spread the load over 2 bolts) our 4 kN goes up to 32 kN, which means it needs to be at least M10, and the female threads need to be long enough that we can actually take 32 kN without stripping them.

But a bolted joint is more complicated than that. My maths is correct if the joint is loose and the bolt itself is taking all of the load, but in reality bolted joints consist of a bolt in tension plus the mating surfaces in compression, and adding load to the joint initially reduces the compression more than it increases tension in the bolt, provided the joint is stiffer than the bolt which it usually will be because the cross-sectional area is so much larger.

OK, so discussing with ChatGPT a bit more, I think if everything is in the worst case, then mounting with the footrest bolts is not adequate unless they're at least M12. But at M10 it's so close to being adequate that I'd probably get away with it if not racing, and if I add any kind of half-hearted brace to pick up the "real" engine mounting holes then it is easily fine.

Apparently these holes are M8, so fine if I use the initial numbers but not fine if you add all the worst-case multipliers. Meh, I think I'll just go with what I was going to do anyway, which is only bolt it with the footrest mounts and worry about it later if it's a problem, but at least now I know it is within the realm of being marginal, rather than wondering if it is obviously not going to work or obviously plenty strong enough.