[Amps] 'red' cores versus 43 cores

[Alex Eban]

Thanks Manfred, this really sums it up very well!
73 and all the best;
Alex	4Z5KS

-----Original Message-----
From: amps-bounces at contesting.com [mailto:amps-bounces at contesting.com] On
Behalf Of Manfred Mornhinweg
Sent: Monday, March 30, 2009 4:33 AM
To: amps at contesting.com
Subject: Re: [Amps] 'red' cores versus 43 cores

Hi Dan, and all,

> Is there a straightforward practical or operational description of the 
> difference between 'red' cores and 43 material.

I will try...!

The "red" material is powdered iron with a permeability of 10, while the 
43 material is ferrite with a permeability of 850. The two behave quite 
differently, not only because of the difference in permebility. Also, 
both materials behave very differently as the frequency changes, so the 
situation is complicate, and most people don't have a good grasp of it!

Imagine that you have two toroids of the same size, say, 2cm diameter. 
Now let's compare three coils: One 15 turn coil wound on the iron 
toroid, one 15 turn coild wound on the ferrite toroid, and the third 
wound just on thin air, but with the same size and shape.

The air coil might have an inductance of roughly 1uH, and at 7MHz the Q 
factor might be on the order of 300, if the wire is thick enough. There 
is no magnetic loss in air, so the loss of this coil comes solely from 
the copper.

The coil on the iron toroid will have much higher inductance. Not 10 
times more, because some of the flux is still in the surrounding air, 
but close enough. Maybe 9 times as much. So this coil will have about 
9uH in this example. Now for the Q factor: The wire resistance is still 
the same, so the loss from the wire is also the same, but the inductance 
is 9 times higher, so if there was no core loss, the Q factor qould also 
be 9 times higher! Unfortunately there IS core loss. But the loss of 
"red" material at 7MHz is so low, that the Q of this coil will probably 
end up a little higher than that of the air core coil, perhaps in the 
neighborhood of 400.

Now let's take the coil on the ferrite toroid. Its inductance will be 
very high, let's say perhaps 700 times as high as the air coil (not 850 
times as high, because there is still some flux in the air). And the 
wire resistance will still be the same, so it will become negligible at 
this high inductance. BUT the loss in the ferrite core is NOT negligible 
at all! The ferrite has a relatively high loss, and 43 material at 7MHz 
might provide as much loss resistance as it provides inductance! So, the 
resulting coil might have an extremely low Q, as low as 1 or 2!

For this reason, type 43 material is totally unsuitable for resonant 
circuits at 7MHz, while "red" material is perfectly suitable, as are air 
coils.

You might then think that "red" material should be used in all cases, 
because it has such low loss? No, not at all!

Let's take the case of a non-resonant transformer operating at 7MHz. 
Let's say, you need a 4:1 stepdown, from 200 Ohm to 50 Ohm. Wind 15 
bifiliar turns on the ferrite toroid, connect them for 4:1, and the 
inductance across the whole thing will be a whopping 2000 to 3000uH! So 
the magnetizing current will be extremely small, the distorsion 
extremely low, etc. And what about the high loss of the ferrite? No 
problem. Even if the unloaded Q factor of this transformer may be just 1 
or 2, that means a loss resistance of the same or twice the inductive 
reactance. At 7MHz, due to the high inductance, this is in the range of 
100 kiloohms! This loss resistance, in parallel with the 200 Ohm source 
impedance, produces an extremely low and totally negligible loss. So, 
the transformer on the ferrite core will be highly efficient, so much so 
that it's almost a perfect transformer!

If instead you use the iron core, the problem will be getting enough 
inductance. Some people say you want a reactance at least 10 times the 
circuit impedance. In this case that would call for 2 kiloohm reactance, 
or about 20uH. In our example, this might require about 22 bifiliar 
turns. That's quite possible to do, but the resulting transformer will 
have twice the wire loss, while the core loss will be as negligible as 
it was for the ferrite transformer, and the big difference is that a 
transformer that requires 10% as much magnetizing current as the whole 
signal current, isn't a anything near to perfect! And if you try to 
obtain enough inductance on the iron core by simply using more turns, 
you quickly run into serious trouble from the high loss in the long, 
thin wire!

Clearly, for such a transformer the ferrite material is a better choice, 
while for a resonant circuit the ferrite material is unuseable, while 
the iron is quite good.

Now if you go up in frequency, the loss of 43 material gets so high, 
that a coil wound on it performas almost like a pure resistance! The Q 
might be 0.1 or even lower! In this case, even the high inductance 
obtained cannot save us, and the material becomes unuseable. That's why 
above perhaps 10MHz it's usually better to use 61 type ferrite instead 
of 43.

Something similar happens to the red iron cores. As you go above 10MHz 
or so, the losses rise a lot, and it's better to switch to the yellow 
material, which has just a slightly lower permeability, but much lower 
loss in the higher HF range.


The above just scratches the surface. There are many other factors. For 
example, powdered iron has a rather stable permeability, while ferrite 
does not. This is one more reason that makes ferrite unsuitable for 
resonant circuits, unless it is airgapped to stabilize the overall 
permeability. But for transformers, the permability changes don't 
matter, because anyway the inductance is way high enough.

Note that these aren't really questions of power level! Bot ferrite and 
powdered iron caqn be used at low and high power. The difference is much 
more in how the core is used. To state it very simply: If you need an 
inductive reactance that is in the same order of magnitude as the 
circuit impedances, such as in filters, tanks, resonant circuits, etc, 
use powdered iron, or air-wound coils. And if you need an inductive 
reactance that's way higher than the circuit impedances, such as in 
non-resonant transformers, use ferrite. And after this basic decision, 
compare material specs within the chosen category, to select the exact 
material. Because both ferrite and powdered iron exist in a wide range 
of different characteristics.

> What would happen if a mixer were made from a 43 core?

This is commonly done! Diode ring mixers for example are very often 
built using ferrite cores. But 43 material is a good choice only for the 
MF to lower HF range. If you want to get into higher frequencies, a 
different ferrite material is surely better. Over 20 years ago I built 
several direct conversion radios for 40 meters, using 43-type ferrite in 
the diode mixers.

> What would happen is an output transformer were made from 'red' cores?

It can be done. But if it were designed to work well (low magnetizing 
current, wide bandwidth, low total loss) it would end up much larger and 
more expensive than a roughly equivalent transformer made on a ferrite 
core. And still, the performance of the ferrite transformer would be better!

But if you want to build the low pass filter after you amp using ferrite 
cores, you are out of luck!  Unless you use such a large air gap, that 
you can hardly talk of a ferrite core anymore! If you use 43 type 
ferrite, you would need a very long and thin core to even be able to 
make a coil with the low inductance needed, and then the equivalent loss 
resistance would be about as low as the circuit impedance, resulting in 
an absolutely huge power loss, and the ferrite exploding if the power is 
high enough!

I hope this helps clear things up!

Manfred.

========================
Visit my hobby homepage!
http://ludens.cl
========================
_______________________________________________
Amps mailing list
Amps at contesting.com
http://lists.contesting.com/mailman/listinfo/amps
No virus found in this incoming message.
Checked by AVG - www.avg.com 
Version: 8.0.238 / Virus Database: 270.11.31/2029 - Release Date: 03/29/09
16:56:00