# Can I put graetz circuits' outputs in series and parallel?

I will have multiple coils that generate alternate current. Those coils are placed around a wheel which has some magnets on it. My current plan is to make it look like this:

I was considering rectifying output of each and then putting the rectified voltage in both series and parallel. For example of four coils, two and two would make series to increase voltage and those pairs would go in parallel for greater maximum current. That would look like this:

simulate this circuit – Schematic created using CircuitLab

I would probably add some capacitors to balance the voltage. But is this even possible?

• Have you considered putting your AC sources in series/parallel and rectifying the result with a single bridge? Commented Sep 21, 2016 at 10:03
• They would cancel each other randomly. The device is not precise enough for that - I just glued some magnets on a wheel and now I'm putting coils all around it. Commented Sep 21, 2016 at 10:09
• Use a sim to see if it works. Commented Sep 21, 2016 at 10:13

That's fine with certain constraints:

• The AC voltages have to be isolated from each other.
• The sum of the AC voltages on the left has to match those on the right.
• Add the capacitance between OUT- and OUT+.

The risk is that if the output voltage of one side is a little higher than the other then it will supply most of the current and may overheat.

And can I do anything to satisfy the second constraint? The individual AC power will come from separate coils and I assume they will never produce equal power, since the whole device is quite crudely crafted.

Parallel connection of transformer secondaries wound on the same bobbin even has this problem. It is resolved by the internal impedance of the coil. As the coil current is increased the voltage droops a little due to the internal resistance. This tends to be enough to balance out the currents. In your application you could add a series low-value resistor on each leg. You could check the effect by measuring the voltage drop across each series resistor to calculate the current from each leg.

• And can I do anything to satisfy the second constraint? The individual AC power will come from separate coils and I assume they will never produce equal power, since the whole device is quite crudely crafted. Commented Sep 21, 2016 at 9:20
• See the update. Commented Sep 21, 2016 at 9:24
• As an extra to the comments of Transistor. Apart from difference in ac voltages the phase of the ac sources also has to be taken into account. If you can connect the ac sources directly in series and parallel without the rectifiers then the system will also work when the rectifiers are added Commented Sep 21, 2016 at 9:33
• @Transistor I hope this isn't a tangential question, but do you think my design is better/worse than if I used polyphase rectifier (2 diods per coil only)? Commented Sep 21, 2016 at 10:11
• This is the first mention of poly-phase and effectively changes the question. Please edit to include context and relevant information. Commented Sep 21, 2016 at 10:14

From the question, the answer and the comments it is clear that we are looking a polyphase system. The solution with the bridge rectifiers is not ideal. Better is to set up a double polyphase rectifier with one side of the coils combined together and the other side of the coil equiped with 2 diodes. This like is done in a alternator for a car to get dc. In a car alternator you find mostly 3 phases but they can be more.

• I'm not sure if I understand how the connection should be done. Connect legs of the coils directly in series? Commented Sep 21, 2016 at 13:00
• Study the diagram of a car alternator. Gives you also the opportunity to improve your design Commented Sep 21, 2016 at 14:43