Reviewing Pulse Motor Circuit Ideas and Theory

[citfta]

Here is a circuit for capturing the kickback and storing it to use for the next pulse.  This circuit does reduce the current draw from the battery by  almost half.  By adjusting the size of the storage capacitor you can reduce the current draw even more.  The size in the schematic is a little too large for most efficient operation but it does work.  A lower value will work even better depending on the inductance of the drive coil.

Carroll

[MileHigh]

It looks to me like both circuits don't work at all.  I am not doubting Carroll's statement that the
current draw is reduced but it certainly isn't because the storage capacitor is being charged and
the charge is being used for the next pulse.

Taking a third look at Carroll's circuit I see one totally bizarre path for charging up the storage
capacitor but it looks like only a fraction of the energy in the power coil would actually make it
into the storage capacitor.  Then Q2 may or may not work, I am not sure.  I am not sure how
an AC-coupled NPN transistor base input will behave because it is a non-standard circuit that
doesn't make sense.

[citfta]

Hello MileHigh,

It has been a while since we passed posts back and forth.

Ok let me explain the circuit operation to you.  This is a schematic for an idea I had on recycling the inductive kickback.  I have used this circuit and it does work.  There are a couple of small changes you can make to make it work better but it will work just as I have drawn it.

This circuit is for a pulse motor that uses a trigger coil that is pulsed by the passing magnet.  Nothing new there.  I have coupled the pulse to the base of the second transistor Q2.  I have found out that the 5K pot is not really needed.  Both the transistors are 2N3055 transistors in the metal case.  The pulse goes from the trigger coil to the 6uf cap to the base of Q2 and then back to ground through the coil and transistor Q1 as it also gets turned on at the same time.  A scope shot shows this to be working correctly.  Of course turning on the transistor like this will only give you a short on time until the cap going to the base gets mostly charged.  But we only need a short pulse anyway.  The value of 6uf was arrived at by trial and error until I got the pulse I wanted from Q2.

As far as the charging of the cap goes it will be easier to understand if we remove a couple of the diodes from the bridge.  Looking at the bridge remove the upper right and lower left diodes.  I only included a bridge to collect any excess ringing pulses that might occur from the quick turning off of the coil.  With only the other diodes in the circuit it should be easy to see the coil discharge current going to the upper left diode to the cap and the return from the cap going back through the lower right diode to the top of the coil.  As I said in the previous post this really works better with a smaller value for the cap.  A smaller value for the cap will of course charge to a higher voltage with less pulses.  Also because of the bridge the cap gets charged to almost the supply voltage as soon as the supply is turned on.  So the inductive kickback does not start out trying to charge a totally discharged cap.

Using this circuit I was able to get a pulse motor that had been running on 120 milliamps down to 40 milliamps with some careful tuning.

Take care and let me know if you have any more questions about this circuit,

Carroll

[MileHigh]

Carroll,

Thanks, you pointed out a conduction path for the discharging power coil that I did
not see.  I saw three unlikely paths and I missed the correct path that you described.

For what it's worth the AC coupling to switch on Q2 is a "cheat" and its stability and
overall reliability are probably questionable.  There may be a better way to do it with
an FET or MOSFET where you bias the gate with a very high impedance resistor divider
network such that you keep the switch off, and then the AC coupled trigger signal
switches it on.  You might even be able to do something similar with the original
2N3055 transistor.

MileHigh

[citfta]

MileHigh,

I agree there are probably several ways to switch Q2 better than the method I chose.  I was asked to design this for a couple of people that have limited electronics experience and therefore tried to keep it as simple as I could so they could copy it and get it to work.  And I know that at least one of them was able to build it and get it work.  I have also run it for extended periods of time with no problem getting Q2 to switch.  But those 2N3055's are some pretty tough transistors for a run of the mile type.  Other transistors probably wouldn't work so well in that configuration.

Thanks for your interest and taking the time to look at it.

Carroll