331 acceleration pitch change thing

[stan claire]

what is it called when a 331 is accelerating and then the pitch of the motor dips down and back up again? ive struggled ages to try to find the "official" term

 

[D365]

Can you point me to a timestamp on a YouTube video? I'm not sure what you mean; the frequency of the motor sound increases slowly and gradually.

 

[stan claire]

 

[swt_passenger]

Welcome to the forum @stan claire - I think it’s called pulse width modulation (PWM) of the AC motor drive. It’s basically the way the variable frequency 3 phase motor supply is constructed from DC within the traction converter.

Theres a video here; check out about 1m 30s in:

 

[D365]

I think it’s pulse width modulation (PWM) of the AC motor drive. It’s basically the way the variable frequency 3 phase motor supply is constructed from DC within the traction converter.

Theres a video here; check out about 1m 30s in:

Thanks for the video and the summary explanation.

 

[Ashley Hill]

Yes,excellent. Is this what is called Chopper Control?

 

[swt_passenger]

Yes,excellent. Is this what is called Chopper Control?

Not usually, AIUI the term chopper control is normally associated with DC traction motors.

The varying frequency warbling sounds associated with PWM mean you have AC motors.

Before the widespread use of 3 phase variable voltage and variable frequency drives, or “traction converters”, AC overhead supplied trains had DC motors. Nowadays all modern EMUs have AC motors, even when running on DC third rail.

 

[Ashley Hill]

Not usually, AIUI the term chopper control is normally associated with DC traction motors.

The varying frequency warbling sounds associated with PWM mean you have AC motors.

Before the widespread use of 3 phase variable voltage and variable frequency drives, or “traction converters”, AC overhead supplied trains had DC motors. Nowadays all modern EMUs have AC motors, even when running on DC third rail.

I’ve heard that phrase used around modern electric traction for years. The presenter here often referred to chopping hence my question. Cheers

 

[swt_passenger]

I’ve heard that phrase used around modern electric traction for years. The presenter here often referred to chopping hence my question. Cheers

Yes I think there’s a subtle difference in the usage, maybe that the “chopping” is only a part of a PWM controller, but in chopper control of DC its basically the whole function.

 

[Pigeon]

"Chopper" is a vague term which usually means something like a phase angle controller as used in boggo domestic light dimmers. It technically is a kind of PWM but it is rather crude. It operates by chopping off a part of the beginning of each half-cycle of the mains waveform so only the remainder is available for supplying power. It was basically the best you could do back when the best high-power switching devices available were thyristors, which are good at handling high currents but are a bit slow and you can only turn them on, not off - you have to rely on the AC waveform going through zero to interrupt the current through them to make them turn off. You can't (well not straightforwardly) use them to generate arbitrary frequencies of AC from a DC or fixed-frequency-AC supply, so you are limited to a fixed PWM base frequency of twice the mains frequency.

When high power MOSFETs first started to be available it was an absolute revelation. Suddenly there were switching devices that you could switch on and off whenever you want, as fast as you want (at least, far faster than there is any need for in a traction motor drive) and they were "easy to drive" (actually they were bloody difficult to drive, but everything else was much worse). All the interesting efficient versatile techniques for driving motors that had previously only been practical for low power machines could now be used on great big things as well, instead of being restricted to slow clunky methods that couldn't do some things at all.

I don't know what's in that video (it's far too much of a pain in the arse to mess about watching any videos except those filmed out of the front of a train) but the screenshot that appears is enough. The motor drive first converts the 50Hz AC off the overhead into DC, and then converts the DC back into variable-frequency (= variable speed) and variable-voltage (= variable power) AC to feed to the traction motors. It does that by switching very rapidly between full-on (positive or negative) and full-off, so the average comes out somewhere in between depending on the proportion of on to off. The switching frequency is much higher than the output frequency you're trying to achieve. You can see that being done on the screenshot: the large rectangular waveform is the rapid-switching-from-full-on-to-full-off bit, and the small wiggly one is the averaged motor current waveform derived from that.

Note that this means there are actually two frequencies at the same time coming out of the drive: the low frequency which you're actually trying to generate, which is what determines the speed the motors are trying to go at, and the remnants of the high frequency from the rapid-switching bit, because the averaging process doesn't remove it perfectly, which is what makes the sort of musical noise you hear in the train. That remaining high frequency component is a nuisance, which increases the inefficiency of the circuit and the stresses on the components, and the higher it is the worse the adverse effects are.

For small changes in speed it is easiest to have the switching frequency be a fixed multiple of the frequency you're trying to produce, so as the speed goes up both frequencies rise together. You would see this on the display in the screenshot as the waveforms being simply squashed in the horizontal direction. But there are limits to how far you can take this; if you go too fast you start running into the adverse effects mentioned in the previous paragraph, and if you go too slow the averaging bit doesn't work properly. So you have to keep the switching frequency between those two limits, but at the same time you still want to be able to make changes in the output frequency which are much larger than that.

So what you do is you change the multiplication factor between the output frequency and the switching frequency. As you speed up, and the switching frequency begins to approach the upper limit, you switch to using a smaller multiplier, so the switching frequency suddenly drops back towards the lower limit while the output frequency can continue to rise. This is what you can hear going on as the train accelerates: the output frequency, which is directly linked to the speed of the train, but is too low to make a noise, continues to rise steadily, while the switching frequency multiplier keeps stepping down, and you hear that going buuuuiiiii-buuuuiiiii-buuuuiiiii as the switching frequency repeatedly rises close to the upper limit and then gets knocked back to the lower one.

 

[D365]

I don't know what's in that video (it's far too much of a pain in the arse to mess about watching any videos except those filmed out of the front of a train)

Pity; it's a good video.