TLV3604: About using negative input signal

Thanks for posting your question.  It is a good one.  The answer depends on how negative your input signal is.  We allow inputs to be 0.3V below GND if you are powering with a single supply.  If you have a schematic and input signal specs to share, we will review.  We also have a new device coming at the end of the year that will allow split supplies on the input structure while still maintaining LVDS compliant levels on the output.

Chuck

Here is where you can find this information in the datasheet (table 6.3 Recommended operating conditions)

Here is my schematic. The input signal ranges from +100mV to -1V. Its compared against a -200mV input.

Sorry, your schematic didn't come through but that is ok.  With a signal that goes down to -1V, you will need to clamp the voltage with a schottky diode most likely.  Can you share any details about the source driving the inputs?  It can't be level shifted?  What is the nature of the signal?  Is it repetitive in nature or random?  what is the source output impedance?  Sorry for all the questions.  

Here it is.

sorry, the way you are uploading I am unable to view it

can you just cut and paste a snip of the circuit or upload as a PDF

this may not really be necessary since what I really need to know more about are details about the input signal (more than just min and max levels).

Chuck

Here is the schematic and the output that I got from Pspice. I have used the spice model for TLV 3604 provided in TI's website. the red line represents the signal applied at the negative input. it is repeatative and expected to get as high as -1V. It compared against a dc voltage of  -200mV (the blue line). The differential output is represented by the green line. Can't I just use a dual supply DC source and connect the Vcc with 3.3 to 5 V and VEE to - 3.3 to -5 V to achieve this result in real?

Unfortunately you cannot because the device is limited to a supply range of 5.5V.  This means that VCC - VEE needs to be less than 5.5V.  For example, you could power VCC at 4V and VEE at -1V and that would make your input range down to -1V inside the recommended range.  The problem with doing this on the TLV3604 is that the LVDS output is relative to VEE.  So if VEE is -1V, the LVDS common mode output voltage will be -1V + 1.2V (-0.2V) which is not LVDS standard.  

Back to your input waveform.  You mentioned that it is repetitive but it sounds like the amplitude swing varies.  If the input isn't too far from 50% duty cycle (from the simulation it looks like that may not be the case), you could ac-couple the input and set the common mode to your reference level on device side of the ac-coupling capacitor.  But if your duty cycle differs greatly from 50%, this trick of level shifting is not a valid option.  

As I mentioned, by the end of the year, we do have a device where you will be able to run the input stage with split supplies and the output will still be ground referenced.  But this device will not be available on the web until late December.  If you have someone from TI that calls on your company, they could request samples earlier.

Chuck

Can you suggest me any fast op-amp to invert the input signal and make it a positive going one without any distortion, keeping the time parameters intact? Thank you.  

Hello,

As Chuck mentioned, there will be a device available in late December that meets your needs. A comparator would be best suited for your application. We can assist you on designs involving comparators, is that what you meant to ask in your follow-up?

Kind Regards,

Joe

All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and is subject to TI’s Important Notice (www.ti.com/.../important-notice.shtml)

As far as high speed amplifiers, we recommend that you take a look at the OPA858.

Kind Regards,

Joe

All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and is subject to TI’s Important Notice (www.ti.com/.../important-notice.shtml)

What about using OPA659IDBVT? I have experience working with that. Is there any other upgrade of OPA659IDBVT with a same footprint? I have printed PCBs compatible with OPA659IDBVT footprint. Thanks for the suggestion.

Hello,

I am reassigning this thread to the high speed amplifiers group so that they can better assist you. If you have any additional questions about comparators please respond to this thread or open a new one.

Kind Regards,

Joe

All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and is subject to TI’s Important Notice (www.ti.com/.../important-notice.shtml)