Two Error messages when fitting mixed effects nonlinear model in nlme

Question

Here is my code:

> pred4<-function(a1,b1,a3,b3,Pa,Temp,D) log(Pa)+a1+b1*(1000/(Temp+273.15) - 0.81)-log(1+Pa*exp(a3+b3*(1000/(Temp+273.15) - 0.81)))+D
> nlme.fit <- nlme(ln.R. ~ pred4(a1,b1,a3,b3,Pa,Temp,Day), fixed = a1+b1+a3+b3 ~1,random = D~1|Day, data =Dat4param, start=start0)

nlme fits the nonlinear model fine if I remove the random Day effect D. However, the above code results in the following error and repeated warnings:

Error in chol.default((value + t(value))/2) : 
  the leading minor of order 1 is not positive definite
In addition: Warning messages:
1: In Ops.factor(log(Pa) + a1 + b1 * (1000/(Temp + 273.15) - 0.81) -  :
  + not meaningful for factors
2: In Ops.factor(log(Pa) + a1 + b1 * (1000/(Temp + 273.15) - 0.81) -  :
  + not meaningful for factors

etc.

Any suggestions??