View Full Version : Discussion Iron losses
andrew b
Dec 19, 2006, 01:35 PM
Can anyone give me a "rough" figure for the frequency at which normal soft iron laminations and common enamelled copper windings start to become inefficient? Do ordinary bridge rectifiers lose afficiency above 100Hz? Do ordinary transformers lose efficiency at 300Hz?
Thanks in advance
vintage1
Dec 20, 2006, 04:45 AM
Iron losses covers several things...hysteresis losses, frictional losses in bearings..these are independent of frequency ..no hat not quite right..these are independent of SWITCHING frequencies and the like but are related to RPM directly and eddy current losses which are frequency related.
I do know that apart from audio transformers, which struggle to make the upper end of the frequency spectrum, most transformers over a couple of KHz tend to be ferrite. You only use iron either at very low frequencies or when you must have the greatest inductance per turn. And the iron has many grades too..
Most rectifiers are petty good up to 20Khz or so, but Schottky are better from a few Khz up..well they are better all the time, but more expensive..
Discharger
Dec 21, 2006, 06:56 PM
The ARRL Radio Amateurs Handbook is a great source of info and may provide further answers to your queries
Lucky Dog Rex
Dec 25, 2006, 05:43 PM
On the rectifier, what happens is there is capacitance in the junction. At higher freqs, this becomes a leakage path. There is some loss, which generates heat, which will fail the junction, typically to a dead short. It all depends on the diode, not being flip, but they are all different. You can dig out specs if you have a part no. Otherwise, just swag, rectifiers should be good to around 1-10KHz.
Schottky diode much faster:
http://en.wikipedia.org/wiki/Schottky_diode
Transformer. Power xfmr designed for 50/60 Hz will rapidly loose efficency at higher/lower freqs, typically 10% or less range. Lower freqs core saturates and heats, higher freqs windings loose coupling, impedence goes up way fast, so its a high z input.
Audio xmfrs work over a wider range, typically something like 50-20KHz.
Copper is copper, but the iron can and is made to a variety of specs.
vintage1
Dec 25, 2006, 09:37 PM
Indeed. To get a broadband audio transformer takes a LOT of iron in very thin laminations..and its pretty specialised iron.
Even at a single frequency, more iron may be needed..one audio project I worked on we had to scarp a whole batch of 'cheap' toroidal mains transformers because they were running the cores to part saturation, and the magnetic fields were leaking into the circuitry instead of staying in the toroid.
A typical QUALITY valve audio amp will have the output transformer several times larger than the mains transformer, though theoretically it delivers less power. It has to go lower in frequency and deliver much less distortion..
vBulletin® Copyright ©2000-2009, Jelsoft Enterprises Ltd.