Just bought 2 miniature audio transformers to convert single ended to fully differential, measured DC resistance between pin 4 and CT (40 ohm), pin 6 to CT (45 ohm), so CT is not in the middle. Will this cause offset and how to eliminate it?
Think about this for a second ..... Wires wound around a former; turns at the beginning are shorter because the diameter is smaller then as the windings increase in number, the diameter increases making the DC resistance larger but the AC resistance or inductance stays the same.
Agree. The inductance / AC impedance depends on the # of windings, that should be the same.
There's no DC across the transformer (there better isn't any!) so the DC resistance is largely irrelevant.
Also, be careful with measuring DC resistance. That sends a DC current through the transformer, and some could get magnetized which is Not Good.
Jan
There's no DC across the transformer (there better isn't any!) so the DC resistance is largely irrelevant.
Also, be careful with measuring DC resistance. That sends a DC current through the transformer, and some could get magnetized which is Not Good.
Jan
If you must have a centre tapped grounded output transformer, use a bifilar wound one from Jensen or similar so the DC resistances of each ½ are near identical. But if all you need is a differential output, why would you ground a center tap?
Also consider a hidden in plain sight problem: bifilar wound transformers have *large* capacitance between both windings, cause by all that loooonnnngggg wire sitting side by side, touching each other and separated by enamel thickness only (which is minuscule)
The very definition of a capacitor: 2 conductive (as a particular case: metallic) objects, separated by a dielectric.
And the shorter the distance, the higher the capacitance.
That´s why "in general", bifilar Audio transformers are avoided, and highest quality ones split into multiple interleaved sections, which minimize stray capacitance.
No problem with "single Audio frequency" power transformers of course.
Bifilars can be acceptable in some applications,where reactive capacitance is high compared to impedances involved and bad effects happen outside the audible band, such as some driver transformers.
In a 600 ohm line out transformer, specially if driven from a lowish impedance source?
Well, there it might be a less stringent situation.
But that must be analyzed case by case.
The very definition of a capacitor: 2 conductive (as a particular case: metallic) objects, separated by a dielectric.
And the shorter the distance, the higher the capacitance.
That´s why "in general", bifilar Audio transformers are avoided, and highest quality ones split into multiple interleaved sections, which minimize stray capacitance.
No problem with "single Audio frequency" power transformers of course.
Bifilars can be acceptable in some applications,where reactive capacitance is high compared to impedances involved and bad effects happen outside the audible band, such as some driver transformers.
In a 600 ohm line out transformer, specially if driven from a lowish impedance source?
Well, there it might be a less stringent situation.
But that must be analyzed case by case.
Last edited:
Bifilar transformers are transmission line transformers, so if you match the impedance to the transmission line's characteristic impedance (typically around 50--100 ohms), all the reactances theoretically tune out to give wideband operation (upto GHz is routinely seen in RF bifilar transformers for instance). In practice you need to reduce inter-winding capacitance between successive turns of the pair for best performance (perhaps using twisted pair bifilar, reducing packing density).
Alas that 50--100 ohm range is rather limiting.
But it is not a 40 Ohm audio transformer, it is? More likely 400Ω, or 1kΩCT.
Usually what you want is equal Turns Count. As Jon says, turns and resistance are not the same when diameters are different.
If you truly needed exceptional balance, such as for duplex speech, you can get that. Costs a LOT. Is pretty much an old telephone company specialty.
Bifilar is not the only way to get balance. As said, capacitance sucks. For low impedances they usually split the winding in four quarters.