Bmax of an iron powder cored transformer ?

[Herschel]

So I'm about to wind a high frequency torroid trasformer for a high power half bridge SMPS powersupply
And for the transformer I'm using two sendust ferrite cores placed an top of each other for more power
 
so while doing the calculations for the transformer, I noticed the Bmax value. How does this affect overall performance of the transformer?
does high Bmax value untill a limit means more efficient? after calculation Bmax value of my transformer is around 650. is this ok for Sendust or Iron powdered cores?

Also for sendust torroid cores, what is the proper range of Bmax value  (like 1200 to 2000 for ferrite cores)

while doing the calculation, the Ae value (cross section area) for one torroid core is 1.990. As I'm using two of them placed on top of each other, I took Ae*2 (3.98)

[TimFox]

The total flux BA in the core is Bx(area).  In SI units, Tesla x m² = Volts x sec.
The emf in one turn of a coil = d(BA)/dt = (2pi)f x B_max x A.
The B_max and area limit the sinusoidal voltage on the entire coil at the operating frequency.

[Xena E]

Over specifying a transformer core to allow a higher saturation threshold will also introduce greater losses.

[Kleinstein]

Sendust is a kind if iron power. Sendust and ferrite are different materials.
Bmax can be relevant with relative low operating frequency (especially mains frequency transformers, but can also be lower frequency SMPS). It there sets the maximum voltage per turn. A higher Bmax would allow fewer truns and thus less restance and loss from that side. For the same number of turns on the core the Bmax does not change anything on the efficienty. It could still help at start up where longer pulese may drive an core to saturation.

For higher operating frequency the useful field strength is often limited by the losses. So the datasheet give the loss for 1000 Gauß as a operating point, way below Bmax. So at the higher frequencies Bmax (of the material) is often no longer relevant as one would stay well below that limit anyway.

How much actual magnetization to use can effect the efficiency, though it is not an easy relation. Using lower Bmax may need more copper and maybe a larger core. A larger core can be be lower or higher efficiency depending on the details, but is usually more expensive. The difference in the efficiency are usually not very large.

[Benta]

Weird data sheet. Who on earth uses gauss and oersteds today?
Tesla and A/m would be appropriate for the 21st century.

[coppercone2]

they are finer units. Tesla makes sense for giant electromagnets

[TimFox]

One can always use prefixes such as milli and micro before large units such as Tesla.
Note that converting between Tesla and A/m (for B and H) to Gauss and Oersted is not merely powers of ten.
In Gaussian cgs units, in vacuum, 1 G = 1 Oe.
See any elementary textbook for the details.

[Benta]

they are finer units. Tesla makes sense for giant electromagnets

The nice thing about the decimal system is that it scales both up and down.
You know, giga, mega, kilo, milli, micro, nano, pico to mention a few that EEs use.
BTW, a normal field in a ferrite transformer would be in the 100s of mT, so it's not just for "giant electromagnets".

[TimFox]

> 1 T electromagnets are common in medical MRI machines, but relatively rare elsewhere.
For B, 1 T = 10⁴ Gauss, but the definitions of H differ for rationalized MKS (SI) and unrationalized CGS (Gaussian) systems.

[Herschel]

Weird data sheet. Who on earth uses gauss and oersteds today?
Tesla and A/m would be appropriate for the 21st century.

these units are still used in my college text books even though they say the syllabus is revised for 2024 to 2028

these units are finer units. And its better to write small values in gauss instead of tesla where there would be a 10^x which makes the answers more confusing

[TimFox]

Physicists often prefer Gaussian unrationalized cgs units (Gauss and Oersted) since the equations for many electromagnetic purposes are simpler.
Engineers usually prefer rationalized mks units (Tesla and A/m) since they relate better to Volts and Amperes in electronic circuits.
When people mix the units, e.g. using Gauss because they are appropriate to the problem's scale with Volts and Amperes, they must be careful since the equations differ.
I recommend sticking with mks and using prefixes ahead of Tesla to deal with flux density below 1 kGauss.
If you don't understand the difference between unrationalized and rationalized units, consult an appropriate elementary textbook.

[Herschel]

Anyway I decided to go with two torroid cores sticked together for more cross sectional area, maybe it would reduce the core losses, not sure about that. But I could end up winding less no of turns(25-30 turns)  in primary...

the first time when I made this smps, the secondary transformer winding in the schematic was too high for my required volts (the transformer was made for 60volts instead of 45volts), So the sg3525 used to skip pulses and I was too late to notice that And to reduce the switching noise, I connected 332J capacitors across the gate and source of the mosfet which did nothing other than destroyed the mosfet under a nice load

also the transformer was creating too much noise under loads, no matter how strong i fasten the EE cores, it still makes the core tapping noise,
So i decided to switch to torroid cores and this is the only core they had(local store) with a datasheet or some identification code

 Also I'm too lazy to do calculations. So I use different calculation apps for faster results. not sure about the calculations too

[langwadt]

One can always use prefixes such as milli and micro before large units such as Tesla.
Note that converting between Tesla and A/m (for B and H) to Gauss and Oersted is not merely powers of ten.
In Gaussian cgs units, in vacuum, 1 G = 1 Oe.
See any elementary textbook for the details.

yeh, Farad, Henry, Bell in also to big for most use without a prefix