Well, a 555 really wouldn't be the right IC to use for a charge pump.
The BJT version would be limited to Vcc-1.2v on a good day with a tailwind while driving downhill.
The CMOS version might work, but it's current source capability is really pretty dismal at less than maximum Vcc/Vdd; even at Vdd=12v you only get about 10mA - but with that high of an output current, you lose anywhere from 0.7v to 1.5v at the output (ref: National Semiconductor LMC555 Datasheet, May 2006)

This is why dedicated gate driver ICs are so attractive; all that stuff is built into them.

Have a read through the attached .pdf - good article on gate drivers. It's somewhat lengthy, but well worth reading through a few times.

 

Why you don't use the BJT H-bridge

 

Using a BJT H-bridge would kind of defeat the purpose of learning about MOSFET H-bridges, wouldn't it?

With BJT H-bridges, you're going to waste about 20% of the load current just driving the transistor's base. After all, Ib=Ic/10, and you have two transistors turned on at any one time... but if you're using MOSFETs, you don't have that constant "wasted" current. Once the gate is charged or discharged, it takes little to hold it there.

 

Hey Bill, something else to look at, this was a setup I used to control a DC motor before I started using Hbridge ic's.... I had to cut out some of the schematic since this is owned by the company I work for.... but it should still give you an idea on how I used fets.....

 

I think its called "driving while distracted"

To use an NFet as the pass device, put it the negative lead of the supply...

Only problem with that is many times you want a ground, a firm ground, for the circuit (things like power supplies). It wouldn't be a series pass if the MOSFET or BJT was going to ground.

Sometimes it is more trouble than it is worth to get a specific part, if you can kludge up an acceptable alternative.

One last question, what is it about this design that bother people? It reliably prevents shoot through from what I can see.

It is basically half of a H-Bridge BJT style.

 

Your LEDs would typically be limited to around 20mA current; which gives a max Ib=20mA and a resulting Ic of 200mA max to guarantee saturation. However, as the LED and transistor heat up, the Vf and Vb is going to decrease. So, you'll have to decrease the initial room-temp current to make up for that. Besides, you'll have to tailor the Vf of the LED's to the particular Vcc in use; what if it's poorly regulated, or from a battery?

Transistors are pretty fast, but if they're deeply saturated, you're still going to have some shoot-through. Providing a positive "dead time" guarantees that you won't.