warning: just talking about it may lead to mental constipation...
build it, test it and confirm correlation between theory and actual practice...
after a while it gets boring merely discussing, after all this is diyaudio,
where members actually build something and then share....
Math and physics explain the majority of electric facts, and this discussion may help to those who want make his own transformers, helping him to save time and money. Some grade of discussion is inavoidable. Theory is unfortunatelly, boring in some cases, but still is necessary.
you are of course correct, but be reminded that theory has been well known as early as the 30's by those who came before us...
and with today's internet, there are lots and lots of books and articles for the taking....
Sure, but I believe that Juan (Popilin) is trying to give an entire new point of of view, including some obscure elements inside the transformer: the parasitic ones, inductances and capacitances, always known, but rarely estimated. And all we know that a good transformer design implies minimizing both of them.
Can you imagine, for example, a design of a large bridge or building ignoring corrosion of iron theory?
warning: just talking about it may lead to mental constipation...
build it, test it and confirm correlation between theory and actual practice...
after a while it gets boring merely discussing, after all this is diyaudio,
where members actually build something and then share....
Here is a measurement:
http://plitron.com/wp-content/uploads/Atcl_3.pdf
In Fig.7 you have the primary inductance (measured at 25H while the standard is usually 50Hz) as function of secondary voltage. Although this is a PP OPT and the variation in mu is much more than in a SE OPT with gap you can see that L is max for 9 V rms only. 9 V rms into 5R is about 16W. That transformer is specified for 100W and power bandwidth (-3dB and so 50W) at 22.7 Hz.
In this case it is not a problem because the primary inductance is huge anyway although it is supposed that there is no unbalance so if you want to make use of it you need a bias servo as he did otherwise you need a gap and things might change....
In the SE OPT discussed here that uses normal M6 mu(dc) will be 330-350 depending on the actual gap to have 80 mA and will about 480-500 max around 0.8T. For simplicity assume 340. This gives Bdc just over 0.41T. It means that inductance will be max for Bac around 0.4T. How much power is this at 20Hz into 3K primary load? It is less 2.4W. Above this output inductance will be more or less constant for just a bit and then decrease. So the effective impedance will be lower and lower and distiortion increases.
This core is simply not suitable for a 3K SE OPT that can deliver 10W at 20Hz with 80 mA DC and low distortion while scoring reasonable numbers in the other departments as well. It's not rocket science.
Last edited:
Take plitron data sheets and other manufacturers, with lots of care. Measurements may be true, but in some particular cases, and as almost all advertisements, misleading. The van der Veen books is full of errors. I purchase the book, and as I know books mustn´t be burned, I wanted to incinerate it.
The whole premise seems odd, so let's step back a bit and see what you and/or Juan are saying - are you basically suggesting that the transformers designed with "old" knowledge do not work well? And the new equations can help to address their shortcomings? But isn't it common knowledge that well designed old transformers are the ones most sought after? And many diy'ers lament that the OPTs "are not made like they used to...", no?
Instead I will take your stament with lots of care, I am afraid!
The fact that one is not able to understand specifications doesn't mean they wrong.
There could be mistakes but I don't think there are conceptual mistakes. EVen without stating that I get the same behaviour from mines this kind of measurements have been done for at least 50-60 years or more and they are no different when pointing at substantial facts. The only difference is that materials have improved but the physics hasn't changed.
Last edited:
The whole premise seems odd, so let's step back a bit and see what you and/or Juan are saying - are you basically suggesting that the transformers designed with "old" knowledge do not work well? And the new equations can help to address their shortcomings? But isn't it common knowledge that well designed old transformers are the ones most sought after? And many diy'ers lament that the OPTs "are not made like they used to...", no?
No, but why not to wait he finishes the exposition, and the critizice if he is wrong. Don't forget since the first design of a transformer, lots of things has been occurred, including new math models and approaches, and new points of view, new materials, and so on. I had try also a new approach in toroidal transformers, search my thread if you want, using RF and switching knowledges.
Not a book, but you can google for my works published in the packet radio network about switchmode power supplies, in spanish. There are 40 small chapters dedicated to explain easy and clearly how they work, and some examples. Free, and oriented to the ham radio media in respect to their electronics knowledge. (search in google for lw1dse+smps+latnet). ASCII drawings for easy packet transmission (1200baud text only).
In respect to write a book, I have nothing new to write, but write a bad book is worse than no write nothing.
I think you are confusing opinion with facts.
The interpretation might change when the laws are not known and there can be different models. Inductance is no such thing, there is no interpretation to be made but just a proper measurement.
Then you can interpret it as you want in terms of microscopic behaviour but this will not change its macroscopic value.
I meant articles or books about audio transformers...
Even with all of these equations, I don't have a clue how to actually specify all of the parameters for an audio transformer and thus make one. Granted, I have not spent a lot of time looking into this, but the title of this thread is "design...." so I assume it's supposed to be about the actual details of how to select this particular core material, that number of windings on this or that bobbin, etc. I also don't have a clue as to what is a "z winding" or other details like where to put interwinding insulation, etc. Theory and equations are great but putting it all into practice is another thing entirely.
Isn't the presentation here a re-hash of other threads by Juan? If not, then perhaps I jump the gun a bit, your points are certainly valid rwt to new core material, special needs for SMPS designs, etc. But we are after all talking about designing/winding audio fequency transformers using well known materials, or are the cores/wires no longer available like the old days? Curious that's all...
Unfortunately, all here (in my country, Argentina) is difficult. I'm currently waiting for 8 compactron tubes purchased in january 6 from Ebay, and payed with paypal. Also waiting for the Teman book, 1 month ago. Currently I had no expectation they came to me, in an usable form (Not a glass puree) Do you imagine the time needed to write a book wasted in an bureau office? Nahhh.
I sympathize with your situation in Argentina. It is a terrible thing that is happening there. I spoke with one of your compatriots over the Christmas holidays about what's going on there, as well as reading a little in the news. So sad the situation there.
At least you do have the internet still available to you. I did not mean any slight at all regarding your posts here, just that for me, they are outside of what I can understand right now. I have a stronger interest in "how to" rather than abstract theory. I need to visualize in my own head what is going on in order to fully understand it. Thanks for your informative posts and wish you well in your home country.
OK, yes, things are too difficult here, but explain them will go to a politic thread which is prohibited here. But believe me that all is too difficult. And not only this, politics varies short time between them. Usually you don't know what to do, this includes some experts in this topic.
I'm Electronic Engineer, ham radio and Electric-Mechanic Technician, so my brain and my way of thinking is essentially mathematically, so I can't understand some politic (wo)men for which any is black now and white tomorrow. But things are this way by the moment and almost for 2 more years. Worst, I have no expectation things go better for a long time.
- Status
- Not open for further replies.