This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TLV3601 PSPICE model output error?

Other Parts Discussed in Thread: PSPICE-FOR-TI, TLV3601

Hi, i realize TI will not directly answer questions re: model operation in LTspice. I am waiting for approval from my IT office to install PSPICE-for-TI... in the meantime i'm doing some simulation in LTspice. I created a model for the TLV3601 from the 'tlv3601.lib' but the output is not performing as expected. I'm operating from a 2.4VDC supply... with the input conditions, i expect to see output pulses from 0 to 2.4VDC, however the output is sitting on a VCC/2 bias and the pulse amplitude is millivolts (see attached image). 

The below comment appears at the beginning of the .lib file:

* INCORRECT INPUT CONDITIONS: If an error condition occurs, such as
* incorrect supply voltages or exceeding input voltage range, the output will go to
* half the supply voltage. The real device will NOT do this.

I think this may have something to do with the erroneous results i'm seeing? the fact that the output IS switching at the expected times, just not at the correct amplitude, leads me to believe it's working. albeit with this issue. My inputs are within the range specified in the datasheet. Does anyone know if this is in fact a problem with the model code? can this be fixed?

thanks in advance,

  • Hi John,

    Thanks for asking. What are your input conditions? Correct me if I am wrong but it looks like your Vin (green) is 3V, which is above the specified input range max limit (VCC of 2.4V + 0.3V). 

    Can you show a screenshot of your schematic? 

    Thanks,

    Chi

  • Hi Chi,

    thanks very much for getting back to me. The Vin (green) is not an input to the comp. a snapshot of this part of the circuit is attached. both inputs (Vsaw and V_err)  are limited to 0 - 2.4VDC. I've also attached a close-up of the waveforms. As you can see, the output (Vpwm) IS switching (a little) when it should.... but it's biased on Vcc/2 and the amplitude of the pulse is very small.. it should switch between VCC (2.4V) and GND. i think there's an error in the model that has something to do with the note above...?

    thanks again!

    John

  • Hi John,

    Thanks for sending me the waveform and schematic. I can see what you mean. Usually the output goes to mid supply when you're violating the input range or the supply voltage range limits. But in this case, I don't see that you are. Do you see still this issue when you disconnect the comparator circuit from the rest? i.e, disconnect PWM output and look at whats happening. It looks like you're pulling the PWM output to a sine wave battery which may cause issues. I've replicated your comparator circuit in TINA (see figure below) and I do not see any issue with the model. The output was able to switch between 2.4 and GND. 

  • thanks Chi... that looks like the same circuit i'm using. Could this be an issue with LTSpice incorrectly interpreting the model code? there's a note at the beginning of the model indicating this mid-scale behavior is just an artifact of the model and not really part of the hardware. I'm wondering if that section were commented out if it would operate nominally, but i'm not sure where that part of the code is. I may try some experimenting, but let me know if you have any thoughts.... again, many thanks for confirming the model works in TINA as expected

  • Hey John,

    That is a possibility because there may be some differences between the PSpice and LTspice programs which might cause the model to not work properly on LTspice. Our models are really designed mainly for TINA and Pspice. And that's right, we include that mid supply error condition just so users know they're operating outside of input or supply limits (but of course the real device doesn't do this). This condition is included in the output_stage subckt. 

    Would you be willing to try to use TINA in the meantime while you get Pspice for TI approved? https://www.ti.com/tool/TINA-TI

    -Chi

  • Hi Chi,

    I think ultimately, we'll move to PSPICE, but I would incur the same delays getting IT to approve TINA, as i'm now facing with PSPICE-for-TI, so i'm stuck with LTSpice for awhile. In the meantime, i reviewed the model code and made a little progress but haven't solved the problem yet.

    I changed the below line of code:

    EVMID VMID 0 VALUE = { ( V(VDD) + V(VSS) )/2 }

    to this:

    EVMID VMID 0 VALUE = {V(VSS)}

    and that removed the VCC/2 bias, so now the output returns to zero.... but did not change the amplitude of the output pulses. As you can see in the attached... the amplitude is ~40mV instead of 2.4V. Any thoughts why that might be happening? is there a disabled output buffer or something that might be limiting the output voltage? i tried increasing the output impedance several orders of magnitude, but that didn't help. I notice there is reference in the code to SHUTDOWN, even though the part has no shutdown pin.

    Note.... i know virtually nothing about SPICE code, so i'm just trying to feel my way through this...:)

    again, thanks much for the helpful insights.

    John

  • Hi John,

    I've made changes to the model which essentially removed this mid supply condition. I've tested in TINA and PSpice and it works. Try this one out and let me know if it works for you. Hope it goes well! See attached.

    TLV3601.LIB

  • thanks Chi... there must be an incompatibility with LTSpice and the TI tools. I've attached some plots for what the output looks like in LTSpice. the 2nd image is a close-up of a section of the 1st image. im not sure if this helps narrow down what might be issue. I'll take a look at the edits you made.

    Again, thanks very much for working this with me...

    John

  • i think there are 2 separate issues... (1) the mid-scale offset and (2) the output only changes by 40mV. I believe we can solve the first, but not sure if you see anything in the code indicating what is causing the 2nd?..... let me know, thanks!

  • Sorry John, this must be frustrating. Like I mentioned, we don't support LTSpice so its difficult to know what the issue is. I'd suggest working with ADI and ask them to help with this issue in the meantime. Hopefully, IT approves PSpice for TI for you soon.

    Just to confirm, you still see this issue when you isolate the comparator from the rest of your circuit, is that right? Just wanted to make sure you its nothing from the battery thats connected to your output.

  • Hi Chi,

    yes, the issue persists even when the comp output is disconnected. thanks very much for the help, hopefully i can get PSPICE installed soon.

    John

  • Hi John,

    I will now close this thread but please create a new thread if your issue persists in PSPICE.

    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)