# Why would an AC motor heavily shake when driven with certain frequencies?

I have an AC motor from an old TCL 160 Boxford NC Lathe, it has three wires named U V and W.

I thought that incorparating the original driver to a modern control interface card would be difficult or unreliable, so I got the motor out and wired it to a Lovato VE1 04 A240 3-phase AC motor driver for tests.

Motor is free on a table and shaft is not connected to anything. When I run the motor using that driver, between the frequencies 20 and 40 Hz, motor starts to shake heavily as if there is an eccentric disk on the motor shaft. But on the other frequencies, 0-20 Hz and 40-100 Hz, there isn't any shake it runs considerably smoothly so I think it's not related to a mechanical problem like a bearing failure.

The original driver was a 0.43 kW one and the one I'm testing is 0.4 kW, and I set the maximum current to 1.5 A just to be safe.

I don't know if the motor is an AC synchronous or AC asynchronous, the only guess I can make is that the motor is AC synchronous because it feels like if the rotor wasn't able to catch up with the frequency fed into it and is shaking like a stepper motor. Also it runs at 1500 RPM with 50 Hz input, and 3000 RPM with 100 Hz input, as far as I can measure with my smartphone stroboscope.

What do you think the reason would be? What would I do to make it run correctly?

• Mechanical resonance. – Oldfart Sep 7 at 16:34
• @Oldfart What would be the source of mechanical resonance? There should be no flexible part inside the motor itself. I've seen examples of rotational mechanical resonance in rotating flexible long shafts but i think this situtation shouldn't be purely mechanical. Would you please explain more about your guess? – Ömer Gezer Sep 7 at 16:40
• What happens if you clamp the motor down instead of letting it swing around freely by the wires like a mass-spring system? – Spehro Pefhany Sep 7 at 16:49
• See Charles Cowie answer below. You don't need flexible parts to get "shimmying". Connect an unbalanced flywheel and your system will shake itself apart: no flexibility anywhere. – Oldfart Sep 7 at 16:55
• @SpehroPefhany when in clamp the motor down the table(a 1x3 meter one) by pushing it down, it shakes the whole table. i also noticed that there are current imbalances in the motor wires. Measured with a clamp meter, two of the phase wires conduct 1.4 A while the other conducts 0.9 A – Ömer Gezer Sep 7 at 16:57

Motor is free on a table and shaft is not connected to anything.

It's most likely an imbalance in the motor plus a mechanical resonance in the "mounting".

Try clamping it down to a nice solidly built table or bench, and repeat your test. Chances are that the problem will go away, or at least be minimized and shifted in frequency.

• So I mounted the motor in it's original configuration, in the lathe, connected to the spindle with the belt. The shaking is almost completely gone as you would suggested. This is the first time I've seen such a thing and it's been a good experience. Thank you for all your suggestions. If you have any other examples on the topic, I'm interested in seeing them because I'm studying in system dynamics masters and really interested in such situations. – Ömer Gezer Sep 7 at 17:18
• All mechanical parts have many mechanical frequency resonances and when excitation spectrum times Q of resonance is amplified , you get vibration g. loading acts as a filter with real and reactive LPF features and also possible resonance. Fourier analysis , Nodal Analysis, are common maintenance damage prevention and design improvement methods. e.g. bearing wear in large machines, generators, motors. turbines – Tony Stewart Sunnyskyguy EE75 Sep 7 at 17:35
• I'm not sure how far along you are, but consider a mounting system that responds to force as a typical 2nd-order LPF: $\frac{P}{F} = \frac{(1/K) \omega_0^2}{s^2 + 2\zeta\omega_0s + \omega_0^2}$. Now let $\zeta$ get really small, and tell me what happens when $s = j\omega$. With structures, it's hard to make the damping for the resonances large, so the usual effort involves making the resonances high enough that they won't be excited. – TimWescott Sep 7 at 17:56
• @TimWescott It makes sense that when ζ gets small, the resonances will be more apparent, but it was counter-intuitive to me since i've run about 20 different AC motors the same way, unmounted on a table, and it's the first time i've seen such a thing. There were also motors that are similiarly sized, similiarly massive and rated for a similiar power among those 20. – Ömer Gezer Sep 7 at 18:40
• I suspect you had just successfully dodged the resonance frequencies, and possibly were always above the first resonance. It's actually a thing on some aircraft tachometers to have red zones that are below the maximum speed that you're only allowed to sweep through, but never dwell on, because there's airframe resonances at those points. – TimWescott Sep 7 at 18:48

The problem is almost certainly due to mechanical resonance. The rotor may be inadequately balanced. There could also be some damage to the motor such as a broken blade on an external or internal fan. There may be a certain amount of imbalance due to the motor being operated without a key in the shaft. It might be possibility that a broken rotor bar could cause a problem like that.

• Okay, i forgot to mention but the motor shaft has the original pulley on it, the shaft key i also in it's place. Maybe there is a broken blade inside but i can't know until i disassemble it. Should i disassemble an AC motor? As far as i know, in case of stepper motors, removing the rotor from the stator is a very delicate task or it would lose its magnetization. – Ömer Gezer Sep 7 at 17:00
• Stepper motors have permanent magnets in them; that's what can lose magnetization. If it's an AC induction motor then there's no magnets in it. There may still be a centrifugal switch, and you need to take normal care of the bearings, but that should be doable. If you do a search for "AC motor disassembly" on YouTube your problem will be sorting out which video to watch, not a lack of videos. – TimWescott Sep 7 at 17:06
• The current imbalance certainly should be investigated. If you can do a coast (freewheel) stop, you can check for vibration during coast-down from maximum speed. That will tell you if there is something electrical causing the problem. If there is some kind of tachometer or a brake connected to the two extra wires, it may be broken. – Charles Cowie Sep 7 at 18:13

Note that even a well-mounted AC motor without load can oscillate around its in-phase position. Typically rotor and the electronics are designed to dampen such oscillations, but with a large discrepancy of available power and mechanical power drawn from the system, you can get problems.