How can I restrict floating point precision?

Hi,

I'm doing some experiments on Section 3.5.4 of SICP.
I'm trying to get an approximation of e with the following
code: 

(define (integral delayed-integrand initial-value dt)
  (define int
    (cons-stream initial-value
                 (let ((integrand (force
  delayed-integrand)))
                   (add-streams (scale-stream integrand dt)
                                int))))
  int)

(define (solve f y0 dt)
  (define y (integral (delay dy) y0 dt))
  (define dy (stream-map f y))
  y)

(stream-ref (solve (lambda (y) y) 1 0.001) 1000)

But it seems that Scheme48 has a very high precision
floating point arithmetic. And this integration process is
extremely slow. 

How can I tell Scheme48 to use lower precision arithmetic?

--

-- 
Wang Yin