1: \begin{abstract} The recently developed method to solve the Faddeev

2: equations in coordinate space is used to study the weakly bound halo

3: nucleus $^{3}_{\Lambda}H (\Lambda + n + p)$. 

4: The long distances are treated carefully

5: to achieve convergence and high accuracy.  We use several sets of

6: two-body interactions which reproduce the deuteron properties and

7: provide the low-energy $\Lambda$-nucleon scattering data close to that

8: of two of Nijmegen potentials. We show that the details of the

9: potentials are unimportant unless the accuracy of the hypertriton

10: binding energy is required to be better than 50 keV.  We find that the

11: most significant parameter of the $\Lambda$-nucleon interaction, the

12: singlet s-wave scattering length, must be within 10\% of 1.85 fm, when

13: the $\Lambda$-separation energy from the deuteron is about 130 keV.

14: Other details of the $\Lambda$-nucleon interaction are less important

15: for the hypertriton structure. The scattering length and effective

16: range are computed for scattering of a $\Lambda$-particle on a

17: deuteron. The folding model reducing the three-body problem to a

18: two-body problem is investigated in this context and found to be

19: inadequate. \\

20: 

21: \end{abstract}

22: