Loading ...
Sorry, an error occurred while loading the content.

Re: [beam] H_Bridge

Expand Messages
  • Duane C. Johnson
    Hi Mark; ... I have never liked this part. OK, it has it s uses. I have used it. My basic problem with the L298 and others like it is the poor efficiency. It
    Message 1 of 19 , Mar 1, 2003
    • 0 Attachment
      Hi Mark;

      Mark Hillier wrote:
      >
      > > Hi,
      > > Does anyone know which transistors(readyly available)
      > > to use in a standard H-bridge configuration to drive
      > > 12v, 2amp motors?
      > > Any idea about FZT605,705 OR BCV27,26 Darlignton
      > > Pairs?
      >
      > How about the L298N ?

      I have never liked this part.
      OK, it has it's uses. I have used it.

      My basic problem with the L298 and others like it
      is the poor efficiency. It uses emitter follower
      transistors on the high side. This about guarantees
      that there will be 3 volts or more across the high
      side transistors.

      This is why the Tilden and other bipolar H-bridges
      are so favored as they have complementary transistors
      with much higher efficiency.

      With frustration with the L298 I looked for a better
      H-bridge. I came up with this circuit:
      http://www.redrok.com/actuator.htm#act01
      OK, it has some warts, but works well if high switching
      speeds are not needed.

      I presented the design to our legal department and
      we decided to patent it. It's still pending.

      Also, the LMD18200 with its internal MOSFET gate
      drive charge pump works much better than the L298.

      > http://www.hvwtech.com/pages/products_view.asp?ProductID=198

      > A complete kit with PCB is also available

      > Mark Hillier
      > President, HVW Technologies Inc.
      > Mark@...
      > Tel:(403)730-8603 Fax:(403)730-8903
      > http://www.HVWTech.com
      > Development Tools for Embedded Design

      Duane

      --
      Home of the $35 LED solar tracker.
      http://www.redrok.com/electron.htm#led3
      CUL8ER \ \ \ \ \ \\ \ \ Receiver
      Powered by\ \ \ \ \ \\ \ \ [*]
      Thermonuclear \ \Solar\Energy\from the Sun \ /////|
      Energy(the Sun) \ \ \ \ \\ \ / / /\/ / /|
      \ \ \ \ \ /\ / \/ / / / |
      WA0VBE \ \ \ \ / /\ \/ / / \/ /|
      Ziggy \ \ \/ / / \ \/ \/ /\ |
      \ / \ \/ / /\ \\ / \ / / |
      "Red Rock Energy" === ===\ / \ / \ === \ / ===
      Duane C. Johnson, Designer=== === \ \ === / |
      1825 Florence St Mirrors,Heliostats,Controls & Mounts|
      White Bear Lake, Minnesota \ \ / |
      USA 55110-3364 \ \ |
      (651)635-5O65 work \ \ / |
      (651)426-4766 home use Courier New Font \ \ |
      (413)556-659O Fax copyright \ / |
      (651)583-2O62 Red Rock Energy Site (C)980907 ===\ |
      redrok@... (my primary email: address) \ |
      redrok2@... (Hotmail address) \ |
      duane.johnson@... (Unisys address) \ |
      http://www.redrok.com/index.htm (My New Web site) \|
      These are my opinions, and not that of Unisys Corp. ===
    • Wilf Rigter
      There are a buncjh more interesting thyristors in dc motor driver circuits in this application note http://us.st.com/stonline/books/pdf/docs/5201.pdf wilf
      Message 2 of 19 , Mar 1, 2003
      • 0 Attachment
        There are a buncjh more interesting "thyristors in dc motor driver circuits"  in this application note
         
         
        wilf
        ----- Original Message -----
        Sent: Saturday, March 01, 2003 8:38 AM
        Subject: Re: [beam] H_Bridge

        Hi Mark;

        Mark Hillier wrote:
        >
        > > Hi,
        > > Does anyone know which transistors(readyly available)
        > > to use in a standard H-bridge configuration to drive
        > > 12v, 2amp motors?
        > > Any idea about FZT605,705 OR BCV27,26 Darlignton
        > > Pairs?
        >
        > How about the L298N ?

        I have never liked this part.
        OK, it has it's uses. I have used it.

        My basic problem with the L298 and others like it
        is the poor efficiency. It uses emitter follower
        transistors on the high side. This about guarantees
        that there will be 3 volts or more across the high
        side transistors.

        This is why the Tilden and other bipolar H-bridges
        are so favored as they have complementary transistors
        with much higher efficiency.

        With frustration with the L298 I looked for a better
        H-bridge. I came up with this circuit:
        http://www.redrok.com/actuator.htm#act01
        OK, it has some warts, but works well if high switching
        speeds are not needed.

        I presented the design to our legal department and
        we decided to patent it. It's still pending.

        Also, the LMD18200 with its internal MOSFET gate
        drive charge pump works much better than the L298.

        > http://www.hvwtech.com/pages/products_view.asp?ProductID=198

        > A complete kit with PCB is also available

        > Mark Hillier
        > President, HVW Technologies Inc.
        > Mark@...
        > Tel:(403)730-8603 Fax:(403)730-8903
        > http://www.HVWTech.com
        > Development Tools for Embedded Design

        Duane

        --
             Home of the $35 LED solar tracker.
            http://www.redrok.com/electron.htm#led3
           CUL8ER  \    \ \     \      \ \\   \      \  Receiver
          Powered by\    \ \     \      \ \\   \      \      [*]
        Thermonuclear    \ \Solar\Energy\from the Sun \ /////|
        Energy(the Sun)    \ \     \      \ \\   \ / / /\/ / /|
                       \    \ \     \      \ /\ / \/  /  /  / |
           WA0VBE       \    \ \     \ /   /\ \/   /   /  \/ /|
          Ziggy          \    \ \/    /    / \ \/   \/    /\  |
                          \ /  \ \/    /     /\ \\ / \   /  / |
        "Red Rock Energy" ===  ===\ /   \ /    \ ===  \ /    ===
        Duane C. Johnson, Designer===   ===     \ \   ===  /  |
        1825 Florence St  Mirrors,Heliostats,Controls & Mounts|
        White Bear Lake, Minnesota                \ \     /   |
        USA         55110-3364                     \ \        |
        (651)635-5O65    work                       \ \  /    |
        (651)426-4766   home  use Courier New Font   \ \      |
        (413)556-659O  Fax                copyright   \ /     |
        (651)583-2O62 Red Rock Energy Site (C)980907  ===\    |
        redrok@...     (my primary email: address) \   |
        redrok2@...              (Hotmail address) \  |
        duane.johnson@...          (Unisys  address) \ |
        http://www.redrok.com/index.htm    (My New Web site) \|
        These are my opinions, and not that of Unisys Corp.  ===


        To unsubscribe from this group, send an email to:
        beam-unsubscribe@egroups.com



        Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

        ---
        Outgoing mail is certified Virus Free.
        Checked by AVG anti-virus system (http://www.grisoft.com).
        Version: 6.0.455 / Virus Database: 255 - Release Date: 2/13/2003
      • Dwayne Reid
        ... This discussion resembles one that took place recently on the Open Source Motor Controller (osmc) group recently. The following is a copy of a post I made
        Message 3 of 19 , Mar 1, 2003
        • 0 Attachment
          At 10:38 AM 3/1/03 -0600, Duane C. Johnson wrote:

          >With frustration with the L298 I looked for a better
          >H-bridge. I came up with this circuit:
          >http://www.redrok.com/actuator.htm#act01
          >OK, it has some warts, but works well if high switching
          >speeds are not needed.

          This discussion resembles one that took place recently on the Open Source
          Motor Controller (osmc) group recently. The following is a copy of a post
          I made there:

          At 09:23 AM 2/24/03 +1100, Brett Paulin wrote:
          >
          >I'm planning on having just a single "leg" of PWM'ed Fets on the ground
          >side of a relay based H-Bridge made with Hi Current SPDT relays. This
          >removes the need for special hi-side gate drivers, Anti-shoot through
          >logic, and cuts the number of Fets to 1/4 of normal, while still allowing
          >nice smooth PWM control.

          We did a design similar to that many years ago (controlled gear motor
          linear actuators) but had a lot of problems with relays failing (welding)
          while braking the motor.

          The design that worked used 2 relays and 2 FETs - the FETs switched the
          ground side, the relays switched the hot side (24V). A little tiny PLD
          (EP320) controlled the whole works.

          We basically had 3 modes with our version: stopped (with active braking),
          forward, reverse.

          Forward and Reverse worked similarly: turn off both FETs (turn off
          braking), close 1 relay, wait, verify relay closed (loop back to wait if
          necessary), turn on 1 FET.

          Stopping was the same for both modes: turn off FET, wait short time, open
          relay, wait, verify relay opened (loop back to wait if necessary), turn
          both FETs on for dynamic braking.

          The relays were P&B T90 series (I'd use American Zettler AZ-2150 form-A
          relays now). The FETs were IRCZ44 with a simple PWM over-current
          detector/limiter.

          There was a simple pull-up from each motor lead to 1/2 supply voltage that
          was used to test for shorted relays or FETs.

          We built several thousands of these for one client (variable pitch fans
          used in Caterpillar earth moving equipment) with great success.

          <end of original post>

          I'd consider something similar for this application, except using Aromat
          JS1E relays. These relays are relatively small and very robust for their
          size: up to 10A rating.

          The problem with SCRs in the top leg is still voltage drop - a typical
          forward voltage for an SCR is much more than a saturated bipolar transistor
          or MOSFET.

          The ultimate in tiny motor drivers still might be complimentary smt MOSFETs
          driven hard and fast by HC/HCT/AC/ACT invertors. I'll have to give it a
          go, I guess.

          dwayne

          --
          Dwayne Reid <dwayner@...>
          Trinity Electronics Systems Ltd Edmonton, AB, CANADA
          (780) 489-3199 voice (780) 487-6397 fax

          Celebrating 19 years of Engineering Innovation (1984 - 2003)
          .-. .-. .-. .-. .-. .-. .-. .-. .-. .-
          `-' `-' `-' `-' `-' `-' `-' `-' `-'
          Do NOT send unsolicited commercial email to this email address.
          This message neither grants consent to receive unsolicited
          commercial email nor is intended to solicit commercial email.
        • Dwayne Reid
          I just made a post regarding using relays as the high side switching element for an H-bridge. I got to thinking (always a dangerous thing!) and realized that
          Message 4 of 19 , Mar 1, 2003
          • 0 Attachment
            I just made a post regarding using relays as the high side switching
            element for an H-bridge. I got to thinking (always a dangerous thing!) and
            realized that one probably does not need (or want) an electronic braking
            mode when driving the motors. If this is the case, one could use a single
            SPDT relay for the HI-side with 2- LO-side FETs. Further - if you made it
            so that the relaxed position of the relay was forward, the relay would
            consume power only while in reverse mode.

            The obvious extension to this is to go back to Brett Paulin's suggestion of
            a single DPDT relay with a single LO-side FET switch. If one takes care to
            change the relay position only while the FET was off, a tiny DPDT DIP relay
            would probably work just fine.

            dwayne

            --
            Dwayne Reid <dwayner@...>
            Trinity Electronics Systems Ltd Edmonton, AB, CANADA
            (780) 489-3199 voice (780) 487-6397 fax

            Celebrating 19 years of Engineering Innovation (1984 - 2003)
            .-. .-. .-. .-. .-. .-. .-. .-. .-. .-
            `-' `-' `-' `-' `-' `-' `-' `-' `-'
            Do NOT send unsolicited commercial email to this email address.
            This message neither grants consent to receive unsolicited
            commercial email nor is intended to solicit commercial email.
          • Bruce Robinson
            ... Not necessarily, Dwayne. I had to use electronic braking on Hider (a relatively fast wheeled robot). It turned by stopping or slowing one wheel; without
            Message 5 of 19 , Mar 1, 2003
            • 0 Attachment
              Dwayne Reid wrote:
              >
              > ... I got to thinking (always a dangerous thing!)
              > and realized that one probably does not need (or
              > want) an electronic braking mode when driving the
              > motors. ...

              Not necessarily, Dwayne.

              I had to use electronic braking on Hider (a relatively fast wheeled
              robot). It turned by stopping or slowing one wheel; without the braking
              it made very wide turns that ruined it's ability to maneuver. On a
              partially built walker with high-efficiency gearmotors, I've had to use
              braking to prevent it from sinking down on it's legs when a leg motor is
              stopped.

              These conditions are highly variable and will not apply in many cases.
              I'd suggest keeping your options open when you design a driver, just in
              case.

              Bruce
            • Wilf Rigter
              Hello Dwayne, Sounds interesting. Do you have some links to the various motor controllers that have evolved during discussions on the Open Source Motor
              Message 6 of 19 , Mar 1, 2003
              • 0 Attachment
                Hello Dwayne,
                 
                Sounds interesting.  Do you have some links to the various motor controllers that have evolved during discussions on the Open Source Motor Controller (osmc) list?
                 
                Here is my contribution to the (beam) H_Bridge discussion: 
                 
                 
                 
                The HyBridge  is a simple experimental (untested)  Hybrid H-bridge circuit that has some nice features.
                 
                It has zero standby current with full active braking. 
                Quiescent current during freewheeling is <20ma. 
                Quiescent current during motor forward / reverse is <10ma.
                 
                With the transistors shown  the HyBridge is conservatively rated at 2A and 20V and should handle motor reversing without too much effort (I hope).  The maximum total saturation voltage at 2A should be a modest 1.5V.  The resistors shown are suitable for a motor supply voltage of 9 -12V. The input signals are assumed to be 0V and 5V.  Both the IRFZ14 MOSFET and the TIP125 darlington devices have build in freewheeling diodes that should be sufficient for motors within the rating of the bridge. Although the power transistor turn on times are relatively slow and any PWM chopping frequency must be kept low but as far as I can see there should be no "shoot through" current.  
                 
                The input logic is as follows:
                 
                In 1   In 2       Motor
                 
                0V    0V        brake
                5V    0V        forward
                0V    5V        reverse
                5V    5V        freewheel  
                 
                Estimated cost per HyBridge is about $5 not including a PCB or heatsink (if needed).  
                 
                I also have a (more complex) PowerSmart Hybridge design with total Vsat of <0.5V at 2A although it has a  quiescent current of 100ma in the forward and reverse mode.
                 
                I look forward to comments and suggestions for improvements but keep in mind this is a beam design that stresses simplicity and low cost.
                 
                regards
                 
                wilf
                 
                ----- Original Message -----
                Sent: Saturday, March 01, 2003 12:42 PM
                Subject: Re: [beam] H_Bridge

                At 10:38 AM 3/1/03 -0600, Duane C. Johnson wrote:

                >With frustration with the L298 I looked for a better
                >H-bridge. I came up with this circuit:
                >http://www.redrok.com/actuator.htm#act01
                >OK, it has some warts, but works well if high switching
                >speeds are not needed.

                This discussion resembles one that took place recently on the Open Source
                Motor Controller (osmc) group recently.  The following is a copy of a post
                I made there:

                At 09:23 AM 2/24/03 +1100, Brett Paulin wrote:
                >
                >I'm planning on having just a single "leg" of PWM'ed Fets on the ground
                >side of a relay based H-Bridge made with Hi Current SPDT relays.  This
                >removes the need for special hi-side gate drivers, Anti-shoot through
                >logic, and cuts the number of Fets to 1/4 of normal, while still allowing
                >nice smooth PWM control.

                We did a design similar to that many years ago (controlled gear motor
                linear actuators) but had a lot of problems with relays failing (welding)
                while braking the motor.

                The design that worked used 2 relays and 2 FETs - the FETs switched the
                ground side, the relays switched the hot side (24V).  A little tiny PLD
                (EP320) controlled the whole works.

                We basically had 3 modes with our version: stopped (with active braking),
                forward, reverse.

                Forward and Reverse worked similarly: turn off both FETs (turn off
                braking), close 1 relay, wait, verify relay closed (loop back to wait if
                necessary), turn on 1 FET.

                Stopping was the same for both modes: turn off FET, wait short time, open
                relay, wait, verify relay opened (loop back to wait if necessary), turn
                both FETs on for dynamic braking.

                The relays were P&B T90 series (I'd use American Zettler AZ-2150 form-A
                relays now).  The FETs were IRCZ44 with a simple PWM over-current
                detector/limiter.

                There was a simple pull-up from each motor lead to 1/2 supply voltage that
                was used to test for shorted relays or FETs.

                We built several thousands of these for one client (variable pitch fans
                used in Caterpillar earth moving equipment) with great success.

                <end of original post>

                I'd consider something similar for this application, except using Aromat
                JS1E relays.  These relays are relatively small and very robust for their
                size: up to 10A rating.

                The problem with SCRs in the top leg is still voltage drop - a typical
                forward voltage for an SCR is much more than a saturated bipolar transistor
                or MOSFET.

                The ultimate in tiny motor drivers still might be complimentary smt MOSFETs
                driven hard and fast by HC/HCT/AC/ACT invertors.  I'll have to give it a
                go, I guess.

                dwayne

                --
                Dwayne Reid   <dwayner@...>
                Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
                (780) 489-3199 voice          (780) 487-6397 fax

                Celebrating 19 years of Engineering Innovation (1984 - 2003)
                  .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-
                     `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'
                Do NOT send unsolicited commercial email to this email address.
                This message neither grants consent to receive unsolicited
                commercial email nor is intended to solicit commercial email.



                To unsubscribe from this group, send an email to:
                beam-unsubscribe@egroups.com



                Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

                ---
                Outgoing mail is certified Virus Free.
                Checked by AVG anti-virus system (http://www.grisoft.com).
                Version: 6.0.455 / Virus Database: 255 - Release Date: 2/13/2003
              • Mark Hillier
                ... Agreed. The L298 has poor efficiency, but in many cases it isn t a huge issue. It *is* farily compact and inexpensive, which is nice. ... True, but these
                Message 7 of 19 , Mar 3, 2003
                • 0 Attachment
                  > Hi Mark;
                  >
                  > Mark Hillier wrote:
                  > >
                  > > > Hi,
                  > > > Does anyone know which transistors(readyly available)
                  > > > to use in a standard H-bridge configuration to drive
                  > > > 12v, 2amp motors?
                  > > > Any idea about FZT605,705 OR BCV27,26 Darlignton
                  > > > Pairs?
                  > >
                  > > How about the L298N ?
                  >
                  > I have never liked this part.
                  > OK, it has it's uses. I have used it.
                  >
                  > My basic problem with the L298 and others like it
                  > is the poor efficiency. It uses emitter follower
                  > transistors on the high side. This about guarantees
                  > that there will be 3 volts or more across the high
                  > side transistors.

                  Agreed. The L298 has poor efficiency, but in many cases it isn't a
                  huge issue. It *is* farily compact and inexpensive, which is nice.

                  > This is why the Tilden and other bipolar H-bridges
                  > are so favored as they have complementary transistors
                  > with much higher efficiency.
                  >
                  > With frustration with the L298 I looked for a better
                  > H-bridge. I came up with this circuit:
                  > http://www.redrok.com/actuator.htm#act01
                  > OK, it has some warts, but works well if high switching
                  > speeds are not needed.
                  >
                  > I presented the design to our legal department and
                  > we decided to patent it. It's still pending.
                  >
                  > Also, the LMD18200 with its internal MOSFET gate
                  > drive charge pump works much better than the L298.

                  True, but these will handle only a single motor each and they are
                  something like triple the cost per unit. If you consider that you need
                  tow of them to equal the function the cost of a single L298, then it's
                  SIX times more expensive !

                  Mark Hillier
                  President, HVW Technologies Inc.
                  Mark@...
                  Tel:(403)730-8603 Fax:(403)730-8903
                  http://www.HVWTech.com
                  Development Tools for Embedded Design
                • Dwayne Reid
                  ... ... I like this. The only drawback is that the TIP125 s are darlington devices - I d expect to see between
                  Message 8 of 19 , Mar 3, 2003
                  • 0 Attachment
                    At 09:40 PM 3/1/03 -0800, Wilf Rigter wrote:
                    >Hello Dwayne,
                    >
                    >Sounds interesting. Do you have some links to the various motor
                    >controllers that have evolved during discussions on the Open Source Motor
                    >Controller (osmc) list?

                    <http://groups.yahoo.com/group/osmc/files/>

                    >Here is my contribution to the (beam) H_Bridge discussion:

                    I like this. The only drawback is that the TIP125's are darlington devices
                    - I'd expect to see between 1-2V drop across them under heavy load. And,
                    of course, if you did not need the free-wheel mode, you could eliminate the
                    extra components needed to create that mode.

                    dwayne

                    --
                    Dwayne Reid <dwayner@...>
                    Trinity Electronics Systems Ltd Edmonton, AB, CANADA
                    (780) 489-3199 voice (780) 487-6397 fax

                    Celebrating 19 years of Engineering Innovation (1984 - 2003)
                    .-. .-. .-. .-. .-. .-. .-. .-. .-. .-
                    `-' `-' `-' `-' `-' `-' `-' `-' `-'
                    Do NOT send unsolicited commercial email to this email address.
                    This message neither grants consent to receive unsolicited
                    commercial email nor is intended to solicit commercial email.
                  • Wilf Rigter
                    Here is the simple pre-cursor design for the HyBridge schematic that helps explain the basic anti-shoot through design. The detail of interest is the way the
                    Message 9 of 19 , Mar 3, 2003
                    • 0 Attachment
                      Here is the simple pre-cursor design for the HyBridge schematic that helps explain the basic anti-shoot through design. 
                       
                       
                      The detail of interest is the way the PN2222 and PN22907 transistors are  used to "break before make"  the source and sink power output devices during input transitions. 
                       
                      This h-bridge is a hydrid MOSFET / Darlington design for 12V-18V supplies.  The nominal continuous current is 4 Amp.  The IRFZ44 is a rugged n-channel MOSFET withVds=60V  Rds= 0.024 ohm,  ID=50A and is designed 10V gate drive signals. . The cheaper  IRFZ48 (Digikey US$1) with Rds= 0.01 ohm and Id=70A is an even better choice.  The TIP 125 is a 5 A 60V device costing about $0.50. 
                       
                      This circuit is interesting in itself because it is capable of forward, reverse, brake and freewheel. It is designed to be driven with rail to rail input signals (ie 0V and 12V).  The The freewheel and chopping  require that the input signals are tri-stated or driven by a 6V signal. Alternatively the four 10K resistors can each  be driven independently.   With both inputs grounded the active brake mode is on and the circuit draws just a few ma.   
                       
                      With just two added resistors the basic circuit can be modfied to include a simple free wheeling mode when both inputs are at 12V as shown here:
                       
                       
                       
                      Next follows a new low supply voltage, low saturation, current limited  HyBridge2 circuit
                       
                      HyBridge2  is a novel H-Bridge design that includes several interesting features
                      :
                      1. Forward, reverse, brake and freewheel using two input pins
                      2. Low (5V) supply voltage operation
                      3. Lossless current limit sensing 
                      4. Low sourcing saturation voltage drop
                      5. Highly efficient proportional drive current.  
                      6. Minimum through shoot current
                      7. Low cost
                       
                      The circuit as shown is suitable for 3 - 5 A motors at 5V.  The circuit can be scaled with component changes to higher current and voltages. 
                       
                      The operating modes are selected by switching the inputs as follows:
                        
                      0V    0V        brake
                      5V    0V        forward
                      0V    5V        reverse
                      5V    5V        freewheel  
                       
                      The saturation control circuit dynamically adjusts base current to the output device depending on load conditions to achieve a low output saturation voltage. The two series diodes is in parrallel with the compound PNP base emitters and shunts input current when the collector- emitter voltage is saturated (<400mV).  Since the TIP42 saturation voltage is proportional to the ratio of Ib/Ic,  a low Ic shunts base current from the PN2907 and reduces the PN2907 emitter /TIP42 base current accordingly.  Since the TIP2 base current is now proportional to the collector current it can be used for lossless current sensing for the current limit circuit.     
                       
                      The lossless current sensing may not be very accurate but it is very simple indeed.  The 10 ohm sensing resistors may be adjusted to suit.
                      With the two 10K resistors acting as a voltage divider at the PN2222 base, the current limit occurs at about 1V across the 10ohm resistor. That corresponds to 100ma of TIP42 base current and is equal to 2A of motor current.  The current limiting controls the coresponding MOSFET gate voltage and that device must have a large enough heatsink to dissipate the IR heating .
                          
                      Saturated bipolar drive is notoriously inefficient at low load currents. The proportional base drive makes this circuit very efficient in variable load applications.   The TIP42 base current can be as high as 100mA near the short circuit limit and with a lightly loaded motor that base current drops to 20ma.   
                       
                      The IRLZ44 is a rugged n-channel MOSFET withVds=60V  Rds= 0.02 ohm,  ID=50A and is designed logic level gate drive signals. It may seem like overkill but at US$2 each think of it as cheap insurance.     
                       
                      Finally here are two implementations of an all MOSFET H-Bridge with forward, reverse brake and freewheel using the anti shoot through design.  MOSBridge1 is meant for 5V or 6V supply applications.
                       
                       
                       
                       
                       
                      and my favorite version of all: the general  purpose  >>> MOSBridge2 <<< design.
                       
                       
                      Use logic level MOSFETs for applications requiring up to 9V motor supply. Use normal MOSFETs for motor supplies  in the 10V - 18V. Higher motor supply applications will require some protection of the MOSFET gates.  
                       
                      High logic level input signals can be anything between 2V to 24V. If necesary add a base emitter resistor to the PN2222 to increase noise immunity for the low level logic signal.    
                       
                       
                      enjoy
                       
                      wilf
                       
                      ----- Original Message -----
                      ----- Original Message -----
                      Sent: Monday, March 03, 2003 12:52 PM
                      Subject: Re: [beam] H_Bridge

                      At 09:40 PM 3/1/03 -0800, Wilf Rigter wrote:
                      >Hello Dwayne,
                      >
                      >Sounds interesting.  Do you have some links to the various motor
                      >controllers that have evolved during discussions on the Open Source Motor
                      >Controller (osmc) list?

                      <http://groups.yahoo.com/group/osmc/files/>

                      >Here is my contribution to the (beam) H_Bridge discussion:

                      I like this.  The only drawback is that the TIP125's are darlington devices
                      - I'd expect to see between 1-2V drop across them under heavy load.  And,
                      of course, if you did not need the free-wheel mode, you could eliminate the
                      extra components needed to create that mode.

                      dwayne

                      --
                      Dwayne Reid   <dwayner@...>
                      Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
                      (780) 489-3199 voice          (780) 487-6397 fax

                      Celebrating 19 years of Engineering Innovation (1984 - 2003)
                        .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-
                           `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'


                      ---
                      Outgoing mail is certified Virus Free.
                      Checked by AVG anti-virus system (http://www.grisoft.com).
                      Version: 6.0.455 / Virus Database: 255 - Release Date: 2/13/2003
                    • Dwayne Reid
                      ... Just a quick note, Wilf: It looks as if you have drawn the P-channel MOSFETs upside down. You show the source lead of the P-channel FETs connected to the
                      Message 10 of 19 , Mar 3, 2003
                      • 0 Attachment
                        At 06:08 PM 3/3/03 -0800, Wilf Rigter wrote:
                        >Here is the simple pre-cursor design for the HyBridge schematic that helps
                        >explain the basic anti-shoot through design.

                        Just a quick note, Wilf: It looks as if you have drawn the P-channel
                        MOSFETs upside down. You show the source lead of the P-channel FETs
                        connected to the motor instead of the supply rail. I get your intent but
                        others may not.

                        dwayne

                        --
                        Dwayne Reid <dwayner@...>
                        Trinity Electronics Systems Ltd Edmonton, AB, CANADA
                        (780) 489-3199 voice (780) 487-6397 fax

                        Celebrating 19 years of Engineering Innovation (1984 - 2003)
                        .-. .-. .-. .-. .-. .-. .-. .-. .-. .-
                        `-' `-' `-' `-' `-' `-' `-' `-' `-'
                        Do NOT send unsolicited commercial email to this email address.
                        This message neither grants consent to receive unsolicited
                        commercial email nor is intended to solicit commercial email.
                      • Wilf Rigter
                        With a little feedback from Wayne (sharp eye!), here are the corrected mosbridge 1 and mosbridge 2 schematics. Breadboarded the mosbridge 2 circuit and it
                        Message 11 of 19 , Mar 3, 2003
                        • 0 Attachment
                          With a little feedback from Wayne (sharp eye!), here are the corrected mosbridge 1 and mosbridge 2  schematics.  
                           
                           
                           
                           
                           
                           
                           
                           
                          Breadboarded  the mosbridge 2 circuit and it seems to work quite well using static switching and a 12V motor supply.  I am getting quite excited about this design A complementary MOSFET bridge with brake and freewheel just doesn't get much simpler than this.  I will do more extensive tests including PWM and report back later. 
                           
                          wilf   
                           
                           
                           
                          ----- Original Message -----
                          Sent: Monday, March 03, 2003 6:21 PM
                          Subject: Re: [beam] H_Bridge

                          At 06:08 PM 3/3/03 -0800, Wilf Rigter wrote:
                          >Here is the simple pre-cursor design for the HyBridge schematic that helps
                          >explain the basic anti-shoot through design.

                          Just a quick note, Wilf:  It looks as if you have drawn the P-channel
                          MOSFETs upside down.  You show the source lead of the P-channel FETs
                          connected to the motor instead of the supply rail.  I get your intent but
                          others may not.

                          dwayne

                          --
                          Dwayne Reid   <dwayner@...>
                          Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
                          (780) 489-3199 voice          (780) 487-6397 fax

                          Celebrating 19 years of Engineering Innovation (1984 - 2003)
                            .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-
                               `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'
                          Do NOT send unsolicited commercial email to this email address.
                          This message neither grants consent to receive unsolicited
                          commercial email nor is intended to solicit commercial email.



                          To unsubscribe from this group, send an email to:
                          beam-unsubscribe@egroups.com



                          Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

                          ---
                          Outgoing mail is certified Virus Free.
                          Checked by AVG anti-virus system (http://www.grisoft.com).
                          Version: 6.0.455 / Virus Database: 255 - Release Date: 2/13/2003
                        • Wilf Rigter
                          Thanks to DWAYNE REID that is (sorry!) wilf ... From: Wilf Rigter To: beam@yahoogroups.com Sent: Monday, March 03, 2003 9:00 PM Subject: Re: [beam] H_Bridge
                          Message 12 of 19 , Mar 3, 2003
                          • 0 Attachment
                            Thanks to DWAYNE REID that is (sorry!)
                             
                            wilf
                            ----- Original Message -----
                            Sent: Monday, March 03, 2003 9:00 PM
                            Subject: Re: [beam] H_Bridge

                            With a little feedback from Wayne (sharp eye!), here are the corrected mosbridge 1 and mosbridge 2  schematics.  
                             
                             
                             
                             
                             
                             
                             
                             
                            Breadboarded  the mosbridge 2 circuit and it seems to work quite well using static switching and a 12V motor supply.  I am getting quite excited about this design A complementary MOSFET bridge with brake and freewheel just doesn't get much simpler than this.  I will do more extensive tests including PWM and report back later. 
                             
                            wilf   
                             
                             
                             
                            ----- Original Message -----
                            Sent: Monday, March 03, 2003 6:21 PM
                            Subject: Re: [beam] H_Bridge

                            At 06:08 PM 3/3/03 -0800, Wilf Rigter wrote:
                            >Here is the simple pre-cursor design for the HyBridge schematic that helps
                            >explain the basic anti-shoot through design.

                            Just a quick note, Wilf:  It looks as if you have drawn the P-channel
                            MOSFETs upside down.  You show the source lead of the P-channel FETs
                            connected to the motor instead of the supply rail.  I get your intent but
                            others may not.

                            dwayne

                            --
                            Dwayne Reid   <dwayner@...>
                            Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
                            (780) 489-3199 voice          (780) 487-6397 fax

                            Celebrating 19 years of Engineering Innovation (1984 - 2003)
                              .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-
                                 `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'
                            Do NOT send unsolicited commercial email to this email address.
                            This message neither grants consent to receive unsolicited
                            commercial email nor is intended to solicit commercial email.



                            To unsubscribe from this group, send an email to:
                            beam-unsubscribe@egroups.com



                            Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

                            ---
                            Outgoing mail is certified Virus Free.
                            Checked by AVG anti-virus system (http://www.grisoft.com).
                            Version: 6.0.455 / Virus Database: 255 - Release Date: 2/13/2003


                            To unsubscribe from this group, send an email to:
                            beam-unsubscribe@egroups.com



                            Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
                          • Wilf Rigter
                            To wrap up the mosfet h-bridge designs, I have posted the three most interesting circuits in the file section. All feature very low voltage drop, forward,
                            Message 13 of 19 , Mar 4, 2003
                            • 0 Attachment
                              To wrap up the mosfet h-bridge designs, I have posted the three most interesting circuits in the file section. All feature very low voltage drop, forward, reverse, brake and freewheel mode. 
                               
                               
                              The mosfets are chosen for the required current and voltage rating  typically up to 10A and 18V.  A suitable automotive type fuse in series with the +V supply should be used to protect against  short circuits. The next generation of these circuits will include a novel current limiting scheme. These h-bridges should cost between $5 and $10 to build from scratch. 
                               
                              enjoy
                               
                               wilf   
                               
                               

                              ---
                              Outgoing mail is certified Virus Free.
                              Checked by AVG anti-virus system (http://www.grisoft.com).
                              Version: 6.0.455 / Virus Database: 255 - Release Date: 2/13/2003
                            • Wilf Rigter
                              I have build and tested the MOSbridge 2 circuit and I am happy to report it works well. I used 2N3904/6 instead of PN2222/2907 driver transistors, 2.2K
                              Message 14 of 19 , Mar 9, 2003
                              • 0 Attachment
                                I have build and tested the MOSbridge 2 circuit and I am happy to report it
                                works well.

                                I used 2N3904/6 instead of PN2222/2907 driver transistors, 2.2K resistors
                                and some "free" Dpak N channel and P channel mosfets had I liberated from a
                                surplus harddisk IDE control board. The motor supply was 12V and the logic
                                signals 5V.

                                The circuit is quite efficient with PWM 0%-100% up to 5kHz. I briefly tested
                                at 50kHz PWM but got significant ringing above 500mA, so I backed off but
                                will investigate higher frequencies later. The anti-shoot through circuit
                                seems to work well, and from o'scope observations breaks motor current for
                                about 5usec before making. The rise and fall times of the winding voltage
                                waveforms were a modest 2usec. Surprisingly PWM was most efficient with one
                                input grounded. The MOSfet internal diodes were used for freewheeling. No
                                heatsinks were used for the MOSfets and the circuit remained cold while
                                switching a 1A motor load. The voltage drop was as expected for the device
                                Rdson specs. I expect, with better MOSfets, 18V motors and >10A currents
                                can be efficiently controlled with this simple circuit.

                                wilf

                                ----- Original Message -----
                                From: "Wilf Rigter" <wrigter@...>
                                To: <beam@yahoogroups.com>
                                Sent: Monday, March 03, 2003 6:08 PM
                                Subject: Re: [beam] H_Bridge

                                ---snip----

                                my favorite version of all: the general purpose >>> MOSBridge2 <<< design.
                                Use logic level MOSFETs for applications requiring up to 9V motor supply.
                                Use normal MOSFETs for motor supplies in the 10V - 18V. Higher motor supply
                                applications will require some protection of the MOSFET gates.

                                High logic level input signals can be anything between 2V to 24V. If
                                necesary add a base emitter resistor to the PN2222 to increase noise
                                immunity for the low level logic signal.


                                enjoy

                                wilf



                                ---
                                Outgoing mail is certified Virus Free.
                                Checked by AVG anti-virus system (http://www.grisoft.com).
                                Version: 6.0.459 / Virus Database: 258 - Release Date: 2/25/2003
                              Your message has been successfully submitted and would be delivered to recipients shortly.