to clarify...
you are trying to double the power by doubling the core area...
there's two issues here.
one is total losses.
the other is material constraint.
stacking the core, in the example above, gets you twice the power, for 25% less copper loss, iron loss is doubled.. but it also cuts your surface area down 25% as well, so it runs a little warmer, not much, but it is significant when you are running the numbers when considering how long a stack of 'E' cores should be before you move up to the next standard size.
end result is that you doubled the iron, saved 25% on the copper and the power went up say 180% for the same temperature rise.
you are trying to double the power by doubling the core area...
there's two issues here.
one is total losses.
the other is material constraint.
stacking the core, in the example above, gets you twice the power, for 25% less copper loss, iron loss is doubled.. but it also cuts your surface area down 25% as well, so it runs a little warmer, not much, but it is significant when you are running the numbers when considering how long a stack of 'E' cores should be before you move up to the next standard size.
end result is that you doubled the iron, saved 25% on the copper and the power went up say 180% for the same temperature rise.
It doesn't change the capacity of the core. It only increases the capacity of the transformer if there is room for more wire or more loss.
AndrewT,
Exactly what don't you understand about Eckhardt's equation. When transformer designer.exe asks for number of turns, in effect, it is asking for current density (J) and winding area (W). The number of turns you can get on a transformer depends on J and W, so N is not an independent variable.
Exactly what don't you understand about Eckhardt's equation. When transformer designer.exe asks for number of turns, in effect, it is asking for current density (J) and winding area (W). The number of turns you can get on a transformer depends on J and W, so N is not an independent variable.
I seem to have a fundamental misunderstanding of transformer science, if I'm to come to accept your statement that doubling the core area gives a doubling of VA capability.
That doubling of VA capability is not what I see within a wide range of toroid and EI types.
The .exe file I referred to confirms my misguided thinking.
I just can't see how your arguments are going to convince me, yet !
Posts 20 & 23 are not helping me to unravel what's happening.
That doubling of VA capability is not what I see within a wide range of toroid and EI types.
The .exe file I referred to confirms my misguided thinking.
I just can't see how your arguments are going to convince me, yet !
Posts 20 & 23 are not helping me to unravel what's happening.
Post 20 and belief in that equation should probably unravel everything. However, including the specific copper arrangement and surface area would be required for a serious theoretical picture of power handling. Ignoring all that does allow you to determine the flux density anyway, which halves itself if you double the core cross section with input voltage, frequency, and number of turns all the same.
Dad Blast it. In a hurry to verify my post I stuffed a milliammeter in series with the primary of a very good quality 200VA toroid (these things might rise a degree with just magnetizing current) and smoked my last 1A fuse for the meter.
Zero joy.
So I get out the Kill A Watt and try that way. It only resolves down to 10mA but that's what it read and presumably it knows something about RMS.
This is on a 120V line.
So let's assume you expect something around 150VA from your transformer. On a 230 volt line, I'd say, a few milliamps. This is supposing power density and cost are not your major design drivers. If you're looking for that then a few to several watts at idle are not out of question, like johansen said.
Zero joy.
So I get out the Kill A Watt and try that way. It only resolves down to 10mA but that's what it read and presumably it knows something about RMS.
This is on a 120V line.
So let's assume you expect something around 150VA from your transformer. On a 230 volt line, I'd say, a few milliamps. This is supposing power density and cost are not your major design drivers. If you're looking for that then a few to several watts at idle are not out of question, like johansen said.
AndrewT,
To answer your last question, assuming there is no load on the secondary the magnetizing inductance will determine the current according to
I=230/(j*omega*Lm). Although the magnetizing inductance will be nonlinear it is given by: Lm=uo ur N^2 Ac/lc where lc is the length of the core. ur is the nonlinear term.
To understand where Eckhardt's equation came from google for "core area product". You should also be able to find something at Ferrite Cores, MPP Cores, Tape Wound Cores, Sendust Cores| Magnetics® or other core vendors web sites.
To answer your last question, assuming there is no load on the secondary the magnetizing inductance will determine the current according to
I=230/(j*omega*Lm). Although the magnetizing inductance will be nonlinear it is given by: Lm=uo ur N^2 Ac/lc where lc is the length of the core. ur is the nonlinear term.
To understand where Eckhardt's equation came from google for "core area product". You should also be able to find something at Ferrite Cores, MPP Cores, Tape Wound Cores, Sendust Cores| Magnetics® or other core vendors web sites.
this new iron is rather interesting stuff... the inductance is exceedingly high, until you get to 1.8 teslas, then it drops real fast..
its the same reason you have to have over-rated slow blo fuses when you plug them in, if you don't use an NTC..
for a core of 60mm by 33 mm you need about .8 volts per turn to get 1.6T.
for 1.2T flux density you need .62 volts per turn
its the same reason you have to have over-rated slow blo fuses when you plug them in, if you don't use an NTC..
for a core of 60mm by 33 mm you need about .8 volts per turn to get 1.6T.
for 1.2T flux density you need .62 volts per turn
I'm not aiming to saturate the core, so I don't want to increase the idling flux in the core.
It looks like the double core is 110mm diam * 60mm thick and needs 500T.
When I enter those two alternatives into .exe they predict:
single core 1000T, 73.84VA, 1.045 flux, 4.348 T/V
dual core 500T, 295.4VA, 1.0465 flux, 2.1739T/V
I expect that when tested at various input voltages, the primary idling current will increase fairly linearly until the core saturates and that the double core version will roughly pass double the idling Primary current cf. the single core version.
Is that realistic?
It looks like the double core is 110mm diam * 60mm thick and needs 500T.
When I enter those two alternatives into .exe they predict:
single core 1000T, 73.84VA, 1.045 flux, 4.348 T/V
dual core 500T, 295.4VA, 1.0465 flux, 2.1739T/V
I expect that when tested at various input voltages, the primary idling current will increase fairly linearly until the core saturates and that the double core version will roughly pass double the idling Primary current cf. the single core version.
Is that realistic?
Last edited:
I am afraid to disagree.
A transformer (and motors) requires a mains fuse that is ~ 3 * VA / Vac. For a 100VA on 230Vac that comes to T1.3A, use the nearest value above this.
If soft start, whether NTC or other and close rated fuse is to be used, you need ~ VA/Vac that comes to T435mA, use the nearest value and see how long it lasts.
Here's the link that recommends Sevart's Xformer.exe
http://sound.westhost.com/articles/xfmr3.htm
and here is the modeling software:
http://sound.westhost.com/articles/xfmr3.htm#s6.0
Last edited:
I forgot to short the meter until the inrush passed.
AndrewT those numbers look like some kind of inverse square is going on. No wonder we're still talking about this. The whole thing is definitely a linear relationship. Throw out that software? I don't know what's going on with that.
Edit: Either I went cross-eyed or you fixed the numbers. But I still don't see how you're getting 4 times the power with twice the core. It would be pretty interesting if that actually worked out.
AndrewT those numbers look like some kind of inverse square is going on. No wonder we're still talking about this. The whole thing is definitely a linear relationship. Throw out that software? I don't know what's going on with that.
Edit: Either I went cross-eyed or you fixed the numbers. But I still don't see how you're getting 4 times the power with twice the core. It would be pretty interesting if that actually worked out.
Last edited:
single core 100T, 73.84VA, 1.045 flux, 4.348 T/V
dual core 500T, 295.4VA, 1.0465 flux, 2.1739T/V
i think you have a typo in the turns somewhere, is that supposed to be 1000 turns?
edit: yes it is.
in any case, there's no reason not to go as high as you can get on the flux.
the core loss vs flux looks more like a hockey stick graph, not an exponential β^1.6 like the old books say.
the inrush current for toroidal transformers can reach 100 times the normal full load current due to the core saturating during turn on.
this has been discussed before in other threads. the cause is both the lack of an air gap and the particularly high permeability of the core, as well as the fact that at turn off it holds its reminiscent field quite nicely, it don't happen every time, as its particularly bad when you just happen to plug it in at the zero crossing IIRC
http://johansense.com/bulk/spreadsheets/
these spreadsheets are continually updated, i wrote the transformer calculator 2 years ago, its not complete by any means.
dual core 500T, 295.4VA, 1.0465 flux, 2.1739T/V
i think you have a typo in the turns somewhere, is that supposed to be 1000 turns?
edit: yes it is.
in any case, there's no reason not to go as high as you can get on the flux.
the core loss vs flux looks more like a hockey stick graph, not an exponential β^1.6 like the old books say.
the inrush current for toroidal transformers can reach 100 times the normal full load current due to the core saturating during turn on.
this has been discussed before in other threads. the cause is both the lack of an air gap and the particularly high permeability of the core, as well as the fact that at turn off it holds its reminiscent field quite nicely, it don't happen every time, as its particularly bad when you just happen to plug it in at the zero crossing IIRC
http://johansense.com/bulk/spreadsheets/
these spreadsheets are continually updated, i wrote the transformer calculator 2 years ago, its not complete by any means.
Last edited:
I never rely on ammeter function. It's too vulnerable to mistakes and is rarely if ever as accurate as voltmeter.
I always use voltmeter function and add in a measuring resistor to allow currents to be determined.
5mA through 1r0 gives a reading of 5.0+-0.1mVac
that 1r0 600mW resistor can pass a short term current of ~780mA and will probably survive 1.6Apk as a starting transient.
A parallel/series combination of 4* 1r0 could probably pass a transient of ~3.1Apk at start up.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.