Level shifter/ moving 0V to -15V

  • Thread starter Thread starter Mauvai
  • Start date Start date
Click For Summary
SUMMARY

The discussion centers on the need for a level shifter capable of converting a CMOS inverter output of 0V to +15V into a signal that alternates between -15V and +15V at a frequency of 200MHz. The user specifies constraints that exclude the use of op-amps and diodes, and ultimately finds a solution using a CMOS source follower and two CMOS inverters powered by +/-15V supplies. However, the achieved bandwidth does not meet the 200MHz requirement, highlighting the challenges of high-speed digital switching.

PREREQUISITES
  • Understanding of CMOS technology and inverters
  • Familiarity with high-speed MOSFET switching
  • Knowledge of signal integrity and transmission line theory
  • Experience with circuit design constraints in high-frequency applications
NEXT STEPS
  • Research high-speed MOSFET driver circuits for improved performance
  • Learn about H-bridge configurations for bidirectional signal control
  • Explore techniques for minimizing stray capacitance in high-frequency designs
  • Investigate alternative level shifting methods for high-speed applications
USEFUL FOR

Electronics engineers, circuit designers, and hobbyists working on high-speed digital signal applications who need to understand level shifting techniques and the associated design challenges.

Mauvai
Messages
9
Reaction score
0
Hi all
I have a CMOS inverter, from which i can produce something like 0V or 15V out - I don't think I can make it -15 to +15, which is what i need
I believe what i need is a device called a level shifter, which would go in after but any searches online have yielded very little
Can anyone even point me in the right direction? (before anyone says it, an op-amp based Schmitt trigger (or something similar), which is an obvious choice, is not an option). I don't need the whole thing done out, just some idea of what kind of circuit i need

(also diodes are not an option)
 
Last edited:
Engineering news on Phys.org
Telling us two things that are "not an option" isn't very helpful.

Tell us what are the constraints on your design that exclude those options, e.g. operating frequency, power requirements, physical size of the circuit, etc.

Otherwise, trying to answer is just a guessing game.
 
Fair answer, i was being intentionally vague because i would like to figure most of it out myself, i just hagve no idea where to start.
The circuit can contain 6 MOSFETs, + resistors and capacitors, has to operate at 200MHz, and can be any size, but the smaller the better, as it has to go onto silicon after.

Also the circuit its intended for is modeled on an ADG201A analog switch
 
Sorry about this:
Thinking further, what i have detailed above isn't actually enough - when the inverter output is high, i need both +15 AND -15V available, and when the output is low, (0v) i need the same outputs but in the opposite places, if that makes sense. Neither will need to supply any current if that's an issue.
Again, sorry - I'm even more confused now
 
You have a logic signal x (0V, 15V). You want it level-changed (-15V, +15V) and also make available its complement ¬x ?

You want this constructed using discrete high-speed MOSFETS?

A fan out of 0 is acceptable?
 
Sounds to me like an H bridge.

Digital switching between +15V and -15V = 30Vpp square wave signal at 200MHz is quite a tall order. The problem is the current needed to charge and discharge the stray capacitance.

The solution to your problem would be obvious to an experienced electronics engineer who knew what you were trying to achieve.
 
Baluncore said:
Sounds to me like an H bridge.

The solution to your problem would be obvious to an experienced electronics engineer who knew what you were trying to achieve.

This is unhelpful - if i was or had and experienced electronics engineer, i would be posting here :-/

In the end i solved it by myself - i used a Cmos source follower to boost the TTL signal to 0 to +15V, then 2 cmos inverters with +/-15V supplies. the first had a +ve switching point, bringing the signal to +/-15, and the second inverter gave the opposite output
 
But you do not get 200MHz bandwidth.

Mauvai said:
This is unhelpful - if i was or had and experienced electronics engineer, i would be posting here
There are experienced electronics design engineers on this forum. If you specified your circuit's environment rather than ideal goal posts, then an engineer could select goal posts that best satisfied the circuit's environment for the minimum power / cost.

Mauvai said:
Neither will need to supply any current if that's an issue.
If we assume it drives a 150 ohm transmission line then for 30Vpp, current = 30 / 150 = 200 mA.
If it drives 50 pF at 200 MHz then since c = i . t / v we get i = c . dv / dt = 50e-12 * 60 / 5e-9 = 600 mA.

Without I/O details it is hard to design a solution that meets the specification.
 
Last edited:

Similar threads

  • · Replies 21 ·
Replies
21
Views
3K
  • · Replies 14 ·
Replies
14
Views
2K
Replies
5
Views
2K
  • · Replies 28 ·
Replies
28
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 6 ·
Replies
6
Views
8K
  • · Replies 21 ·
Replies
21
Views
3K
  • · Replies 14 ·
Replies
14
Views
84K
  • · Replies 13 ·
Replies
13
Views
78K
  • · Replies 12 ·
Replies
12
Views
2K