Bidirectional DCDC Converter

[Tom M]

I've been wanting to build a bidirectional DCDC converter. I'm thinking if
i make a full-bridge DCDC boost converter i could replace the rectifiers on
the secondaray side with MOSFETS and use them as diodes to allow the current
to flow from the source to the storage device. Then reverse the opration
and the power could flow from the storage device to the source again.

The application 'm thinking of right now for it is a battery to a motor.
Normal flow is from battery to motor. Then to stop, regen the energy from
the motor and dump it back to the batteries (as much as possible).

Can this actually work as i'm thinking it will? Or am i missing something
vial here?

- Tom

 

[Robert Baer]

I've been wanting to build a bidirectional DCDC converter. I'm thinking if
i make a full-bridge DCDC boost converter i could replace the rectifiers on
the secondaray side with MOSFETS and use them as diodes to allow the current
to flow from the source to the storage device. Then reverse the opration
and the power could flow from the storage device to the source again.

The application 'm thinking of right now for it is a battery to a motor.
Normal flow is from battery to motor. Then to stop, regen the energy from
the motor and dump it back to the batteries (as much as possible).

Can this actually work as i'm thinking it will? Or am i missing something
vial here?

- Tom

You are missing a lot.
Start with power going into port A on a presumed device; this gives
power out of port B.
But there is a filter capacitor on port B, so port B sees a power
source - and acts like an input and converts that to power at port A.
But port a must have a filter capacitor when it is acting like an
output, so....
See the contradiction?

 

[Mochuelo]

I've been wanting to build a bidirectional DCDC converter. I'm thinking if
i make a full-bridge DCDC boost converter i could replace the rectifiers on
the secondaray side with MOSFETS and use them as diodes to allow the current
to flow from the source to the storage device. Then reverse the opration
and the power could flow from the storage device to the source again.

The application 'm thinking of right now for it is a battery to a motor.
Normal flow is from battery to motor. Then to stop, regen the energy from
the motor and dump it back to the batteries (as much as possible).

Can this actually work as i'm thinking it will? Or am i missing something
vial here?

That is already used in some electric/hybrid vehicles. Search for
"regenerative braking bidirectional converter". You'll find tons of
information.

 

[BFoelsch]

You are missing a lot.
Start with power going into port A on a presumed device; this gives
power out of port B.
But there is a filter capacitor on port B, so port B sees a power
source - and acts like an input and converts that to power at port A.
But port a must have a filter capacitor when it is acting like an
output, so....
See the contradiction?

No. Please explain more details of this problem.

Thanks in advance.

 

[Mochuelo]

You are missing a lot.
Start with power going into port A on a presumed device; this gives
power out of port B.
But there is a filter capacitor on port B, so port B sees a power
source - and acts like an input and converts that to power at port A.
But port a must have a filter capacitor when it is acting like an
output, so....
See the contradiction?

Can't find the sense in what you wrote, and especially in your attempt
to relate capacitors with power sources, or with outputs. The Buck
converter, for instance, does not need a capacitor at its output.

 

[Ol' Duffer]

I've been wanting to build a bidirectional DCDC converter. I'm thinking if
i make a full-bridge DCDC boost converter i could replace the rectifiers on
the secondaray side with MOSFETS and use them as diodes to allow the current
to flow from the source to the storage device. Then reverse the opration
and the power could flow from the storage device to the source again.

The application 'm thinking of right now for it is a battery to a motor.
Normal flow is from battery to motor. Then to stop, regen the energy from
the motor and dump it back to the batteries (as much as possible).

Yes, you can build bi-directional synchronous converters.
But for your application, I think you may find that your motor
used as a generator produces significantly less voltage than
it requires as a motor, so you may be better off with two
separate circuits each optimized for its own job, and logic to
control them.

 

[Tim Wescott]

I've been wanting to build a bidirectional DCDC converter. I'm thinking if
i make a full-bridge DCDC boost converter i could replace the rectifiers on
the secondaray side with MOSFETS and use them as diodes to allow the current
to flow from the source to the storage device. Then reverse the opration
and the power could flow from the storage device to the source again.

The application 'm thinking of right now for it is a battery to a motor.
Normal flow is from battery to motor. Then to stop, regen the energy from
the motor and dump it back to the batteries (as much as possible).

Can this actually work as i'm thinking it will? Or am i missing something
vial here?

- Tom

Do a web search for "synchronous rectification" -- it's pretty much what
you describe.

 

[Joerg]

Hello Mochuelo,

That is already used in some electric/hybrid vehicles. Search for
"regenerative braking bidirectional converter". You'll find tons of
information.

Like this one:
http://pearlx.snu.ac.kr/Publication/0135_11634.pdf

Regards, Joerg

 

[Terry Given]

I've been wanting to build a bidirectional DCDC converter. I'm thinking if
i make a full-bridge DCDC boost converter i could replace the rectifiers on
the secondaray side with MOSFETS and use them as diodes to allow the current
to flow from the source to the storage device. Then reverse the opration
and the power could flow from the storage device to the source again.

The application 'm thinking of right now for it is a battery to a motor.
Normal flow is from battery to motor. Then to stop, regen the energy from
the motor and dump it back to the batteries (as much as possible).

Can this actually work as i'm thinking it will? Or am i missing something
vial here?

- Tom

Yep, it can and does work just fine.

I have a few small planar forward SMPS sitting in front of me that use
synchronous rectification (FETs instead of diodes). At no load, because
the FET happily conducts current in either direction when ON, the output
inductor current is a triangle wave centred on zero amps - no
discontinuous inductor current . When the primary FETs turn ON, the
initial primary current is negative. Voila, reverse power flow.

Cheers
Terry

 

[Robert Baer]

Can't find the sense in what you wrote, and especially in your attempt
to relate capacitors with power sources, or with outputs. The Buck
converter, for instance, does not need a capacitor at its output.

....and it *cannot* work in reverse!

 

[Robert Baer]

Yep, it can and does work just fine.

I have a few small planar forward SMPS sitting in front of me that use
synchronous rectification (FETs instead of diodes). At no load, because
the FET happily conducts current in either direction when ON, the output
inductor current is a triangle wave centred on zero amps - no
discontinuous inductor current . When the primary FETs turn ON, the
initial primary current is negative. Voila, reverse power flow.

Cheers
Terry

....which is not whathe was relating to at all.

 

[Terry Given]

...which is not whathe was relating to at all.

Yes it is. Who cares where the reverse power comes from (in this case
the output inductor) its still flowing backwards.

(+ve Dc bus volts)*(-ve primary current) = -ve power.


Cheers
Terry

 

[Terry Given]

...and it *cannot* work in reverse!

Yes it can. Use a synchronous rectifier. Then turn off the main FET
(which has an anti-parallel diode) and what do you have - a BOOST
converter, moving power from the output to the input. Then turn the main
FET back on, giving a boost converter with a synchronous rectifier.

Cheers
Terry