Setting the testing potential window is completely depends on the redox reaction of the active materials. For the case of Sn, the alloying reaction takes place at 0.25 V vs. Li, so you can reduce the same in both sides or must be in upper potential  of the cathode (in your case to 2.55). But, this approach may not applicable for conversion anodes owing to the huge polarisation. Moreover, I don't understand, what is the point of testing the Sn for such wide potential window?. You can limit it for 1 V vs. Li range for the mass balance.

As an Electrical Engineer in the field of batteries, I always thought that stability and safety was the main constraint. So, if your electrodes are stable at higher voltages  then I dont see any reason why you shouldnot try it.

Similarly, in the case of discharge the constraint is the lowest voltage where the reaction can be reversible. Some literature suggests discharge voltages of 3v/cell for Lithium. I have tested it successfully down to 2.5-2.7v/cell for Li-Pol BUT my batteries were new. Some of my results can be see here: https://www.researchgate.net/publication/254708421_Discrete_linear_constrained_multivariate_optimization_for_power_sources_of_mobile_systems

Thesis Discrete linear constrained multivariate optimization for po...

Your old range for the half cell Li/Li2S (Pot. Window 2.8-1.8V) must shift (for the full cell) to about  : 1.5 to 3.0 V. This change is due to the replacement of the anode (the LOW Pot. electrode, the new "reference"). For example, if you electrochemically stress (the full cell) to a voltage well below 1V, then you risk many scientific and safety issues, like some new Li(-metal and other) complex phases formation on the surface of the electrodes, etc.

I agreed with Stelios, safety is the prime priority. but voltage window depends on the inherent properties of the electrode and stability of the used electrolyte.

Setting the testing potential window is completely depends on the redox reaction of the active materials. For the case of Sn, the alloying reaction takes place at 0.25 V vs. Li, so you can reduce the same in both sides or must be in upper potential  of the cathode (in your case to 2.55). But, this approach may not applicable for conversion anodes owing to the huge polarisation. Moreover, I don't understand, what is the point of testing the Sn for such wide potential window?. You can limit it for 1 V vs. Li range for the mass balance.

Q1: For the sulfur cathode: According to the 2015 review by Wild et al. (doi: 10.1039/c5ee01388g), potentials (vs. Li/Li+) >2.3 V are necessary for the formation of S8, with a plateau at ca. 2.7 V. Potentials

I agree with Vanchiappan, the potential range you choose depends completely on the positive and negative active materials, which you use in your system of choice. 

I am not too familiar with Li-S systems, but a state of the art Lithium-Ion Cell with a graphite/NMC chemistry is cycled in voltage range of 3.0-4.2V. A wider voltage range, especially higher anodic cut-off voltages (>4.2V) may deteriorate the stability of the layered cathode as well as the electrolyte.

Please mind that the term potential may only be used in a system in which you are using a reference electrode. In full cells you can typically only see the potential difference (i.e. voltage) of the cell. You may use a three electrode cell with a Li-metal (quasi-) reference in order to get an idea about the true potential of your working and counter electrode. This would help you to find you to find your cut-off conditions in a full cell, which ideally should include the full redox process of anode and cathode.

Good luck :)

First, I would like to thank all of you for your comments. 

Answer to some of your concerns about the stability window of my electrolyte: I am using an ionic liquid as a solvent for my electrolyte solution which are known for their wide range stability as compare to ether or carbonate based electrolyte. This is the first time i am using an ionic liquid so I am not sure but i read that they are stable upto 6 V. However, they are quite sluggish when it come to the mobility of Lithium ions, I guess due to their viscosity.

My first experiment went horrifically wrong. I started by charging my cell upto 4.5V as Li2S needs high activation energy to be converted to sulfur, I guess during this time i also lithiated tin to balance the charge. While discharging i got negligible capacity and the cell dies. In my opinion the culprit is ionic liquid which in my case is n-butyl-n-methylpyrrolidinium bis(trifluoromethanesulfonyl)imide. The electrolyte has been used for Sn/NMC full cell before. Other possible reason for the failed experiment would be mass imbalance of the active materials on the cathode and anode side. 

Is there any guideline paper which discusses these technical issues related to the preparation of full cell? If not, its good to write one for some citations. 

Refer my numerous paper on full-cells (in web), but not Li-S system

I think there are already good pieces of advice in the other answers, but due to your complicated system, even those don't cover the problem completely. So let me try again:

Q1: As mentioned, the potential window for a full cell depends on the electrode materials and the balancing of your materials.

Q2: see Q1

Q3: What is your OCV, when you do this measurement. If you start with full Li2S (1.8V? vs Li) vs. empty Sn (~2.7V vs. Li) it should even be negative if you connect tin as the anode.

In the end, when your Li2S is just sulfur (~2.2V vs. Li) and your tin is fully charged (0.4V vs. Li), it ends a (+1.8V).

So the measured potential is always the difference of your anode and cathode potential.

But since the potential range of tin is so large this values can change: If you overbalance tin it might only get charged until e.g. 0.7V when your cathode is empty, and you get a potential in the end of +1.5V.

So the best way is to take the absolute voltage vs capacity curves of your measured anode and cathode materials and calculate the resulting curve. The cut-off potentials follow out of this (for example not overcharging tin to avoid lithium plating).

An easier way: Since you have a liquid electrolyte, use a lithium reference, overbalance the tin slightly and set the cut-off potentials in between the reference and your cathode (Li2S) exactly how you did in the half-cell. Then you can record your full-cell potentials in this cell if you can record also the reference to anode potential. (depending on your potentiostat)