Hi,

I am looking at the current reference proposed in the paper by Oguey and Aebischer in JSSC, July 1997, pp. 1132-1136, Fig. 2.

The big advantage of this source apart from it being purely-MOS is that the temperature coefficient of the resulting current can be usually set to zero. At least in principle, that is the case.

However, I am unable to do this for an arbitrary current level. It seems as if the zero tempco is achievable for a given typical current level only. Of course, I do not understand the dynamics of the tempco in this paper yet.

Could anyone shed light on how to decouple the current level of choice from zero tempco, if they have had experience with this circuit?

Thanks,

Vivek