OK, I want to look at improving the 555 output drive. Here is something I'm working on...

One core problem is the Darlington on the top of the output driver is soaking up too much of the power supply voltage. So here are my thoughts on the subject.

I would like to reduce the parts count if possible. Any ideas?

 

The transistor drive in the large 555 diagram is differential, thus negating the need for the classic PNP/NPN pushpull stage. It seems that they couldn't get enough current to drive the upper stage's base-emitter enough, so they chose a darlington arrangement in order for enough output current.

I'm not sure what you're trying to do, the interface from the 555 to a driver of some sort? Why not take a fet driver (pretty much a high power push-pull) to drive a fet half-bridge?

Steve

 

The output on the high side is not to rail, and there are applications that need it. The only reason I have collector resistors is to minimize shoot through.

Consider this, if I were to connect the output of the 555 to a PNP transistor it would stay on even in the high state. This is a good example of the problem.

The other thing is I want to maintain the wide power supply characteristics.

Looking at it, I wonder if the discharge could be incorporated to replace one of the transistor drivers?

 

If you can tolerate a few hundred nanoseconds of propagation delay, I've got a circuit that, in simulation, swings rail-to-rail and gives <20ns rise and fall times into a 1nF load, with almost zero shoot-through.

 

Please, let's see it.

**********

I just realized my possible solution is a magic smoke releaser. Both transistors will turn on at the same time. In addition it also inverts the signal. Let me think about this....

OK, how about this.

 

The output on the high side is not to rail, and there are applications that need it. The only reason I have collector resistors is to minimize shoot through.

Consider this, if I were to connect the output of the 555 to a PNP transistor it would stay on even in the high state. This is a good example of the problem.

The other thing is I want to maintain the wide power supply characteristics.

Doesn't the CMOS version (e.g., ICM7555) do those things with a little loss in drive current capability? John

 

Actually it is a big loss in drive current capability. 200ma max is a hard act to follow for CMOS.


*****************

OK, I take it back, it can handle 100ma. Just looked at the datasheet.

I already have 100's of 555s though. Have I mentioned I really like this chippie?

While I haven't stated it, this will be used in my function generator eventually. I might give this chip an honorable mention though, and eventually buy a few.

The other reason I started this thread is I'm working on a 555 series for the AAC eBook projects. This might be used there for a driver chapter. Come to think of it, the function generator falls under that catagory too.

 

Please, let's see it.

This circuit builds in some dead time to prevent shoot through. There may be an IC that does the same thing. The output swings rail-to-rail with input from a bipolar 555, and simulated rise and fall times are less than 20ns into a 1nF load, and around 100ns into 10nF.
I included the .asc file just in case anyone wants to simulate it in LTspice.

 

This was a scheme I came up for something similar on a class D amp...

This is also using the bad function generator I came up with.

 

According to Intersil's datasheet for their ICL7555 Cmos 555, the ordinary 555 has a supply shoot through current spike of 400mA. Then it can switch a 200mA load high or low quickly.