Ways for shifting pulse levels

[uvaraj]

Hi all,

I have to shift my pulses(0 to 3.3V) to pulse(-3V to +5V) to drive Electro-Optic Modulator. I tried with comparator but changing the rail voltage is not a advisable.

Pulse width - 1us (rise time-2ns) and pulse repetition rate is 100us.

So could anyone suggest any other way of doing this.

Thanks in advance.

Uvaraj.

 

[permute]

Re: Pulse level shifting

do you have either +5V or -3V. Or really, voltage other than +3.3V?

 

[uvaraj]

Re: Pulse level shifting

sorry permute i can't reach your question.:-|

1. I am having the pulse changing from 0 to 3.3V and now i have to shift the level of the pulse to change between -3 to +5V.

2. And i am having these voltages(+ve and -ve variable voltage regulators) in my board.

 

[permute]

Re: Pulse level shifting

you can use the chained open-collector type output then.

eg, the 3.3V signal drives an N-FET. the drain connected to a small resistor (eg 300R) to 5V. this node connects to the gate of a P-FET. the source is connected to +5V, and drain is connected by a small resistor (eg 1000R) to -3V. The drain/resistor node of the PFET is the output.

devices should be sized appropriately. the FETs must have a low enough RDS on to drive the resistors. the resistors must be low enough in value that the RC time constant associated with the circuit is not significant. very low values will waste power. high voltage (8V) cmos parts can be used past this point if buffering is required to drive other, resistive loads.

there are variations on this theme.

other methods exist.

 

[uvaraj]

Re: Pulse level shifting

Thank you very much for your idea. I will try this.:grin:

Will u please tell some other methods? so that i can know.

 

[permute]

Re: Pulse level shifting

you can also use RCD networks. Or more generally optos/transfomers. the latter two are more useful when galvanic isolation is also required. High-speed comparators are also an option.