Hi everyone, I've just signed to this forum. I have some questions about H bridge.
I constructed a H bridge circuit by following this picture, connected point A and B to pin 7 and 8 of Arduino. It worked.
However, I know there is another H bridge circuit, that is 2 PNP at the upper side, 2 NPN at the lower side.
Why there are 2 types of H bridge? What are the advantages and disadvantages to these 2 H bridge?
I changed the 2 upper side's transistors to PNP, and connected to Arduino, but it didn't work.
How should I connect this type of H bridge?
You get a problem when V (for motor > 5volts)
Logic HIGH (on A) is OK for Q3, but is NOT OK for Q1 (voltage at emitter rises)
Q1 will not satutate fully.. and get HOT.
..and: add flyback diodes
compare: what happens if you try the 'illegal' combination.. using NPN only vs NPN/PNP
Yes, that's a hopeless circuit for driving motors, it's quite good at burning out
transistors though.
Switching devices for power should be driven to saturation, the top transistors
in that circuit are emitter followers and without a bootstrap-supply for the base drive
so they will get hot and waste a few volts.
More importantly without flyback diodes it will simply destroy itself through
inductive kick-back spikes.
Single IC solutions are much easier to use, but if you want a better BJT circuit
with saturating transistors and flyback diodes, how about:
The layout's a bit odd, the H has been twisted over between top and bottom drivers
to show how the input transistors switch on one NPN and one PNP. Only two input
signals are needed, but they must not be HIGH at the same time of the bridge is
shorted out ("shoot through") which will destroy it. A full-wave
rectifier chip cleverly provides all 4 flyback diodes in one package.
Super-beta transistors have been chosen (its an all-surface mount design though).
resistors R5 and R6 are chosen to match the drive level and supply voltage, the above
values correspond to 6V supply with 20mA of base drive to the output transistors
(which have gain > 250 I think and several amps max). With more humble
transistors the max load will be less and R5 and R6 should be chosen for more like
50mA.
I am sorry the above circuit is difficult to know for me because I just know very very basic about H bridge.
I found a PNP H bridge circuit.
Would it develop voltage drop across resistor of Q1 and Q2 if both A and B are low?
if I add a capacitor parallel with the motor to solve back EMF. What are the advantages of this circuit compare with 4 NPN H bridge?(the first circuit I posted)
..this works (add diodes!) fine for motors running from 5V.
For voltages >5V, you need another kind of circuit.
(seach ebay/google for "L298", and see its schematic)
verystrong1991:
I am sorry the above circuit is difficult to know for me because I just know very very basic about H bridge.
I found a PNP H bridge circuit.
Would it develop voltage drop across resistor of Q1 and Q2 if both A and B are low?
if I add a capacitor parallel with the motor to solve back EMF. What are the advantages of this circuit compare with 4 NPN H bridge?(the first circuit I posted)
The complementary pair works significantly better than just 4-NPN types. The problem happens when the 2 that are supposed to be PNPs try to turn on or off. Since the emitter voltage also involves the drop across the motor, you could end up with not enough voltage and possibly not making it to saturation. This means you are burning energy in the transistor and I expect that it would turn the motor slower than expected and probably burn out the transistor.
Use flyback diodes as well (you can probably use the caps, but diodes are far superior and typical for this application), just reverse polarity from the way you expect current to flow if the transistor is on, Cathode pointed up in this case.
A disadvantage of the pnp / npn bridge is that the pnp transistors usually have worse stats compared to npn, so they probably have lower beta and more voltage drop. From a schematic point of view the complimentary h bridge also has to be driven either with HI or LOW, but it can't be hi-z since the top pnp will look like a diode and activate the bottom one through the 2 * 1k resistors. That's why that HC14 inverter is there . Since an arduino starts in hi-z while the bootloader initialises this might not be the best idea. I used such a bridge with 4 arduino pins, or you could use 2 more transistors ( or 2 inverters) and 2 pins.
The all npn bridge can be left with the inputs floating. As such you can connect it to 2 arduino pins. On the other hand driving the all npn bridge with 2 HI = self destruct. The pnp bridge does not have that function. Instead it brakes.
Thank all of you!
I still have some questions.
Why flyback diodes is more suitable than capacitor to this application?
Would the PNP be activated if A and B are low? Because these are connected to the output of 74HC14. Aren't these reserve biasing?
Both input high (or low) is legal. Motor stops - no short circuit.
..and U're right, when inputs are low- the PNPs connects positive voltage to motor (both sides)
Do you know how the coil in a car works? - (generate kilovolts).
A condeser doesn't handle that enegy spike as well as a diode.
(Working with small motors..small currents it may not be a problem. But 'history' says: use flyback diodes)
knut_ny:
Both input high (or low) is legal. Motor stops - no short circuit.
..and U're right, when inputs are low- the PNPs connects positive voltage to motor (both sides)
Do you know how the coil in a car works? - (generate kilovolts).
A condeser doesn't handle that enegy spike as well as a diode.
(Working with small motors..small currents it may not be a problem. But 'history' says: use flyback diodes)
That's not the reason - using just a capacitor will increase the switching current
spikes in the transistors to high levels. Typically a small capacitor across the motor
terminals is used to reduce EMI - it mustn't be too large of the switching devices
will start to struggle charging it.
An RC snubber circuit can be used to squash inductive spikes, but has to be tuned
to the actual load - the C limits the voltage excursion, the R soaks up the energy
to reduce or eliminate ringing.
A diode simply shorts out the spike, but takes a lot longer to absorb the energy
as the current falls more slowly due to the low forward voltage of a diode
(rate of change of current proportional to voltage).
Typical high-performance H-bridges use the switches themselves to deal with
such inductive back EMF, the diodes take over during dead-time only. This way
the inductive spikes actually put power back into the battery and the current
falls much faster as the back EMF = supply voltage. Look up fast-decay mode
and mixed-decay mode on wikipedia for more info I think.
