A Suslin algebra is a generalization of a Suslin tree: it is a ccc, $(\omega,\infty)$-distributive boolean algebra.  Jech proved that every Suslin algebra has size at most $2^{\omega_1}$.  A proof is given in the current edition of his book "Set Theory."  He also mentions that it is consistent to have Suslin algebras of size $2^{\omega_1}$ but does not sketch an argument.  I have been having trouble finding a proof of this fact.  I looked in the older edition of Jech's book, and he says a little more about it there, but still no proof.  I would be very grateful if someone could point me to a paper that has a construction of a large Suslin algebra, or sketch a proof here in an answer.  I am also curious about the following:

1) Is it consistent to have Suslin trees but no Suslin algebras of density larger than $\omega_1$?

2) Can the existence of large Suslin algebras be proved from several diamonds, say $\diamond_{\omega_1}$ + $\diamond_{\omega_2}(cof(\omega_1))$?

Thanks!