FIGUERA'S AETHER MAGNETIC FIELDS LINEAR PUMP, REVIVED

[lfarrand]

I've reviewed your design for the MBB scheme and it shouldn't be a problem with the AC MOSFET switches.

I guess you will only have to deal with inductive kickback when switching off the device, since the magnetic field won't collapse if the next switch is closed before the previous switch is opened. You should be able to deal with this easily by placing two diodes across the coil, going into a capacitor (if you want to recover the energy). That said, since the inductive kickback won't be something that is happening frequently, I'd recommend just snubbing it out using a RC snubber. You're unlikely to gain much, you might light an LED for a second if you're lucky!

[Ufopolitics]

Hello All,
Great @lfarrand, that overlapping is possible on this method!

Ok, I did some fast research about your great idea of using BIDIRECTIONAL POWER SWITCH(ES), based on two NFETS back to back...

Then I have put together this simple Diagram (that I got from the web, and I installed both Cycles on the same page): (Got info from: Bidirectional Switch website )

On FIG 1, A is Positive and B is Negative, and whenever the Common Gate turns ON, the Flow (from positive to negative) will take place, where NFET on LEFT will be just conducting (near to D1) as on RIGHT NFET the Flow will go through the Diode (D2)

On FIG 2, A is Negative and B is Positive, and whenever the Common Gate turns ON, the Flow (from positive to negative) will take place, where NFET on LEFT the Flow will go through the Diode (D1) and as on RIGHT NFET the Flow will be just conducting through FET body (near to D2).

However, when we are applying this Bidirectional Switching Circuit to OUR SETUP, is  where I get a little bit 'lost'...So, this is - so far- what I get to understand:

When GATE ON, and A POSITIVE the Flow will travel through, from A to  B (Positive Flow) As shown on FIG 1 above.
When GATE OFF, and A STILL Positive, the Bidirectional Power Switch BLOCKS ALL BIDIRECTIONAL FLOW WHEN GATE IS OFF) (B becoming Positive) REVERSED VOLTAGE from Coil), while Gate is OFF, D2 will oppose .

In my understanding, and applying the BPS (Bidirectional Power Switch) to our setup, we will need ONE BIDIRECTIONAL SWITCH PER EACH SEQUENTIAL COIL...On the case of Kampen setup, He has 13 Sequential Coils Total, where Seven Terminals are ALWAYS NEGATIVE, and Seven (7) are ALWAYS POSITIVE Switching, as each MATING TERMINAL PAIR EG:  (#1+/#1-) MUST ALWAYS TURN AT UNISON.

Then we will need 14 of these BPS, ONE BPS per each terminal...where Seven (7) will be conducting Positive Flow ALWAYS to Coils ( 1 to 7 Positive Block) and Seven (7) would be conducting Negative  ALWAYS ( 1 to 7 Negative Block).

Then basically we will be driving each Sequential Coil, by Turning the BPS Gate ON & OFF.

As we could use the same arrangement shown above (FIG 1 & FIG 2) for Positive Block and Negative Blocks?...

I am seeing it very clear, when it comes to A being Always Positive (FIG 1) and when GATE OFF, still the Right Diode (D2) will OPPOSE/Block Reversed Voltage (B Becoming Positive by Coil V Reversal).

And when looking at FIG 2, where A being Always Negative (for Negative Sequential Coils Block), there, when GATE IS OFF, I see Diode (D2) NOT OPPOSING to B becoming Negative, HOWEVER, I see left Diode (D1) opposing to the Reversed Negative Flow...

Please feel free to correct me if my understanding explained above is somewhere wrong...

Anyways, I see this arrangement as a Brilliant, and very Elegant Approach...as very SOLID as well.

And I want to thank Member @lfarrand again for sharing this Method with us!

Regards

Ufopolitics

[Ufopolitics]

[...]

I've built my own MOSFET driver boards - one for single MOSFET and another with two MOSFETs for AC switching. The board uses an isolated DC/DC converter and isolated gate driver, thus removing the need to use optical tranceivers. You simply provide the PWM input via the onboard BNC connector. The PWM signal can come from a signal generator, or more commonly a microcontroller such as an Arduino. I use a Teensy 4.1 since it has quite a nice PWM block onboard and runs at 600MHz.

[...]

Best regards,

Lee

Hello again,

As for the Driver Board, originally we discussed all possibilities, however, both of us (Member Kampen and me) liked the idea of NOT having a Programmed Microprocessor like Arduino or others, for such simple algorithm's which is based on simple sequencing...or switching from 1 to N then driving back from N to 1...and repeat and repeat.
So Kampen designed a beautiful Logic Board, where we could regulate a few parameters like frequency, as on the later one (overlapped version) we could also regulate the % of overlapping from zero to a max of 20%...

Now with your Bidirectional Power Switch, this Logic Board will be even more simple to design and build than both previously designed versions...as we no longer need to drive PFET GATES...only NFET's Gates.

I want this Generator to be a simple 'Plug and Play' Device, after all parameter regulations have been tested for best performance and left untouched.

Of course, we will still need to work on an Automatic Regulation Control, which rises Frequency according to Load applied...same way Generators have a governor rod to adjust the gas feeding to carburator or fuel injection throttles, whenever load is applied and RPM's drop or Frequency, in our case.
So that our 'governor' (Automatic Control) will always keep the frequency at the same operating level whenever generator is loaded.

Thanks again

Regards

Ufopolitics

[lfarrand]

I have a slight correction to your diagram with the red arrows. The body diodes should not conduct at any point. Current will only ever flow through the channel.

Here's a longer AI answer:

In the described configuration—two N-channel MOSFETs connected back-to-back with their sources tied together (drains as the outer terminals)—this setup is commonly used as a bidirectional switch for AC or reversible DC applications. The body diodes are inherently present in parallel with each MOSFET channel, oriented to allow current from source to drain when forward-biased.

When the MOSFETs are off (gates low), the body diodes are effectively oriented in opposing directions across the overall path. For either polarity of applied voltage:

• One diode will be forward-biased.
• The other will be reverse-biased.

This prevents current flow through the diodes in either direction, as there's always a blocking diode. The switch thus blocks bidirectionally without diode conduction.
When the MOSFETs are on (gates high, assuming sufficient Vgs), the channels provide a low-resistance bidirectional path for current (since MOSFET channels conduct symmetrically). The voltage drop across each MOSFET (Vds) is typically much less than the diode forward voltage (~0.7V), so the body diodes remain reverse-biased or unbiased and do not conduct. Current flows through the channels instead.

In summary, under normal operation for an AC circuit, the body diodes would not conduct at any point. They are present but do not carry current due to the opposing configuration and channel dominance when on. If unusual conditions like excessive voltage drops or faults occur, diode conduction could theoretically happen, but that's outside standard use.

[lfarrand]

[...]

I've built my own MOSFET driver boards - one for single MOSFET and another with two MOSFETs for AC switching. The board uses an isolated DC/DC converter and isolated gate driver, thus removing the need to use optical tranceivers. You simply provide the PWM input via the onboard BNC connector. The PWM signal can come from a signal generator, or more commonly a microcontroller such as an Arduino. I use a Teensy 4.1 since it has quite a nice PWM block onboard and runs at 600MHz.

[...]

Best regards,

Lee

Hello again,

As for the Driver Board, originally we discussed all possibilities, however, both of us (Member Kampen and me) liked the idea of NOT having a Programmed Microprocessor like Arduino or others, for such simple algorithm's which is based on simple sequencing...or switching from 1 to N then driving back from N to 1...and repeat and repeat.
So Kampen designed a beautiful Logic Board, where we could regulate a few parameters like frequency, as on the later one (overlapped version) we could also regulate the % of overlapping from zero to a max of 20%...

Now with your Bidirectional Power Switch, this Logic Board will be even more simple to design and build than both previously designed versions...as we no longer need to drive PFET GATES...only NFET's Gates.

I want this Generator to be a simple 'Plug and Play' Device, after all parameter regulations have been tested for best performance and left untouched.

Of course, we will still need to work on an Automatic Regulation Control, which rises Frequency according to Load applied...same way Generators have a governor rod to adjust the gas feeding to carburator or fuel injection throttles, whenever load is applied and RPM's drop or Frequency, in our case.
So that our 'governor' (Automatic Control) will always keep the frequency at the same operating level whenever generator is loaded.

Thanks again

Regards

Ufopolitics

I agree with your approach. Avoiding a MCU makes sense if the control logic is going to be simple. The best part is no part, after all! All the MOSFET really cares about is the signal to turn it on and off, it doesn't really care much for how those signals are generated.