Fast Transistor Switching using Cascode configuration

Go To Last Post
8 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi guys, I was reading in a textbook called the "The Art of Electronics". Under Transistors on page 103 it talks about the miller effect (the capacitance between the base and collector), which greatly reduces switching speed (it creates an RC time interval circuit with the Collector load). It has a 3 circuits on page 868 for "beating" the miller effects, 1 of them seems to fit what I'm trying to do, its called the famous "cascode" circuit.

 

But when I try simulate it with LT spice it performs worse (as in slower switching) than a normal NPN common emitter circuit!

Please see the attached circuit of trying to switch 12V at 5MHz from a 4.5V input signal.

 

Any ideas what I'm doing wrong?

Kind Regards

 

 

Attachment(s): 

Just a noob in this crazy world trying to get some electrons to obey me.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Well, one of your issues is that the upper transistor in the cascade pair will only pull down to about 2.3V compared to 0.1V or so for the single transistor. That is inherent in the cascade config. 

 

It also appears that you are never turning off the current in the cascade pair. Use an LTspice current probe and look at the current through Q3. It looks like Q2 is not turning off, though it should. I would also check the actual base drive on Q2. There may be something happening there.

 

The base boost circuits seem to have a time constant that is way too long. 10K and 100pf makes 1us. That does not fit with a 5MHz drive (though why it would be different between the two circuits is not yet clear). I think that is the source of the changing voltage (current) while Q2 should be on.

 

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

Last Edited: Mon. Feb 22, 2016 - 12:27 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

fever2tel wrote:

Any ideas what I'm doing wrong?

 

You need to add a resistor (10K?) in series with the base of Q3.  Also add a capacitor (0.1uF?) from the base of Q3 to Gnd.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for the reply, ah yes the base resistor helped a lot!

However a capacitor on Q3 base to GND acts like a smoothing cap and slows things down.

Just a noob in this crazy world trying to get some electrons to obey me.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For cascode, the upper base really wants to be connected to a stiff source. What "improvement" are you seeing with the resistor added? 

 

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:

For cascode, the upper base really wants to be connected to a stiff source. What "improvement" are you seeing with the resistor added? 

 

With a 5MHz square wave input, the cap from base to ground provides a low impedance.

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Jim, on second thought ...

 

Here's fever2tel's cascode circuit from post #1:

 

The point of the cascode circuit is to keep the collector to base voltage of Q2 near constant.

But this inherently limits the low voltage at the collector of Q3 (Vout).

The cascode is intended as a small signal amplifier, not for logic level output signals that need to drop to ground.

 

Last Edited: Mon. Feb 22, 2016 - 11:22 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Chuck99 is correct. Cascade is intended as a high frequency small signal amplifier. It happens to work nicely as a non-inverting level translator with an NMOS FET in the Q3 position but that form does not give you particularly fast switching.

 

In the waveform in the original post, I am convinced that Q2 is NOT getting fully turned off.

 

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net