Feedback To Source

[Yucca]

@Jesus and All,

I am thinking about a similar thing with my pulsemotor.

How to recover any BEMF generated charge and feed it back to the input?


An idea I have at the moment is this:

Instead of collecting the BEMF in a cap, feed it into a transformer. Use a diode to capture the BEMF as normal but just put a transformer where the cap normally goes.

The transformer output is then fed through a diode bridge into a large electrolytic that is sat over the motors supply input. Could use a single diode but the bridge may catch a little more ringing.

The large electrolytic will soak up the rectified pulse and make it available to the motor drive coil, adding voltage to the cap that is already charged by the supply.

Put a beefy diode on the supply side of the cap to prevent it leaking back into the supply.

It should be more efficient than collecting in one cap and then transfering using switches (FETs) into the supply side because there will be no switching losses and no cap to cap losses.

The cap should be large enough to hold one pulse with room to spare. The cap should also be low ESR like used in switch mode PSUs.

Toroidal transformer is prefered because it is efficient and will not interfere with mag field of the motor. I think if the mass of copper on the transformer is comparable to the motors coil mass then it should be ok. The core should be large enough so that it doesn´t saturate with the pulse. Of course toroids are a pain to wind and so initial tests could be conducted with El core power transformer. Could maybe use 120V to 12V transformer out of a cheapo power adapter.

P.S.
This idea may have already occured in this thread, apologies for not reading all of this thread yet, it has grown quite long.

[nievesoliveras]

@ Jesus:

Maybe if we used supercaps as opposed to those regular caps we could bleed off the energy in a slower, more usable manner?

Bill       PS my huge amount (due to the minimum order) of parts may be in by the end of the week.  In there are 50 germanium diodes and a bunch of other stuff.  I will give this a go.

Thank you @pirate.
I bought a package of three from goldmine and am doing experiments with them. I used the camera's 350v caps and lost transistors with the high voltage spikes. They are very powerful, but does not last long.

@Jesus and All,

I am thinking about a similar thing with my pulsemotor.

How to recover any BEMF generated charge and feed it back to the input?


An idea I have at the moment is this:

Instead of collecting the BEMF in a cap, feed it into a transformer. Use a diode to capture the BEMF as normal but just put a transformer where the cap normally goes.

The transformer output is then fed through a diode bridge into a large electrolytic that is sat over the motors supply input. Could use a single diode but the bridge may catch a little more ringing.

The large electrolytic will soak up the rectified pulse and make it available to the motor drive coil, adding voltage to the cap that is already charged by the supply.

Put a beefy diode on the supply side of the cap to prevent it leaking back into the supply.

It should be more efficient than collecting in one cap and then transfering using switches (FETs) into the supply side because there will be no switching losses and no cap to cap losses.

The cap should be large enough to hold one pulse with room to spare. The cap should also be low ESR like used in switch mode PSUs.

Toroidal transformer is prefered because it is efficient and will not interfere with mag field of the motor. I think if the mass of copper on the transformer is comparable to the motors coil mass then it should be ok. The core should be large enough so that it doesn´t saturate with the pulse. Of course toroids are a pain to wind and so initial tests could be conducted with El core power transformer. Could maybe use 120V to 12V transformer out of a cheapo power adapter.

P.S.
This idea may have already occured in this thread, apologies for not reading all of this thread yet, it has grown quite long.

Thank you @yucca
That idea of yours sounds good. Could you post a schematic of it?

Jesus

[nievesoliveras]

@all

In the meantime we find the way to get feedback to the source, I will post some useful information about multivibrators. The credit goes this time to lady @jeanna, who gave the link for this information.

Jesus

[Yucca]

Here is a schematic, I haven´t tried it yet. The principle is fairly simple and any transformer should work but best would be a high freq. handling toroidal core wound as 1:1 transformer, maybe 100 wraps on each half.

If your motor uses aircore then fairly small transformer should work. If motor coil has a core then the transformer will need to be a little bigger.

For best performance the transformer size will need matching to the coil BEMF pulse. I´m thinking a transformer with about the same weight as the motors drive coil would be a fair starting point for experiments.

[nievesoliveras]

@yucca

Thank you!

What will you use as the drive signal?

Jesus