next topic »

[on:]

Is it even possible? How can it be achieved? I built an H bridge using 4 NPN IRF520 transistors but my multimeter reads 2.2v even though the power supply is a 9v battery !

I would be very grateful if you could help me

Thanks.

[Reply #1 on:]

I have not seen any working H-Bridge when the high-side and low-side transistors were all NPN or PNP.  I think you need PNP on the high side and NPN on the low side.


See:
http://www.w9xt.com/page_microdesign_toc.html

See particularly articles 7, 8 and 12 about low-side and high-side switching.

[Reply #2 on:]

I'd agree for bipolar...  For MOSFET H-bridges all-n-channel is common however, but special circuitry is in place to generate gate voltages
higher than the supply for the top MOSFETS.

Here's a bipolar H-bridge suggestion:
 
The middle resistors (R5 and R6) are used to set the maximum base current to the switch transistors,
so their value depends on the supply voltage and desired output current (and current gain of the
switch transistors).

This actual circuit I've used with surface mount NPN/PNP superbeta transistor pairs for the switches, which
have very low Vsat, and a clever trick of using a small surface mount bridge rectifier to provide all 4
free-wheeling diodes in one go.

You must be careful not to drive both inputs high at the same time!   Some means of monitoring over-current
would be a wise addition.

[Reply #3 on:]

Is it even possible?

Yes. Pretty much any combination of PNP/NPN can be made to work.

[Reply #4 on:]

IRF520 is an n-channel mosfet, not an npn transistor.

[Reply #5 on:]

IRF520 is an n-channel mosfet, not an npn transistor.

Fine, but how does that make any difference? Can I build my own h bridge using 4 IRF520? Maybe using two of these and two transistors?

It appears that an h bridge can be built because I actually have done so, the problem is that there is a considerable voltage drop that would cause heat and energy waste so it definitely is not the way to go.

[Reply #6 on:]

Can I build my own h bridge using 4 IRF520? Maybe using two of these and two transistors?

I think h-bridges using all NPN MOSFETs use a bootstrap circuit to keep the upstream MOSFET gate voltage higher than the h-bridge supply voltage. If you are thinking of making a DIY MOSFET h-bridge, then the below discussion might be of interest to you. Lots of inexpensive h-bridges now on ebay.

http://arduino.cc/forum/index.php?topic=53425.0

[Reply #7 on:]

also I don't think IRF 520 is even logic level

[Reply #8 on:]

Is it even possible?

Yes. Pretty much any combination of PNP/NPN can be made to work.

I, for one, would be interested in seeing an All NPN H-Bridge circuit. Can you point to one or draw one?

[Reply #9 on:]

Below is a bootstrap setup for a high side MOSFET.

[Reply #10 on:]

Below is a bootstrap setup for a high side MOSFET.

Note that that bootstrap circuit imposes two restrictions on the H-bridge:

1) When top MOSFET is idle (off) you should turn on the lower MOSFET to keep the capacitor charged (in the diagram
   the load resistor is performing that role).
2) The PWM duty cycle of the upper MOSFET must not reach 100% - in fact you have to ensure the off-periods are
   long enough for the capacitor to recharge through the diode and lower MOSFET, and that the on-periods are
   short enough that the capacitor doesn't self-discharge through leakage currents.  You need to keep the
  capacitor C charged all the time.