Thanks for that.
I see that you are describing what can be seen when one plots the Iprimary vs Vprimary curve.
The S shape that results shows the steepest I/V when the tangent to the S curve is used as the Turns per volt.
I have used that same S curve to determine a much higher flux that still stays away from the saturation region. This nearly always requires some extra primary turns for our typical UK voltages. I have in the past recommended adding 20T to the primary. I find that few extra turns makes a significant difference to the core losses.
But it never gets down to that target shown with the tangent to the S curve you have described. I suspect using the highest impedance would require 50% extra primary turns and result in 2/3 of the output voltage.
Hi Andrew,
First of all thanks for having a fruitful discussion now.
Your interpretation of what i described is correct. Offcourse in any transformer design you have to set the priorities first and then adjust the unavoidable compromises accordingly. Given the priorities of a manufacturer it is for us DIY-folks, in a mains transformer that we use in a audio project, usually beneficial to increase the primary turns. By how much, 20T or whatever, depends on our individual priorities and the reqirements set by the load and the actual tranformer at hand. Most propably, despite the added turns, you will still be in the region nearer to saturation than to maximum inductance. To add primary windings to the point to get the highest impedance would seldom be practically possible because it would in most cases require even much more than 50% extra turns with the result of accordingly lower output voltage, power a.s.o.
Never the less, if size, weigth and purchasing costs does not restrict, even in a mains transformer with M6 working at 0.8T has his merits as long as you can cope with the downside effects in a non degrading way.
In a OPT with M6 iron 0.8T at the lowest frequency givs the highest impedance at the frequency where it is most needed, but it is good to remember also, that 0.8T is only the steady state value. Music is not steady state so to prevent excessive distortion at the very start of a low frequency full power event it would not be a entirely bad idea to use even lower flux densities.
Then again, it will bite you at the other end of the passband....
What is the best compromise depends on so many things, source impedance, feedback a.s.o.
I think you know all that, may be others dont, so i just putted here what came into my mind of possible being useful...
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.