# Simulating Car Vibrations

1. Jul 11, 2015

### BillIsTheDill

Hi everyone! It's my first time working with vibrations (and on this site) and I'd like to know what the best device is for simulating a car engine's vibrations is. Basically I want to generate a random range of frequencies. I would like to keep it as small and for it to use as little power as possible.
Thanks!

2. Jul 11, 2015

### Dr. Courtney

I'd look up (or measure) what the typical range of frequencies is at or near the point where I was hoping to simulate them.

And do you mean numerically simulate or mechanically simulate?

Finally, the frequencies will not be random, but likely be dominated by a fundamental (the engine speed in rpm converted to Hz) and harmonics (integer multiples of the engine speed).

Last edited: Jul 11, 2015
3. Jul 11, 2015

### meBigGuy

Not sure what your requirements are. You want a signal that represents the vibrations in some way? You want to drive a shaker table? You want to build a shaker table?

How do you want to control the vibrations? They will vary with engine speed. You want 1 speed? Any speed? Ramping speed? Random speed for random periods?
Vibrations during warmup are different than normal operation,
You want vibrations on the engine itself, or on the car body.
Also depends on the engine. A 4 cyl Volkswagen is different that a 12 cyl Jaguar.

Regardless, you will need to make measurements to derive suitable equations for whatever it is you are actually doing.

4. Jul 12, 2015

### anorlunda

First, follow the good advice meBigGuy gave. Second, the simplest and least expensive variable frequency vibrator I can imagine, is a bent nail locked into the chuck of a variable speed cordless drill.

5. Jul 12, 2015

### OldEngr63

The cylinder gas pressure function in an IC engine requires harmonics out to about order 18 to get a reasonable representation. That means that, for a two stroke engine cycle, order 0 through 18 give a pretty good representation. For a four stroke engine, the same range applies, except that the orders increment by 1/2, rather than by 1. Since cylinder pressure is the ultimate driver behind most vibrations (except for road induced vibration), you have to be looking at a pretty complicated physical simulation if that is in fact your goal.

6. Jul 17, 2015

### BillIsTheDill

I want to simulate a Mitsubishi air compressor, and I've taken a few measurements. It doesn't need to be very close, basically I just need to simulate ambient vibration in certain situations. Is there any way I could use something much smaller, like a sound emitter at the same frequencies? I'm not sure if I can get the vibrations to be strong enough though...

7. Jul 17, 2015

### Nidum

Record a real one working over a range of conditions and play back through a stereo sound system . May need to be a high power amp and speakers though to get full effect .

8. Jul 17, 2015

### Nidum

Alternatively do a frequency analysis on same recordings and use data to synthesise an input signal for sound system .

9. Jul 18, 2015

### meBigGuy

The sound emitter needs to be loud enough to couple to the mass you want to vibrate. Non-trivial, to my mind, for a stock speaker system.

But, you can use a speaker to build a cheap shaker table. lay something in the cone, or tape it to the cone, and it will shake up a storm.

BTW, I've never tried this.

If you use a cheap speaker to do this, be sure to isolate it from front to back (basic infinite baffle theory). Speakers cancel themselves out if they are not isolated front to back. (that's why they all have enclosures, and sound tinny without them).