In Reply to: Re: Shorted Turns in RF Tank Circuits posted by John Dunn - Consultant on 04/24/00 at 9:41 PM:
: : John,
: : I notice your well worded and carefully reasoned posts here, and I thank you for your response.
: : I will belabor the point just a bit further, and state that if I was not unequivocal in my last post, I should have been -- the shorted turns method of adjusting tank inductance is the standard method used in RF tube type induction heaters.
: : I will venture that this method probably prevails for one main reason. Since the overwhelming majority of the tank coils in these units are constructed of copper tubing and are water cooled, the circuit is much easier and less expensive to design and construct this way, considering that issues of cooling water and high voltages are involved. I will guess that if the coil is designed with many turns, and high currents are only run through a few of them, then maybe the coil will in fact act as an autotransformer, developing some alarmingly high voltages out at the far, unconnected end (which must be connected to a water hose). This would necessitate the addition of an extra foot per thousand volts of cooling hose, designed for the worst case scenario, plus insulation of the whole apparatus for much higher voltages than otherwise required.
: : It is not unusual to operate one of these units with one or two out of 10 to 15 total turns connected. Even under these conditions, a peak voltage in excess of 9,000 volts is not unusual at the output of the unit (one end of the tank coil is the output). The more I ponder it, the more it appears that unmanageable voltages could result from leaving one end unconnected. A pseudo Tesla coil, if you will. Arcing and insulation failure are always problems with these units, and the design principles have evolved over many years, with a good number of companies applying their best efforts at improvements.
: : It seems highly likely that economic considerations have made the shorting method the historical choice. These units are only 50% efficient on a good day, and the losses in the shorted turns are probably negligible considering all of the other design trade - offs that have to be made in order to design and produce one that keeps working for any reasonable length of time.
: : Thank you again for your response,
: : John Danforth
: Hi, John.
: You make a very ggod point indeed! An autotransformer effect on open ended turns would be a serious mischief maker indeed.
: I guess what it comes down to is that if the resistive losses in the main part of the coil, if I can call it that, are kept low, the same techniques of construction would keep them low in the shorted section too. Then, when the shorted section is reflected into the main section, you simply get a new inductance value against which the input and output capacitors of the network can be chosen.
: This has been an education! Thank you.
: Now to get in touch with Mike. If he hasn't read all of this, I'm sure he'll find it interesting.
: Good luck.
: John Dunn - President
: Ambertec, Inc.
WOW! This certainly has been an education. Until now I haven't seen many logical or credible reasons for the practice. The bottom line is, it is a trade-off -- i.e., the minor losses that result from shorting turns out-weigh the greater problems that can occur with NOT doing it! I thank you all who have contributed to this perplexing question.