2
\$\begingroup\$

I am not an EE. Most resolvers I come across have 6-wires:

enter image description here

I have even come across 5-wire resolvers. Could we get this down to 4 wires?

For example, an RVDT (Rotary Variable Differential Transformer) requires only 4 wires: 2 for the primary coil, 2 for the secondary coils if they are wired in series.

Of course, with RVDTs, the secondary coils run parallel to each other.

enter image description here

What if the secondary coils were wired in series, where the secondary coils were perpendicular to each other like in a resolver?

Could I still determine the angle of the rotor?

enter image description here

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Since you'd get the sum Sin(θ) and Cos (θ), how would you distinguish quadrants?? \$\endgroup\$ Commented Feb 14, 2022 at 19:54

2 Answers 2

1
\$\begingroup\$

You can get down to 4 wires, if you have star or Y connected 3 windings, and you have steel rotor, and you calculate the inductance of the windings. Assuming you have a microcontroller smart enough to read such a thing - use AC signals at high frequency to messure inductance, rather than slow signals that are used in your examples. And either read the windings one by one or try to use different AC frequencies on each winding.

3 wires if you have delta connected 3 winding, but reader is more complex - adc and dac on the same outputs. And reading is more complex too - need to account for a long path too

\$\endgroup\$
7
  • \$\begingroup\$ Will the secondary windings still remain perpendicular to each other? \$\endgroup\$ Commented Feb 16, 2022 at 21:30
  • \$\begingroup\$ @apprenticeprogrammer 3 windings with 120* between each is easy to do and what is usually done. 2 windings with 90* between them would require excentric rotor and would be sensitive to noise and require much more complex program. But yes, both are possible. \$\endgroup\$ Commented Feb 16, 2022 at 21:34
  • \$\begingroup\$ Thanks for the reply. Where can I find more information about these configurations? \$\endgroup\$ Commented Feb 17, 2022 at 23:47
  • \$\begingroup\$ @apprenticeprogrammer i've not seen them being used in this role, i just assume it is possible. But closest things are mri-q.com/real-v-imaginary.html and en.m.wikipedia.org/wiki/Synchro but neither is popular at precise angle measurment today afaik. Because they require much more complex program to operate than simpler options with more wires. \$\endgroup\$ Commented Feb 18, 2022 at 1:33
  • \$\begingroup\$ I see. Unfortunately, I have to use the resolvers we already have so I wouldn't be able to use the 3-winding method. I also don't think I have the option in modifying the rotor. It seems to me the cheapest option for me is too use a 5-wire resolver. Thanks for your help. \$\endgroup\$ Commented Feb 18, 2022 at 15:59
0
\$\begingroup\$

Could we get this down to 4 wires?

Yes. If each coil has one end connected to ground, you would have only 4 wires, sine, cosine, rotary and ground. Obviously, however, the circuit that interfaces with such a 4 wire resolver needs to supply and interpret single ended signals appropriately.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.