1

We are making a rotating machine where we need to measure the flux in the air gap, however, the flux sensor is too thick to fit in the airgap so a slot will need to be cut into the face of the pole so the flux sensor can slide into it.

But before we do that, we need to determine how much those modifications will distort the flux readings so I was going to make a mock-up.

There would be two setups:

  1. The control setup would consist of a UI core with a DC-excited winding and an air-gap machined into it that would fit the thickness of the flux sensor.
  2. The second setup would be identical except it would have a slot machined into one face of the airgap that the flux sensor could rest in.

Then we would compare the readings of the flux probe at the same DC excitation. Our only goal is to see how much a slot would throw off the flux readings of the flux in the air gap. In this mock setup, we can make the airgap large enough to fit in the flux probe to compare, but on the actual machine the airgap is too small so we would have to make an inset/slot for the flux probe to nestle into but we have concerns about the slow affecting accuracy.

My question is this: Does the location of the airgap in the middle or at the corner of the core matter? Because if the airgap could be at the corner, we could put the coil on a bobbin and on to an I-core and move it back and forth between the two setups (an unmodified U-core and a U-core with an airgap machined into it), rather than needing to wind two separate cores.

I suppose we could do it share a bobbin and coil either way, but if the airgap should be mid-length then that's three fiddly pieces and we might choose against that.

schematic

simulate this circuit – Schematic created using CircuitLab

DKNguyen
  • 56,670
  • 5
  • 69
  • 160
  • I think this is way too little information for any kind of meaningful response. "Rotating machine" is very ambiguous. Also, no idea where the coil windings are at. (How the fringing field interacts with nearby coils matters.) No idea of the range/dynamic range of measurements. (Which means there's no way to answer "does it matter" kinds of questions.) Do you think you could add a lot more detail? – jonk Dec 10 '20 at 00:43
  • It doesn't matter. The results will be equally poor either way. –  Dec 10 '20 at 00:46
  • I hope the core material isn't ferrite - you'd have to use diamond and coolant. – rdtsc Dec 10 '20 at 01:15
  • you can split up the gap into 2 gaps in series by just inserting some Kapton or mylar at both ends of the U. Also why use a flux sensor when you can just wrap a couple turns of a sense winding and use an integrator (or integrate on a computer)? – electrogas Dec 10 '20 at 04:40
  • @rdtsc No. We would choose something like steel or iron. No material engineered for a core. – DKNguyen Dec 10 '20 at 05:02
  • @electrogas A sense coil won't work because this will basically be a locked rotor test with DC field excitation only. – DKNguyen Dec 10 '20 at 05:03
  • @jonk Rotating machine just means a generator or motor. I can't actually describe the machine itself but I don't think it's important here since this will essentially be a locked rotor test with DC excitation of the field. The coil would be wound around the I-core. I don't know the range because I am not very concerned about the range (should I be?). We would just pump enough current into it to get a reading on the probe since it's a comparative test to see if having a slot would drastically throw off the reading of flux in the airgap. That's the only thing we're after. – DKNguyen Dec 10 '20 at 05:09
  • @jonk Fringing of the coil with coil so close to the airgap in the diagram on the left was what prompted me to ask the question. – DKNguyen Dec 10 '20 at 05:11

0 Answers0