Engine Torques in FEA

[Shaun]

I'm preparing FEA on our chassis and we're arguing over where to apply torques to the engine in order to obtain reaction forces at the engine mounts.

Do I apply a moment equal to the engine torque only at the crankshaft axis or to I apply the torque after all the gear reductions on the PTO axis or other method?

Thanks,
Shaun

[Marshall Grice]

If you can answer where a particular torque is being reacted then you can put it in FEA. For instance, the raw engine torque is reacted by the transmission(all reductions) then the output of the transmission is reacted by the chain, to the diff, etc. The detail of your analysis depends on, among other things, how realistically you apply your loads.

How much time to you want to spend on FEA of engine loads? Realistically the engine load is reacted at the tires through the entire drive train. If i were to take a guess though i'd say it's probably not a good use of time to chase the engine loads that far. There are usually larger loads that will determine the overall size of the engine mounts(torsional rigidity).

[awhittle]

our chassis and we're arguing over where to apply torques to the engine in order to obtain reaction forces at the engine mounts.

Do I apply a moment equal to the engine torque only at the crankshaft

I would think that it all reduces down to the CF of the tire is say 1.6 or so times the weight of the car times the radius of the tire divided the radius of the rear gear. That is about the most tension that can end up on the chain. After that you smoke the tires.

AW

[Storbeck]

I used the output shaft torque, that would be required to smoke the tires with a coeficient of friction of 1.5.

Then I chickened out and made the mounts stronger. Lots of vibration going on there that is difficult to account for.

[JP Venturi]

Here's a little story of what happened to us this year one day before detroit:

Usually our shock mounts are right behind the engine so they react the tension forces between the sprocket and the wheels...This year we moved the shocks and the first thing that happened during our first burnout is that the engine translated longitudinally about 3/4 of an inch.

I would say calculate the torque required to get the wheels to break away (spin out) and that is the maximum torque your chain should see (then make it a little bigger just to be safe).

We had to take about 4 or 5 links out of our chain when it happened...

[Conor]

Originally posted by McGill_Man:
Here's a little story of what happened to us this year one day before detroit:

Usually our shock mounts are right behind the engine so they react the tension forces between the sprocket and the wheels...This year we moved the shocks and the first thing that happened during our first burnout is that the engine translated longitudinally about 3/4 of an inch.

I would say calculate the torque required to get the wheels to break away (spin out) and that is the maximum torque your chain should see (then make it a little bigger just to be safe).

We had to take about 4 or 5 links out of our chain when it happened...

Can you go a littler futher in depth on this or if anybody else would like to elaborate my problem is: I designed and built new differential mounts for our car this year and when we first built the car and went to competition, the chain remained tight when I tensioned the differential mounts to a certain position. Now, at that same position, the chain is very lose, almost as if it has been stretched out. Do you think we're experiencing the same problem? The engine still seems to be solidly in place leaving me completely stumped...

[Marshall Grice]

chains will stretch to a point. More like all of the slop goes the same direction. just retension the chain, you should be fine.

[JP Venturi]

Chains tend to run in just like belts do, but not as much i would think. I know at work when we put a new belt on it has to be re-adjusted a day or so later cause it loses almost half its tension.

If your engine moved you would know...just look around the engine at the tubes, is any paint cracking off? In our case we just had a really long tube with two brackets right in the middle...didnt even have to do the bending moment calculations after it failed...sigh, but this is how we learn..

[Conor]

The engine mounts look fine. I'm very confused because I didn't think a chain could stretch this much. Either way, tensioning further is no longer an option because the way our system is setup, tensioning the chain raises the differential and driveshafts and the driveshafts start to rub on the frame.. poor design, I know. Maybe I'll just remove some links?

[BA]

As far as engine loads go in frame FEA, this is the method I followed:

1. Calculate your chain tension for a max tire friction force.
2. Translate that tension into a longitudial force, vertical force, and moment at the output shaft based on your chain angle and sprocket size.
3. Model the engine in the FEA as a set of frame elements interconnecting all the engine mounts and the output shaft (the size of these elements will determine the engine stiffness -- engines aren't rigid).
4. Apply your translated chain tension loads at the output shaft in your model.
5. Let the computer distribute these loads to the frame model.

It makes the FEA model a bit more complex, but if you're using software like ALGOR that can handle frame and truss elements -- it won't take but a few seconds to solve anyways.

In addition to modelling the engine, I would also place suspension (using a mixture of frame and truss elements) with uprights and simulated wheels that make contact with the ground (good place to apply loads), and other structural elements such as the seat.

Side topic -- I posted a thread a couple of days ago on my frame optimization software, but here's another shameless plug:

http://sourceforge.net/projects/esop

The software also does mixed frame and truss element FEA analysis, in case anybody needs a really archaic piece of FEA software http://fsae.com/groupee_common/emoticons/icon_wink.gif

Happy analysizing!

[SR-Mike]

Just a small point that appears to have been overlooked. The maximum tension that your chain will see is actually related to the maximum torque available at the output sprocket of the gearbox, and not the maximum torque that your tyres can handle without slipping. This is because your diff assembly, driveshafts, wheels, tyres etc all have mass. Therefore if the applied torque is greater than what the tyres can transfer to the ground, the excess torque from the engine will simply be used to accelerate the rotating assembly, ie cause more wheelspin. Hence all the torque available from the engine/gearbox will still be transferred through the chain.

Cheers

Mike

[mtg]

Originally posted by SR-Mike:
Just a small point that appears to have been overlooked. The maximum tension that your chain will see is actually related to the maximum torque available at the output sprocket of the gearbox, and not the maximum torque that your tyres can handle without slipping. This is because your diff assembly, driveshafts, wheels, tyres etc all have mass. Therefore if the applied torque is greater than what the tyres can transfer to the ground, the excess torque from the engine will simply be used to accelerate the rotating assembly, ie cause more wheelspin. Hence all the torque available from the engine/gearbox will still be transferred through the chain.

Cheers

Mike

I'll second that. If you don't take into consideration what SR-Mike describes and design the drivetrain to withstand only what the tires will put to the ground, either your F.O.S. is really high, or something will break.

I've done those calculations myself, and the drivetrain inertia is a pretty big deal. Well designed cush-drive mounts can reduce this load, but it is still there.

[VFR750R]

As any off-road guy will tell you, there's alot more going on then a simplified friction model can tell you. If you are spinning the tires and hit a piece of pavement with higher friction, the load will spike as the tires regain grip. Similarly with bumps, or a diff locking up. I agree with the chain theory. The force on the engine mounts is directly related to the tension in the chain. If you want to reduce the load, increase the small sprocket size.