can negtive feedback reduce the ppm/K of bandgap reference?

[wanily1983]

hi, guys
can negtive feedback reduce the ppm/K of bandgap reference?
with regards
wanily

 

[sunking]

it can not.

 

[huojinsi]

Re: can negtive feedback reduce the ppm/K of bandgap referen

Vref=nVt+Vbe,Temperature Coefficient=ΔVref/ΔT,so i feel the ppm/K of bandgap reference doesn't has relation with negative feedback!

 

[rfsystem]

Re: can negtive feedback reduce the ppm/K of bandgap referen

The best results are actually achived with a noncurvature corrected bandgap and a temperature sensor built from the same bandgap. The temperature sensor is converting a*VT/VBDG into the digital domain and the temperature correction is done in the digital domain. Correction training is required at 3-5 temperature values but with process dependend presettings a similar performance as laser trimming could be achieved.

 

[wanily1983]

Re: can negtive feedback reduce the ppm/K of bandgap referen

this is the schematic, it can get 1.5 ppm/K using the standard 0.6u UMC model, but it only has first order compensation. what is impossible!

 

[|IAngel|]

Re: can negtive feedback reduce the ppm/K of bandgap referen

The best results are actually achived with a noncurvature corrected bandgap and a temperature sensor built from the same bandgap. The temperature sensor is converting a*VT/VBDG into the digital domain and the temperature correction is done in the digital domain. Correction training is required at 3-5 temperature values but with process dependend presettings a similar performance as laser trimming could be achieved.

+1

Using the same Bandgap as a temp sensor and putting it into the digital domain and using that as a controller can generate a super accurate bandgap.