I'm working with a audio isolation transformer and it hit me, how exactly does a audio transformer maintain a flat frequency response when by definition an inductor is frequency dependent. I'm guessing its a simple answer and I Googled it but no good hits. Ether way, I thought it was an interesting question.
Thanks
Micheal
Thanks
Micheal
The self inductance of the primary is engineered to be very high so that the magnetising current is insignificant compared to the load current. If the magnetising current is insignificant, it can be ignored, which means, the secondary current is what determines the primary current. In other words, a primary coil with sufficiently high self inductance makes a transformer to behave like an ideal transformer.
A transformer isn't an inductor (it has 4 wires, not 2), and doesn't follow the same equations. Well for a theoretically perfect transformer. At lower frequencies the inductance (or more correctly lack of it) will lead to a tail off in performance. At higher frequencies the nature of the magnetic core will determine performance, as does inter-winding capacitance.
(Ideal) transformer equations: Vp * Ip = - Vs * Is, Vp / Ip = Zp = N^2 * Zl
where N is the turns ratio (primary turns / secondary turns), Zl is load impedance on secondary.
The actual physics is that the back EMF per turn is propotional to (N * dIp/dt + dIs/dt), which means N * Ip + Is is forced to be very small (this is the magnetizing current - or more correctly the MMF). Or in other words. Ip ~= - Is/N, where N is the turns ratio (primary turns / secondary turns). Combining this with Vp = N * Vs leads to the transformer equation.
Or put another way a transformer blocks the AC component of the sum of N * Ip and Is... In the middle of its operating range of frequencies this is effectively complete blocking, leading to Ip = - Is/N as the current relationship.
A point of note is that the inductance of a transformer core is not at all critical, as it is effectively huge compared to signal impedances, so transformer cores usually aren't gapped (unless they carry DC and need saturation control), an ungapped inductor would be pretty unstable in value, so they are usually gapped so that the gap controls/stabilizes the total permeability.
(Ideal) transformer equations: Vp * Ip = - Vs * Is, Vp / Ip = Zp = N^2 * Zl
where N is the turns ratio (primary turns / secondary turns), Zl is load impedance on secondary.
The actual physics is that the back EMF per turn is propotional to (N * dIp/dt + dIs/dt), which means N * Ip + Is is forced to be very small (this is the magnetizing current - or more correctly the MMF). Or in other words. Ip ~= - Is/N, where N is the turns ratio (primary turns / secondary turns). Combining this with Vp = N * Vs leads to the transformer equation.
Or put another way a transformer blocks the AC component of the sum of N * Ip and Is... In the middle of its operating range of frequencies this is effectively complete blocking, leading to Ip = - Is/N as the current relationship.
A point of note is that the inductance of a transformer core is not at all critical, as it is effectively huge compared to signal impedances, so transformer cores usually aren't gapped (unless they carry DC and need saturation control), an ungapped inductor would be pretty unstable in value, so they are usually gapped so that the gap controls/stabilizes the total permeability.
The transformer reflects the secondary impedance to the primary side and therefore makes it appear as though the source is driving the load directly, in spite of the galvanic isolation that exists in between.
To keep it simple, put N = 1 in the following picture.
Google, ChatGPT (and thereabouts) are all idiots trying to make answers by copying and pasting sentences written by real men.