Jim,
That's exactly the way the tech at Palomar Engineers explained it to me years
ago. The problem was that 43 material got too hot to the point of ruin. The
only help for that would be use a good amount more of toroids in the stack
(dropping the flux density) than what was shown in some of the amps shown by
the Motorola app notes. 61 material, even though having a lower permeability,
didn't heat up this way. Helge selected 43 because of its higher permeability,
and even used 77 in some places for this reason. Neither one is really rated to
operate at 2-30 MHz using the 10X upper frequency spec for broadband
transformers. Palomar states that the upper frequency can be increased X10 for
use in broadband transformers, and that the published ratings are for
transformers in narrowband circuits (IF, etc). The lower frequency limit counts
more as to not go below it. But like you say, even 61 material, over a certain
frequency, the losses really start going up which does in any transfor
mer material at a certain point (permeability starts falling off rapidly, and
magnetizing current raises more expotentially until saturation). I brought this
up as I wanted to find out more about it before writing some new transformer
software.
Best,
Will
*********** REPLY SEPARATOR ***********
On 2/19/06 at 11:31 AM Jim Brown wrote:
>I'm a bit late to the table on this issue, because I've been on the road
>almost
>continuously for three weeks. But I have some comments that might shed
>some light.
>
>First, I agree with John's comments re: misuse of the word "saturation."
>Second, I
>strongly suggest that anyone interested in the use of toroids for any
>purpose go to the
>source, www.fair-rite.com and download their pdf catalog. These are the
>people who MAKE
>the ferrite materials and form them into the parts that people like Amidon
>and Palomar
>re-sell. The curves in that catalog are very useful. They show the series
>equivalent
>values for complex permeability for all of the materials they make.
>
>Third, I strongly suggest that anyone interested in understanding ferrites
>download the
>tutorial on my website, which includes a lot of measured data and a new
>equivalent
>circuit for a ferrite coil. This equivalent circuit should go a long way
>towards helping
>us understand how ferrites work, which parts should be used, how to use
>them in
>circuits, and how to analyze their performance.
>
>http://audiosystemsgroup.com/SAC0305Ferrites.pdf
>
>In short, the loss component seen by any coil wound around a ferrite part
>can be quite
>significant, and becomes dominant with increasing frequency. This is true
>whether the
>coil is part of a transformer, a balun, or a "choke" used for suppression
>or decoupling.
>If the core is handling significant power, this loss component must be
>taken into
>account, and can often be the most important design parameter. For this
>reason, the
>Fair-Rite #61 material is the most practical one for high power
>applications at HF and
>above. And even this material begins to show significant loss above about
>20 MHz.
>
>What defines high power? Any application that causes significant heating
>of the core,
>which in turn reduces permeability, burns transmitter power, and can cause
>overheating
>of the wire (or coax)!
>
>One place where I disagree with John's post is his advice with respect to
>use of the
>core in a transformer. The loss component IS quite important IF the
>transformer, and
>thus the core, is carrying significant power. The transfer of energy from
>one winding of
>a transformer to another is through the core, so any power being
>transferred WILL heat
>that core and power will be dissipated. One watt in the component as large
>as the one
>that Amidon and Palomar call an FT240-61 or FT240-43 is no big deal, but
>100 watts can
>be, depending on the frequency. If you doubt this, wind 6-8 turns around
>both components
>and transmit through them at HF. The FT240-43 will get HOT. The FT240-61
>won't.
>
>Jim Brown K9YC
>
>, On Thu, 02 Feb 2006 11:54:21 -0500, John Popelish wrote:
>
>>First of all, there is some confusion using the word "saturates".
>>Magnetic saturation involves a sharp drop in permeability as the
>>magnetic flux passes through some large positive or negative value.
>>It has almost nothing to do with frequency.
>
>>The two important aspects of any ferromagnetic material for high
>>frequency use are permeability versus frequency and loss versus
>>frequency and flux swing.
>
>>If a core is use for energy storage, for instance as the inductor in a
>>filter or tank, it stores that energy by accumulating flux. So
>>significant flux swing is implied, and all the loss mechanisms that
>>take place during a flux swing cycle. But if a core is used as a
>>transformer core that only couples energy from one turn to another,
>>then there may be very little flux involved (which goes up by the
>>volts*seconds of each half cycle, so higher frequencies imply lower
>>flux swing) permeability and conductivity (that provides a one turn
>>sneak load across the transformer), are about all that matter.
>
>>That is why they say that type 61 is usable for wide band transformers
>>(turns coupling applications) up to 200 MHz, where its permeability
>>goes to crap, but is usable for high Q (energy storage, since Q is the
>>energy stored divided by the energy lost, each cycle) only up to 15
>>MHz, because its loss per cycle for large flux swings goes to crap there.
>
>>So the first stem in defining an appropriate material is to put the
>>application in the turns coupling or energy storage category.
>
>>_______________________________________________
>>Amps mailing list
>>Amps@contesting.com
>>http://lists.contesting.com/mailman/listinfo/amps
>
>
>
>_______________________________________________
>Amps mailing list
>Amps@contesting.com
>http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|