Author Topic: Would this bidirectinal switch ciruit work?  (Read 6303 times)

0 Members and 1 Guest are viewing this topic.

Offline hkBattousaiTopic starter

  • Regular Contributor
  • *
  • Posts: 117
  • Country: 00
Would this bidirectinal switch ciruit work?
« on: February 05, 2015, 09:48:38 am »
I'm trying to design a general purpose bidirectional switch with MOSFET transistors.

I have designed the following circuit.



In my design, two MOSFETs are placed back-to-back. It is assumed that two isolated DC supplies are available. The two MOSFETs are simultaneously turned on and off by the same digital control signal. When the MOSFETs are on, the ports INOUT1 and INOUT2 will be short circuited in both direction. When the MOSFETs are off, the MOSFETs will block any current through the aforesaid ports.

Would this circuit work? Is it necessary to isolate the digital pulse on both sides too? Please make some suggestions for me and correct my mistakes.
 

Offline Paul Price

  • Super Contributor
  • ***
  • Posts: 1433
Re: Would this bidirectinal switch ciruit work?
« Reply #1 on: February 05, 2015, 12:24:26 pm »
is bidirectional:  No
Will work:          No
 

Offline dannyf

  • Super Contributor
  • ***
  • Posts: 8221
  • Country: 00
Re: Would this bidirectinal switch ciruit work?
« Reply #2 on: February 05, 2015, 12:31:50 pm »
Quote
Would this circuit work?

Why do you think it will work?
================================
https://dannyelectronics.wordpress.com/
 

Offline penfold

  • Frequent Contributor
  • **
  • Posts: 675
  • Country: gb
Re: Would this bidirectinal switch ciruit work?
« Reply #3 on: February 05, 2015, 12:52:25 pm »
As long as that ground is floating, of course it should work.  I personally wouldn't rely on the anti-parallel diodes built in the MOSFET and would add something with a low forward drop and fast recovery externally.
 

Offline voja

  • Contributor
  • Posts: 45
  • Country: 00
Re: Would this bidirectinal switch ciruit work?
« Reply #4 on: February 05, 2015, 01:07:44 pm »
DC1 and DC2 are shorted, I guess? There's no need to isolate them. Same is with digital input signals.
Check if MOSFETs already have Zener protection at input and add them externally if they don't.
:( + :( = :) :)
 

Offline mij59

  • Frequent Contributor
  • **
  • Posts: 693
  • Country: nl
Re: Would this bidirectinal switch ciruit work?
« Reply #5 on: February 05, 2015, 01:14:13 pm »
You need only one mosfet driver.
In the on state the mosfet will conduct in both directions of current flow, no need for extra diodes.
 

Offline hkBattousaiTopic starter

  • Regular Contributor
  • *
  • Posts: 117
  • Country: 00
Re: Would this bidirectinal switch ciruit work?
« Reply #6 on: February 05, 2015, 01:16:54 pm »
As long as that ground is floating, of course it should work.
I get it. The ground must be only for this circuit block and must be isolated from the main circuit ground. It also means that the driving digital signal must be isolated by an optocoupler or something like that, isn't it?

I personally wouldn't rely on the anti-parallel diodes built in the MOSFET and would add something with a low forward drop and fast recovery externally.
I didn't understand this part of your answer. When the switch is off, one of the MOSFETs and the other's body diode together will be blocking the current. It is not the body diode alone. Can you please explain your warning a little more.

You need only one mosfet driver.
In the on state the mosfet will conduct in both directions of current flow, no need for extra diodes.
But if I use only one MOSFET, I can't only block the current in one direction. I want to be able to switch both AC and DC currents.

Thank you for your answers. I will update my schematic soon and share it here.
 

Offline penfold

  • Frequent Contributor
  • **
  • Posts: 675
  • Country: gb
Re: Would this bidirectinal switch ciruit work?
« Reply #7 on: February 05, 2015, 07:52:11 pm »
You're right there, you do need two mosfets, because it has two directional states,
 

Offline hkBattousaiTopic starter

  • Regular Contributor
  • *
  • Posts: 117
  • Country: 00
Re: Would this bidirectinal switch ciruit work?
« Reply #8 on: February 05, 2015, 09:22:46 pm »
How about this one?

 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22433
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Would this bidirectinal switch ciruit work?
« Reply #9 on: February 05, 2015, 09:41:11 pm »


This is all you need.  Instead of a switch, you can use a logic source and single gate driver IC, and whatever value gate resistors you wish.  The logic and supply must be isolated, that is all.

Note that some snubber or protection circuitry may be desirable, especially with quick switching times and around inductive loads (e.g., AC line inductance).  You could use a voltage peak, or dV/dt, type snubber per transistor, or An AC variation on either, connected from drain to drain; or just a boring R+C or MOV.

Tim
« Last Edit: February 05, 2015, 09:43:07 pm by T3sl4co1l »
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 29432
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Would this bidirectinal switch ciruit work?
« Reply #10 on: February 05, 2015, 10:00:00 pm »
For what sort of "general purpose"?
More than a *4066 can do?

Current & voltage requirements?
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline hkBattousaiTopic starter

  • Regular Contributor
  • *
  • Posts: 117
  • Country: 00
Re: Would this bidirectinal switch ciruit work?
« Reply #11 on: February 06, 2015, 07:55:59 am »


This is all you need.  Instead of a switch, you can use a logic source and single gate driver IC, and whatever value gate resistors you wish.  The logic and supply must be isolated, that is all.

Very nice idea. I redesigned my circuit according to your suggestion. Is it OK now? Please tell me if still there is a problem in the circuit.

 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22433
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Would this bidirectinal switch ciruit work?
« Reply #12 on: February 06, 2015, 02:50:58 pm »
Yes, except you should still have independent Rg's because the transistors will switch at different voltages.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline MF-jockey

  • Regular Contributor
  • *
  • Posts: 55
  • Country: de
Re: Would this bidirectinal switch ciruit work?
« Reply #13 on: February 10, 2015, 09:47:27 pm »
You should have a look to Panasonic's power PhotoMOS relays, it works itself

AQY211EH 30 V-AC/DC 1000 mA, ... AQY216EH 600 V-AC/DC 50 mA

ore bigger types like

AQZ204D  400 V-AC/DC 450 mA
 

Offline lewis

  • Frequent Contributor
  • **
  • Posts: 704
  • Country: gb
  • Nullius in verba
Re: Would this bidirectinal switch ciruit work?
« Reply #14 on: February 10, 2015, 10:02:43 pm »
There are photovoltaic optocouplers available designed precisely for this purpose: https://www.google.co.uk/search?q=photovoltaic%20optocoupler

Having the MOSFET sources and system ground at the same potential significantly limits the usefulness of this circuit, isolated is the way to go. Especially if you want to switch AC.

The circuit you'll need is on the last page of this PDF: http://www.irf.com/technical-info/designtp/dt94-5.pdf
I will not be pushed, filed, stamped, indexed, briefed, debriefed or numbered.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf