I have hand wound a transformer which has a 220v primary winding, two 25v secondaries and a 15v secondary:


simulate this circuit – Schematic created using CircuitLab

I have centre tapped the 15v secondaries (to later rectify to +-25v) and when measuring the voltage between A-B or B-C, I get 25v as I would expect. Similarly between D-E I get 15v.

However, If I measure the voltage between B-D I get 5v, and B-E gives 15V.

A-D is also at 24v and A-E at 34v. I would have expected these to all be ~0v.

Why am I getting these voltages? I would have assumed that the 15v winding would be isolated from the 25v centre-tapped winding? I have checked there are no shorts between any the secondaries. I should also point out that this is under no load.

As I said I hand wound this transformer (well the secondaries), so is this something to do with the direction of the windings? Or unbalanced windings? I would like the 15v winding to be isolated from the 25v ones. Is this not possible with a shared core?

  • \$\begingroup\$ Maybe capacitive coupling. What is the resistance you measure between the windings? \$\endgroup\$
    – jippie
    Dec 24, 2013 at 16:16
  • \$\begingroup\$ about 3ohm A-C and 2.2 D-E. 25v winding also has a larger wire diameter than the 15v. Meter OL's between the 25 and 15v windings (so no shorts) \$\endgroup\$ Dec 24, 2013 at 16:22
  • \$\begingroup\$ I mean the resistance between A-D or A-E, ... \$\endgroup\$
    – jippie
    Dec 24, 2013 at 16:23
  • \$\begingroup\$ Infinite. Well the meter shows OL so I guess at least 100Mohm (dont know the specs of the meter) \$\endgroup\$ Dec 24, 2013 at 16:26
  • \$\begingroup\$ What happens when you load the A-D, A-E, ... voltage with a 1k or 10k resistor? \$\endgroup\$
    – jippie
    Dec 24, 2013 at 16:27

2 Answers 2


A high impedance meter in most cases will give more accurate readings because it does not draw much power from the circuit under observation, but it can give misleading readings when the circuit produces high voltages without a load. What you are observing is the result of an overly sensitive meter combined with inductive and capasitive coupling. If you put a load on your transformer, you won't notice this.


You are looking at capacitive coupling between the windings. This is clear from the fact that when you load the 'ghost' voltage with a small load, it entirely disappears. This is expected behavior with a high impedance voltmeter and a transformer windings on the same core. I believe there are ways to minimize capacitive coupling between windings, but I'm not entirely sure how. Probably a conductive screen between the windings, physically seperating the windings by placing them in another location on the core or a smart other way of separating the windings that I am not aware of Maybe someone with more hands-on transformer knowledge is able to elaborate on that.


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.