I've just had a idea. Instead of tweaking the drive frequency to a Baxandall inverter, one could tweak the resonant frequency by switching in extra capacitance with a PhotoFet. https://www.mouser.com/datasheet/2/149/h11f1m-185284.pdf Sadly. even the HF11 can only take 30V, so in Timo's inverter you'd need to add a extra 11 turn winding to his transformer and put the tuning capacitors, each in series with a PhotoFet, across that. As usual, you'd a series of progressively smaller capacitors each a bit more than half the value of the next one up. As before, you start the inverter in a state where the resonant frequency is higher than the drive frequency and look at shape of the voltage waveform across the inductor to work out how much higher , and switch in tuning capacitors to get to the point where it is only just higher, rather than fiddling with the drive frequency. I doubt if it is a practical approach, but it is sort of feasible. -- Bill Sloman, Sydney
Tuning a Baxandall inverter.
Started by ●July 6, 2021
Reply by ●July 6, 20212021-07-06
On Monday, July 5, 2021 at 8:20:52 PM UTC-7, bill....@ieee.org wrote:> I've just had a idea. Instead of tweaking the drive frequency to a Baxandall inverter, one could tweak the resonant frequency by switching in extra capacitance with a PhotoFet.Switching isn't a great idea; unless you do it zero-voltage, the capacitor has too much clock feedthrough. How about using a little DC to start saturating an inductor? Or just an inductive weak-link? That's a switch that can be gradual. Also not a great idea for power, but the other idea is to servo a ferrite slug in a coil, which is fairly easy at a few hundred kHz. A suitable motor/advance mechanism can be found in any old floppy drive...
Reply by ●July 6, 20212021-07-06
On Tuesday, July 6, 2021 at 2:47:06 PM UTC+10, whit3rd wrote:> On Monday, July 5, 2021 at 8:20:52 PM UTC-7, bill....@ieee.org wrote: > > I've just had a idea. Instead of tweaking the drive frequency to a Baxandall inverter, one could tweak the resonant frequency by switching in extra capacitance with a PhotoFet. > Switching isn't a great idea; unless you do it zero-voltage, the capacitor has > too much clock feedthrough.I was thinking of leaving the capacitor switched in or switched out for lots of cycles, not switching it during each cycle.> How about using a little DC to start saturating an inductor? Or just an inductive > weak-link? That's a switch that can be gradual.An inductor which was partially saturated would get more and less saturated as you went through the sinusoidal cycle. This doesn't strike me as a good idea.> Also not a great idea for power, but the other idea is to servo a ferrite slug in a coil, which is fairly easy at a few hundred kHz. A suitable motor/advance mechanism can be found in any old floppy drive...There are gapped pot cores with a centre hole, for which you can buy a turning slug that can be screwed up or down the centre-hole to adjust the magnetic path length and thus the inductance. Too many moving parts ... -- Bill Sloman, Sydney
Reply by ●July 6, 20212021-07-06
On 06/07/2021 6:23 am, Anthony William Sloman wrote:> > An inductor which was partially saturated would get more and less saturated as you went through the sinusoidal cycle. This doesn't strike me as a good idea. >The DC contol current can be arranged to be much greater than the AC component so the distortion is not so much. Or the core saturation control flux can be orthogonal to the AC winding as in parametric transformer. piglet
Reply by ●July 6, 20212021-07-06
tirsdag den 6. juli 2021 kl. 05.20.52 UTC+2 skrev bill....@ieee.org:> I've just had a idea. Instead of tweaking the drive frequency to a Baxandall inverter, one could tweak the resonant frequency by switching in extra capacitance with a PhotoFet. > > https://www.mouser.com/datasheet/2/149/h11f1m-185284.pdf > > Sadly. even the HF11 can only take 30V, so in Timo's inverter you'd need to add a extra 11 turn winding to his transformer and put the tuning capacitors, each in series with a PhotoFet, across that. As usual, you'd a series of progressively smaller capacitors each a bit more than half the value of the next one up. > > As before, you start the inverter in a state where the resonant frequency is higher than the drive frequency and look at shape of the voltage waveform across the inductor to work out how much higher , and switch in tuning capacitors to get to the point where it is only just higher, rather than fiddling with the drive frequency. > > I doubt if it is a practical approach, but it is sort of feasible.Rube Goldberg would be proud, changing the frequency is trivial
Reply by ●July 6, 20212021-07-06
On Tue, 6 Jul 2021 14:57:22 -0700 (PDT), Lasse Langwadt Christensen <langwadt@fonz.dk> wrote:>tirsdag den 6. juli 2021 kl. 05.20.52 UTC+2 skrev bill....@ieee.org: >> I've just had a idea. Instead of tweaking the drive frequency to a Baxandall inverter, one could tweak the resonant frequency by switching in extra capacitance with a PhotoFet. >> >> https://www.mouser.com/datasheet/2/149/h11f1m-185284.pdf >> >> Sadly. even the HF11 can only take 30V, so in Timo's inverter you'd need to add a extra 11 turn winding to his transformer and put the tuning capacitors, each in series with a PhotoFet, across that. As usual, you'd a series of progressively smaller capacitors each a bit more than half the value of the next one up. >> >> As before, you start the inverter in a state where the resonant frequency is higher than the drive frequency and look at shape of the voltage waveform across the inductor to work out how much higher , and switch in tuning capacitors to get to the point where it is only just higher, rather than fiddling with the drive frequency. >> >> I doubt if it is a practical approach, but it is sort of feasible. > >Rube Goldberg would be proud, changing the frequency is trivialSwitching tank caps with 200 ohm photofets should be interesting.
Reply by ●July 6, 20212021-07-06
On 7/6/2021 12:23 AM, Anthony William Sloman wrote:> On Tuesday, July 6, 2021 at 2:47:06 PM UTC+10, whit3rd wrote: >> On Monday, July 5, 2021 at 8:20:52 PM UTC-7, bill....@ieee.org wrote: >>> I've just had a idea. Instead of tweaking the drive frequency to a Baxandall inverter, one could tweak the resonant frequency by switching in extra capacitance with a PhotoFet. >> Switching isn't a great idea; unless you do it zero-voltage, the capacitor has >> too much clock feedthrough. > I was thinking of leaving the capacitor switched in or switched out for lots of cycles, not switching it during each cycle. > >> How about using a little DC to start saturating an inductor? Or just an inductive >> weak-link? That's a switch that can be gradual. > An inductor which was partially saturated would get more and less saturated as you went through the sinusoidal cycle. This doesn't strike me as a good idea. Take some of the output, invert it and add it to the DC saturating current to keep the total flux the same. Just a WAG idea. Mikek -- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus
Reply by ●July 6, 20212021-07-06
On Wednesday, July 7, 2021 at 7:57:26 AM UTC+10, lang...@fonz.dk wrote:> tirsdag den 6. juli 2021 kl. 05.20.52 UTC+2 skrev bill....@ieee.org: > > I've just had a idea. Instead of tweaking the drive frequency to a Baxandall inverter, one could tweak the resonant frequency by switching in extra capacitance with a PhotoFet. > > > > https://www.mouser.com/datasheet/2/149/h11f1m-185284.pdf > > > > Sadly. even the HF11 can only take 30V, so in Timo's inverter you'd need to add a extra 11 turn winding to his transformer and put the tuning capacitors, each in series with a PhotoFet, across that. As usual, you'd a series of progressively smaller capacitors each a bit more than half the value of the next one up. > > > > As before, you start the inverter in a state where the resonant frequency is higher than the drive frequency and look at shape of the voltage waveform across the inductor to work out how much higher , and switch in tuning capacitors to get to the point where it is only just higher, rather than fiddling with the drive frequency. > > > > I doubt if it is a practical approach, but it is sort of feasible. > > Rube Goldberg would be proud, changing the frequency is trivial.Changing the drive frequency is trivial. If you don't want to change it, changing the resonant frequency of the tank circuit could be helpful. -- Bill Sloman, Sydney
Reply by ●July 6, 20212021-07-06
On Wednesday, July 7, 2021 at 8:07:00 AM UTC+10, John Larkin wrote:> On Tue, 6 Jul 2021 14:57:22 -0700 (PDT), Lasse Langwadt Christensen > <lang...@fonz.dk> wrote: > > >tirsdag den 6. juli 2021 kl. 05.20.52 UTC+2 skrev bill....@ieee.org: > >> I've just had a idea. Instead of tweaking the drive frequency to a Baxandall inverter, one could tweak the resonant frequency by switching in extra capacitance with a PhotoFet. > >> > >> https://www.mouser.com/datasheet/2/149/h11f1m-185284.pdf > >> > >> Sadly. even the HF11 can only take 30V, so in Timo's inverter you'd need to add a extra 11 turn winding to his transformer and put the tuning capacitors, each in series with a PhotoFet, across that. As usual, you'd a series of progressively smaller capacitors each a bit more than half the value of the next one up. > >> > >> As before, you start the inverter in a state where the resonant frequency is higher than the drive frequency and look at shape of the voltage waveform across the inductor to work out how much higher , and switch in tuning capacitors to get to the point where it is only just higher, rather than fiddling with the drive frequency. > >> > >> I doubt if it is a practical approach, but it is sort of feasible. > > > >Rube Goldberg would be proud, changing the frequency is trivial. > > Switching tank caps with 200 ohm photofets should be interesting.You'd turn them or or off when there was zero volts across the resonant tank. These are going to be smaller capacitors than the one that would always be connected. There's a peak current of about 150mA through the tank capacitor in Timo's inverter, which is a lot more than you'd want to put through a photofet. The trimming capacitors you'd switch in and out would be smaller. It might make sense to throw in a select-on-test capacitor to compensate for the production tolerances on the inductor and the main capacitor, to allow you to get away with appreciably smaller trimming capacitors. That's the kind of design decision that people who actually design their circuits make. -- Bill Sloman, Sydney
Reply by ●August 1, 20212021-08-01
whit3rd wrote:> How about using a little DC to start saturating an inductor?Hey, that's my line! ;-)> Also not a great idea for power, but the other idea is to servo a ferrite slug in a > coil, which is fairly easy at a few hundred kHz. A suitable motor/advance mechanism can > be found in any old floppy drive...If a mechanical solution is not a problem, then there was that airborne military transceiver that used a motor for winding a coil with silver tape as a way for tuning the LC tank. Something along these lines, but not the exact model I have in mind: http://www.lb3hc.net/archives/2376 More inspiration: https://g3ynh.info/comps/Vari_L.html Best regards, Piotr
