hi,all
I need a constant current reference in my circuit, it's value is about 1uA.I summed a PTAT current from bandgap and a CTAT current to achieve this target, it's TC performance is ok, but the resistance's value will departure 20%, so the current value will change at different corners.Anyone has an idea to design a current reference which only relevant to resisitance's ratio more than it's absolute value.Any advice is welcome,thanks!
I think, the below paper might be useful for you. It's not exactly the same you wanted. But a current reference with low variation across process corners. Most importantly it's implementable. The paper has some minor errors in explanation and deduction of some conditions/settings for the circuit (I didn't remember the exact things). So just use the architecture and tweak the parameters (kind of trail & error) such that you get good response.
_____________________________________________________________________
A New Low Voltage Precision CMOS Current Reference With No External Components
Rasoul Dehghani and S. M. Atarodi
Abstract—A novel current reference with low temperature and
supply sensitivity and without any external component has been
developed in a 0.25 m mixed-mode process. The circuit is based
on a bandgap reference (BGR) voltage and a CMOS circuit similar
to a beta multiplier. An NMOS transistor in triode region has
been used in place of a resistor in conventional beta multiplier to
achieve a current which has a negative temperature coefficient and
only oxide thickness dependent. The BGR voltage has a positive
temperature coefficient to cancel the negative temperature coefficient
of the beta multiplier. The simulation results using Bsim3v3
model show max-to-min fluctuation of less than 1% over a temperature
range of 20 C to +100 C and a supply voltage range of
1.4 V to 3 V with 30%tolerance for all of the used on-chip resistors.
The maximum current variation is slightly less than the oxide
thickness variation in the process corners.
____________________________________________________________________
You can download the paper from:
hi,ipsc
This paper i have read already.The circuit reported in this paper is somewhat complicated, it uses two folded-cascode two stage opamp, and a bandgap voltage reference.The circuit's performance i simulated is not very good, variation at all corners is very large, additionally, the power consumption has exceed my assignment.
The reason for keeping constant Ibias to make circiut work more reliably.. There are three ways, make contact-current-source-circuit real "constant".. or make circuit less susceptible to changes in Ibias (by "over-designing") or do both of them ...
hi,ipsc
This paper i have read already.The circuit reported in this paper is somewhat complicated, it uses two folded-cascode two stage opamp, and a bandgap voltage reference.The circuit's performance i simulated is not very good, variation at all corners is very large, additionally, the power consumption has exceed my assignment.
Hi caosl,
I think, then the only way is to go for "BGOUT+I/V converter with external resistor". Because among the 'complete onchip constant current source' papers, I have seen so far, this is the only fair good one. But, I might have missed some. See if you can find one. ALL THE BEST.
hi,ipsc
There is another paper talked about current reference,
"A New Compact Temperature-Compensated CMOS Current Reference",IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS-II:EXPRESS BRIEFS,VOL.52,NO.11,NOVEMBER 2005.
Mora's book "voltage reference" has some related discussions too.
hi,ipsc
There is another paper talked about current reference,
"A New Compact Temperature-Compensated CMOS Current Reference",IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS-II:EXPRESS BRIEFS,VOL.52,NO.11,NOVEMBER 2005.
Mora's book "voltage reference" has some related discussions too.