There is a pulse shown in the picture.The pulse voltage are -2V to -5V and the width is 200ns.
Is there a component which could take this pulse and relpicate it three times?
Three outputs which are equal to the pulse i entered.
Thanks.

 

What supply voltages are available?
What are you driving?
You will need help choosing which op-amp to use.

 

Hello , I have tried to use buffered opamp configuration of AD8034 as shown bellow and in attached ltspice.
As you can see my rise and fall time are getting worse.
is there a way to make the edges sharper?
Or if there is a digital component which will do a good job its also a great solution .
My goal is to have -2 to -5 200n width input and gat the same pulse on the output as well but with rise and fall time as fast as possible.
maybe a flip flop or a latch could do the job?
Thanks.

 

There is a pulse shown in the picture.The pulse voltage are -2V to -5V and the width is 200ns.
Is there a component which could take this pulse and relpicate it three times?
Three outputs which are equal to the pulse i entered.
Thanks.


View attachment 327155

Equal in what way?
What tolerances for delay, skew, risetimes, and amplitude?
You could use ECL or CML with a negative supply.

 

Hello, equal like the input pulse.
delay there is no tollerance.amplitudes from -5 to -2
Rise and fall time are as small as possible.As i showed befor i tried to implement what ronsimpson said , but my rise and fall got worse .
I could stay with that but if you have better compononet maybe digitalrecommendation it will be great.
Thanks.

 

An important question:
Are you looking to buffer (regenerate) a digital input signal? or an analog input signal?

 

Hello eetech00,its an analog signal,-5 to -2 rise as shown below. I need to bufferr it and to try make the edges as sharp as possible.
Thanks.

 

I need to bufferr it and to try make the edges as sharp as possible.

Buffering implies you want 3 copies of the same signal.
Make the edges as sharp as possible is different. It implies you want sharper edges then you now have.
There is a big difference.

Before we put any more time into this. We need to know what you are driving. How much current? Will there be coax cable? We need more information.

 

So you are starting with a 0,3v digital signal and converting it to -5v,-2v?

 

Eetrch00 , yes .
Robsimpson ,it drives at one end a gab transistor with almost zero gate current .

And other out put goes to be an input to
LM339 comparator. Data sheet shown below.
https://www.ti.com/product/LM339?bm...ywczLRUACvTLUTG2yGaUFT4qTvGhzTfGlVC3Q_FCvzMb0

 

Your odd mixture of analog and digital signals is confusing.
Why do you need analog?

 

Eetrch00 , yes .

So, conceptually, something like this:

 

Hello eetech00, so basickly its a comparator I have tried to use the following LT1016 comparator.Ltspice file is attached
When 1>0 output is -5
When 1<3.3 output is -2
but on the output i get straight -1 .
Where did I go wrong?
Thanks.

 

Hello eetech00, so basickly its a comparator I have tried to use the following LT1016 comparator.Ltspice file is attached
When 1>0 output is -5
When 1<3.3 output is -2
but on the output i get straight -1 .
Where did I go wrong?
Thanks.

View attachment 327190
View attachment 327191

Why not use an analog switch like I showed in post #12

 

Hello ,because analog switch to me is the same as comparator.why my lt1026 idea implementation is not working?
Thanks.

 

First lets fix your spice file.

You have the plus supply at -2V. The inputs are "+"=1V and "-"=0 to 3V. The IC works if the inputs are 1.25V above the - supply and 1.5V below the + supply. Your inputs are above the + supply so it will not work at all.

Change the + supply voltage from -2V to +5V. Now it will work. I know the output is not what you want but you can see the circuit function.

200pF load on the output will slow down the comparator.

The output of the IC pulls up by the + supply. Not all the way to the + supply.
The output of the IC pulls down by the GND pin. Not all the way to gnd supply.
Green trace is the "-" input. Red is the "+" input Blue is "out".

 

Hello ,because analog switch to me is the same as comparator.why my lt1026 idea implementation is not working?
Thanks.

No. It is not.
Per post #9,#10, you are basically making a logic level translator. But I guess how it will be used is a secret.

 

The important specs so far are ;
Tr = 20 ns
Vo=3Vpp
Load = TBD
I assume you can provide any supply voltages necessary.

Pls advise

 

Assuming you have the waveform in post #1. I would buffer it with a SN74LVC1G17. Which is something like a sn74xx17 but only one in a 5-pin package. It has about 4nS delay and 2.5nS edges. It will work with supply voltages of 1.65-V to 5.5-V. I am using them in production. I use the SOT-23 package because I can hand solder it. As shown below the buffer will pull down to -5V and up to -2V. It has 24mA of drive. I know the 2V Zener is a problem, but I know how to fix that. This is not a analog buffer but a digital buffer.

 

What about AC series CMOS running "upside down"? The TS hasn't said anything about propagation delay, so 1/6 of a hex inverter as the input stage, driving three more inverter inputs. Nanosecond edges, stiff output stages. I think this is a less finnicky option than those in #4.

ak