What kind of circuit can be used to swap the polarities of an input signal

[bhl777]

Hi All, I have a voltage coming out from a voltage regulator. I am looking for a specific kind of circuit to provide the polarity controlability from this voltage to the load.

An idea is shown in this picture. If we have an analog/digital signal to determine how we want to swap the polarity, by closing S1 and S2, or closing S3 and S4, we can achieve the controllable polarity swap.

Could anyone give me some suggestion any available ICs I can use to achieve this? The simplest design with accepatble precision is good enough. The current to go through these switches are around 100mA.

Thank you!

 

[FvM]

100 mA is slightly beyond the capability of analog switch ICs, I would use a MOSFET H-bridge. Gate drive solution depends on the LDO voltage range.

 

[CataM]

Hi,
Your schematic looks like an "H bridge". On the conduction state of the transistors there would be some loses, how sensitive is your load ?

 

[bhl777]

100 mA is slightly beyond the capability of analog switch ICs, I would use a MOSFET H-bridge. Gate drive solution depends on the LDO voltage range.

Hi FvM, Thank you! I have some additional questions
(1) if my current can go down to 50mA or 25mA, is it still possible to use analog switch ICs? Can I say generally the overall size of this switch is much smaller than power MOSFETs?
(2) If the LDO voltage ranges from 1.2V to 5V, would you give me some recommendation of the gate drive solutions?
(3) is this the simplest solution? Or some other commercially available ICs can do the similar work?

- - - Updated - - -

Hi,
Your schematic looks like an "H bridge". On the conduction state of the transistors there would be some loses, how sensitive is your load ?

Hi CataM, is it OK to have some loses in the H bridge. The precision of the output is not required to be high. Do you have any suggestions? Thank you!

 

[FvM]

1 - 5 V LDO voltage is good, because it allows to drive the MOSFET switches with e.g. 0V/12 V level independent of actual LDO voltage.

You'll find some low ohmic / high current analog switches that can work up to 50 or 100 mA. I suggest to review the selection guides of major manufacturers like Analog, Vishay.

 

[CataM]

To control the MOSFETs of the H-Bridge, the more voltage you have available on your "control signal" the better because it reduces the losses in the conduction state (MOSFETs as switches i.e. Triode/Ohmic region).

 

[KlausST]

Hi,

something like TS5A23160, rated for 5V, 200mA.

But be careful they are Make-before-break.


Klaus

 

[bhl777]

Hi Klaus, what should I pay attention when I put this switch on the breadboard? I found the term "Make-before-break" but did not understand what I should do initially to make it functional. Thank you!

Hi,

something like TS5A23160, rated for 5V, 200mA.

But be careful they are Make-before-break.


Klaus

 

[KlausST]

Hi,

Make-before-break means that during switching there is a short time where all the terminals are shorted together: NO+COM+NC.

When you connect NO and NC to the supply rails, then this causes a short. So don't do this.

Either you choose a connection where the short causes no problem, or you choose an other switch.

Klaus

[

 

[crutschow]

A double-pole, double-throw mechanical relay will also do what you want.

 

[bhl777]

Hi crutschow, will “Make-before-break” happen in mechanical relay as well? If not, would you recommend one part I can use, to have at least 50mA continuous switch current capability?

A double-pole, double-throw mechanical relay will also do what you want.

- - - Updated - - -

Thank you Klaus, is there any other analog switches that do not have "Make-before-break"? The part you recommended for me have 200mA continuous current capability, that is good enough for my application. In my application, any switches with >50mA continuous current capability would be good.

Hi,

Make-before-break means that during switching there is a short time where all the terminals are shorted together: NO+COM+NC.

When you connect NO and NC to the supply rails, then this causes a short. So don't do this.

Either you choose a connection where the short causes no problem, or you choose an other switch.

Klaus

[

 

[crutschow]

Hi crutschow, will “Make-before-break” happen in mechanical relay as well? If not, would you recommend one part I can use, to have at least 50mA continuous switch current capability?

...........................

Yes, all mechanical are break before make.
Just about any small DPDT relay should work. (Where do you buy parts?)
Just pick one with the operating coil voltage you want.
You will likely need a transistor driver to provide the coil current.
Be sure you put a diode (such as a 1N4148) across the coil (cathode to plus voltage side of coil) for transient suppression.

 

[bhl777]

Hi crutschow, can I also use normal break-before-make analog switches? I want to minimize the external components count. Do you think this one can achieve the similar function as DPDT relays? I will buy parts from digikey or mouser.

https://www.vishay.com/docs/67590/dg2735a.pdf

Yes, all mechanical are break before make.
Just about any small DPDT relay should work. (Where do you buy parts?)
Just pick one with the operating coil voltage you want.
You will likely need a transistor driver to provide the coil current.
Be sure you put a diode (such as a 1N4148) across the coil (cathode to plus voltage side of coil) for transient suppression.

 

[crutschow]

Hi crutschow, can I also use normal break-before-make analog switches? I want to minimize the external components count. Do you think this one can achieve the similar function as DPDT relays? I will buy parts from digikey or mouser.

https://www.vishay.com/docs/67590/dg2735a.pdf

Yes, that Vishay analog switch should work for 0-100mA as long as your voltage is no higher than 5V.
Note that the minimum supply and control voltages are 1.65V.

 

[bhl777]

Thank you crutschow. The last question, when you are saying the minimum control voltage, you are talking about minimum logic "1", is it correct? If I have 0V and 5v to this control pin, it will not be a problem, is it correct?

Yes, that Vishay analog switch should work for 0-100mA as long as your voltage is no higher than 5V.
Note that the minimum supply and control voltages are 1.65V.

 

[crutschow]

Thank you crutschow. The last question, when you are saying the minimum control voltage, you are talking about minimum logic "1", is it correct? If I have 0V and 5v to this control pin, it will not be a problem, is it correct?

Yes to the first, no to the second.
If you look at the Absolute Maximum Ratings chart on page 2 of the data sheet, the maximum input voltage is the supply voltage +0.3V. Above that you forward bias the input protection diodes.
But it also says the maximum input control current is 30mA, so if you put a resistor in series with the input you can prevent damage.
Since the maximum normal input current is 1µA, I would use a 100kΩ or so resistor in series with the control input. That will safely limit the input current to <50µA when the supply voltage is low.