There is no such relation with VA rating.
For a very rough estimation of L you need # of turns, core permeability, core cross section, air gap and length of the magnetic path. With this you can do some calculation which give you an order of magnitude. And yes, it probably will be high, in the several H range.
There is no absolute value for a transformer's inductance. It's value varies with both frequency and level at which it is measured. So modelling a transformer as coupled inductor will not give very accurate results. Might be no big deal if you just want to model a power supply but for signal transformers it's useless.
For better results consider using a more complex model that takes winding resistance, leakage, internal capacitances and core losses into account. A few terms into google should help you on the way. Still, even the more complex ones do not include the level and frequency dependency of L.
For a very rough estimation of L you need # of turns, core permeability, core cross section, air gap and length of the magnetic path. With this you can do some calculation which give you an order of magnitude. And yes, it probably will be high, in the several H range.
There is no absolute value for a transformer's inductance. It's value varies with both frequency and level at which it is measured. So modelling a transformer as coupled inductor will not give very accurate results. Might be no big deal if you just want to model a power supply but for signal transformers it's useless.
For better results consider using a more complex model that takes winding resistance, leakage, internal capacitances and core losses into account. A few terms into google should help you on the way. Still, even the more complex ones do not include the level and frequency dependency of L.
maudio said:
Core cross section and lenght of magnetic path scale with the transformer VA. Frequency and signal level are obviously that of the mains. Obviously when mains transformers are designed, they must calculate the needed minimum inductance.
Why would I model a signal transformer when I specifically wrote "power transformer"?!
Nixie said:
Mains voltage and frequency are not the same everywhere so things are not as obvious as you suggest. Anyway, this small program can help you reverse enginering the transformers design which can help you along with your question.
Only trying to help, you didn't tell me what you want to use it for. Many people 'abuse' power transformers for other purposes, for instance as chokes or to drive esl panels.
Also, by answering your question in a more general context the answer might be usefull for others as well. After all, this is a public forum. Sorry to offend you
Oops... pulled the wrong link from my bookmarks, I intended magdemo. No wonder you got lost. Sorry...
The OPT program I accidently linked to before is intended to design output trannies, Rp being plate resistance from the output tubes which should be equal or < to Zprim at lowest feq of interest to prevent the output power being wasted in the primary inductance. Anyway, not very helpfull for calculating power transformers..
Magdemo may look complicated at first sight (well it did to me) but it comes with a pretty good quick design guide (in help function) which makes things rather easy. The good thing about magdemo is that it will generate a spice model for your design.
A nice article on transformers, if you're prepared to do some more reading on the subject, can be found on http://www.geofex.com/Article_Folders/xformer_des/xformer.htm
May I ask why you want to model your power transformer in spice? What do you want to achieve?
The OPT program I accidently linked to before is intended to design output trannies, Rp being plate resistance from the output tubes which should be equal or < to Zprim at lowest feq of interest to prevent the output power being wasted in the primary inductance. Anyway, not very helpfull for calculating power transformers..
Magdemo may look complicated at first sight (well it did to me) but it comes with a pretty good quick design guide (in help function) which makes things rather easy. The good thing about magdemo is that it will generate a spice model for your design.
A nice article on transformers, if you're prepared to do some more reading on the subject, can be found on http://www.geofex.com/Article_Folders/xformer_des/xformer.htm
May I ask why you want to model your power transformer in spice? What do you want to achieve?
Nixie said:
Ah, that makes thing a bit clearer
That's not a bad idea. But you don't need Lprim for that at all. Well, that is unless you are interested in modeling the load it puts on the mains but that seems rather pointless (you don't know the mains impedance anyway). All you need is an equivalent schematic for the transformer's secondary side. In other words: what does it look like seen from the rectifier.
To do that, take a model of a transformer and reflect everything on primary side to the secondary. At 50 or 60 Hz you can safely ignore any capacitances and leakage inductances. So all that remains is a voltage source and a series resistance (being Rsec + Rprim / N^2). Both winding resistances can be easily measured (just make sure the meter doesn't pulse the measurement current). And there you are.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.