TLV6703: Hysteresis range

Hi Kai,

can you show a schematic?

Kai

Hi Kai 

Thanks for your help 

I would like to know how to modify this circuit VIT+ = 6.05V ,VIT- = 5.95V

if have excel calculator ,it will be better 

Kai,

Please try this excel tool for calculating hysteresis. I'm not sure what your output high level was in this case so I wasn't able to test it out first. Please let me know if you have any feedback on the tool!

Regards,

JonnyComparator Hysteresis Tool.xlsx

Hi Jonny

Thanks for your reply and tool 

I use the tool to calculator ,it is weird

1. TLV6703 is internal reference ,my condition is below ,if I set the Vdiv voltage under 5V ,it will have warning information

2. Resistor range I set 10khom , but R3 result is 20ohm ,which one is correct 

So far our schematic is using TLV6703 (internal reference) ,could you help to double confirm how to use TLV6703

Thanks

Kai,

R3 is an input (in yellow), so the default that had showed up was 20 ohms. You can change that to whatever value you plan to use for your pullup. From there, the "resistor range" tab will calculate the remaining resistors as close to the range you select.

Did you want your VIT+ at 6V or 6.2V? I see in the screenshot you provided, you have Vh set at 6.2 and Vl set at 5.7, which is different from your initial post.

Because the part has internal reference, you can ignore whatever values it generates for R4 and R5 - I believe R4 is the trouble maker for your case that is causing the warning to show up. Changing Vdiv to less than 6V doesnt change the values it calculated for R1, R2, and R3. Thank you for the feedback and I will look into this... not sure why it is throwing out a warning for this case.

Below was what I was able to verify to work for the part:

Hi Chuck 

OK

We will check your suggestion 

Thanks

Hi Kai,

unfortunately, the PSPICE model of TLV6703 cannot be imported into TINA-TI. So, I made a simplified TINA-TI simulation:

This means the following: In your schematic choose R58 = 33k + 910R, R59 = 2k4 and solder a 10M resistor from the "sense" input (pin 3) of TLV6703 to the output (pin 1) of TLV6703.

To be able to compensate for the manufacturing tolerances of resistors and threshold voltages of TLV6703, you could use a 2k precision multiturn cermet trimmer instead of the 910R resistor.

Kai

Hi Kai

Thanks for your fully support 

Is it possible to provide the excel calculator to us ? due to we may change the VDD and pull up voltage and hys voltage range 

it will be easy to use TLV6703 

Thanks

Hi Kai,

I have found the recommended values in the simulation by trial and error. I just played a bit with the values. Deriving the correct formulas for the resistances consumes more time.

The correct formulas are:

R2 = R3 x D x (0.395V x B - 0.4V x A) / (0.4V x A x D + R3 x B x C)

R1 = (R2 x R3 x B) / ((R2 + R3) x 0.4V)

with:

A = 5.95V - 0.395V

B = 6.05V - 0.4V

C = 5V - 0.395V

D = R3 + 10k

5.95V in "A" is your lower threshold. Change it as desired. 6.05V in "B" is your uppper threshold. Change it as desired. 5V in "C" is the pull-up voltage. And 10k in "D" is the pull-up resistance.

The run of the calculation is as follows: Choose R3 arbitrarily. The higher R3 compared to R1 and R2, the lower the hysteresis. I have chosen 10M here. Then, enter R3 into the formula of R2. Afterwards, enter R3 and R2 into the formula of R1. Of course, you could create an excel sheet which does all these calculations automatically for you.

Here is the result of the simluation with the improved values:

And here is the according TINA-TI file:

kai_tlv6703.TSC

Kai

Kai

Just to be clear, when you resistor divide down your input voltage to the 0.4V reference level (approximately 15:1), that gains up the internal hysteresis of the TLV6703 by the same ratio.  So the internal hysteresis of 5.5mV becomes approximately 82.5mV without adding any feedback. 

However, to your point we should add some equations to our datasheet to help in the design of circuits that require additional hysteresis.

Chuck

Hi Chuck

1. if Vih+=6.8V ,Vil-=6V ,May I know the resistor value? this application is using for 2cell battery voltage monitor 

2. If 15:1 ratio ,the internal hys will be 82.5mV ,what's the side effect ? sorry ,i have a little confuse 

I use kai_tlv6703.TSC to calculator ,but seems like the result is not correct 

Hi Kai,

I think you have made a mistake during the calculations. Applying the formulas by myself, I get:

However, taking into account the input current of TLV6703 of up to 25nA, I would decrease these resistor values by at least a factor of 20:

Kai

Thanks Kai for your support on this.  As usual, your support is excellent.

Kai

Now that you are requesting much more hysteresis than 100mV, introducing positive feedback is required as Kai has shown.

Chuck

Thank you, Chuck!

Kai

Hi Kai / Chuck

Thanks for your fully support 

just one question 

how do you use TINA to calculate the value?

If you are using TINA, then that is a trial and error approach.  Reducing the size of the positive feedback resistor will increase your hysteresis.  However, it will also slightly impact your initial switching threshold.  It is possible to fine tune your circuit with TINA but I would recommend doing some calculations so you have an expectation for your results before running your sim.

Chuck

TLV03.xlsxhI Sir

I follow your formulas then got R1 = 11K ;R2=4.7K ; R3 =470K 

But got fail reslut after implemented to test , the pin allways Hi ,no chance goes to lo ?

Could you please teach me whats problem ?

Attached is my calculation form !!

The correct formulas are:

R2 = R3 x D x (0.395V x B - 0.4V x A) / (0.4V x A x D + R3 x B x C)

R1 = (R2 x R3 x B) / ((R2 + R3) x 0.4V)

with:

A = 5.95V - 0.395V

B = 6.05V - 0.4V

C = 5V - 0.395V

D = R3 + 10k

Toby

It appears that you are new to this thread.  Please recognize that the equations you shared have an upper threshold of 6.05V and a lower threshold of 5.95V.  I just want to make sure that you were aware of this when you ran your simulation.  If you want to change to the previous design plan of Kai which had Vih+=6.8V ,Vil-=6V, you will need to alter your equations.

Chuck

Hi Chuck

Yes  of cause : I am new on this thread.

My applicatio now have to set up the upper threshold of 6.3V and lower threshold of 5.5V!

Can you please help on this ?

Thanks

Toby