Hi Again,
Anybody care to comment about the Impedance matching of the particular circuit.
I need to make this but i would like to get a better understanding of it before just jumping in head first & just making it without a full understanding of it.
I have built the driver for it already & have purchased some components but i am missing the knowledge to understand it without help.
I realise that some will not comment on things like this for obvious reasons & the dangers involved, i am a newby to electronics only.
I need to build this to heat treat components that i make for Automotive purposes.
Thanks
Hello,
The impedance matching aspect of the transformer is basically to be
able to drive the output coil resistance with the maximum current that
is possible. Without the transformer the coil is subject to be driven
by the series impedance of the transistors and the 1uf capacitors,
and even the power supply,
which can be high relative to the resistance of the coil. By using
a step down transformer between driver and load we effectively
lower the output impedance of the driver by 400 (20 to 1 transformer)
and so are better able to drive the coil. The best match is obtained
when the transformer presents the same impedance as the
raw output of the transistors, to the transistors, with the true load
connected, and this will be 400 times higher than the actual load
impedance because the impedance reflected to the primary is
as the turns ratio in that direction squared (20 squared = 400).
The other view is that we want to load the power supply as much
as possible, but not overload it, and this can only happen when the
load on the very output of the transistors is a certain value, and this
only happens when the transformer ratio is such that the reflected
impedance is that certain value.
There are practical considerations that make this a little harder to
do then simply sit down and do a few calculations. This is partly
due to the proposed frequency of operation (125kHz) and also because
there are component imperfections abounding. The frequency
of operation here means that skin effect is going to have a
significant impact on the actual component parameters.
The component imperfections include this skin effect as well as
inherent deficiencies like ESR and leakage inductance.
What this means is getting the right transformer ratio is going to
take some experimentation, unless suboptimal performance can be
tolerated. This also means it would be best to 'tune' the frequency
in using a current meter to measure dc current from the power
supply, or possibly just looking for a dip in the power supply voltage.
As the frequency is tuned, the power supply current increases
as resonance is obtained.
To tune the transformer, it would be necessary to try different
ratios to see which one caused the maximum power in the coil.
This may have already been done and 20 seemed like a good
ratio which causes enough heating so that it doesnt have
to be changed. If not, or if your setup is unique, changing
the turns ratio could result in more heating.
I would try it with 20, and also tune the frequency and
see if it flies like that first.
Simple Example...
We have a 10v power supply that can put out 10 amps max.
To get this 10 amps max, we need a load of 1 ohm because
anything larger wont draw the full power and anything smaller
will draw too much current.
Our load, unfortunately, is 100 times smaller (0.01 ohms),
so if we connect this load to the output we will attempt
to draw 1000 amps, which obviously will blow the fuse.
If we use a series resistor of 0.99 ohms, we get the
right current (10 amps) but now we loose much of the
power in the 0.99 ohm resistor, so instead we use a
matching transformer. Since we want to step 'up' the
load from 0.01 to 1, this is 100 times, so we need a
turns ratio of 10 to 1 (because 10 squared is 100).
This transformer will allow the 10 amps to be drawn from
the power supply and the load will get the max power
available without damaging the power supply.