And there is the heat due to the primary DCR x primary current.
And there is the heat due to the secondary DCR x secondary current.
Keep in mind, the "current" is the integral of current over a complete cycle of AC.
And the heating loss is current squared x DCR.
Cap input filters have very large peak currents versus the DC current on the B+.
Even though the current is over a short time, current squared makes for lots of heat loss in the DCR of the windings.
Hotter windings.
And, these large peak currents also heat the laminations.
Choke input filters have more average current versus time, and not so very large versus the DC current on the B+, so current squared has less loss in the DCR of the windings.
Cooler windings.
And, these more average currents heat the laminations less.
And there is the heat due to the secondary DCR x secondary current.
Keep in mind, the "current" is the integral of current over a complete cycle of AC.
And the heating loss is current squared x DCR.
Cap input filters have very large peak currents versus the DC current on the B+.
Even though the current is over a short time, current squared makes for lots of heat loss in the DCR of the windings.
Hotter windings.
And, these large peak currents also heat the laminations.
Choke input filters have more average current versus time, and not so very large versus the DC current on the B+, so current squared has less loss in the DCR of the windings.
Cooler windings.
And, these more average currents heat the laminations less.
i do not see the copper heating transfer to the core that much, copper losses are radiated in parts where they are exposed, and conducted where the coil assembly is in contact with the core...
Last edited:
In a normal design under load the majority of losses should be of course in windings ... where current flow ( Ohm's law ) , so those are hotter and transmit heat to the core . If not you have problems , crappy core or so 😛
Last edited:
core losses are pretty much fixed due to the weight of the core and the flux density...the coils transmit its heat to the core but most of the heat is right in the core itself...
Laminations radiate heat to the air around the core.
Windings either radiate heat to the laminations, or the wire retains all of their heat.
I never thought about the thermal resistance of windings to laminations before.
I should have.
But, according to Physics, the rules of thermal dynamics rule.
Heat does not just dissipate, without heating something else, even it is radiated into vacuum space (where the heat still exists) . . .
Conservation of energy.
If the primary DCR is 5% of the rated primary impedance, and the secondary DCR is 5% of the rated output impedance, and the load = rated output impedance, then the insertion loss from DCRs, is 1 dB.
12.5 Watts in, 10 Watts out.
Any loss due to flux causing heating of laminations will add to that total insertion loss of the transformer.
Windings either radiate heat to the laminations, or the wire retains all of their heat.
I never thought about the thermal resistance of windings to laminations before.
I should have.
But, according to Physics, the rules of thermal dynamics rule.
Heat does not just dissipate, without heating something else, even it is radiated into vacuum space (where the heat still exists) . . .
Conservation of energy.
If the primary DCR is 5% of the rated primary impedance, and the secondary DCR is 5% of the rated output impedance, and the load = rated output impedance, then the insertion loss from DCRs, is 1 dB.
12.5 Watts in, 10 Watts out.
Any loss due to flux causing heating of laminations will add to that total insertion loss of the transformer.
Last edited:
heat in traffos are radiated off the surface area of the the periphery of the core, and physics says, the bigger the surface radiating area, the lower the temperature rise per square inch of surface....
paper, like the bobbins and layer insulations used in insulating the windings are a poor conductor of heat, that is why we strive to make the copper current density as low as possible, using a bigger circular mills per ampere of the wire, 500 cm/A being a typical starting point, but 300 to 750 is also used...
paper, like the bobbins and layer insulations used in insulating the windings are a poor conductor of heat, that is why we strive to make the copper current density as low as possible, using a bigger circular mills per ampere of the wire, 500 cm/A being a typical starting point, but 300 to 750 is also used...
Last edited:
