|
Reply for mt7344
I would like to reply to our friend mt7344.
( [( r" y+ I+ t+ }5 n* U/ r: }2 ^. q: n" R
They are multiple factors that you need to consider! K0 h* K( o) [; j! `- i, E! a+ k: ?: e
(1) your requirement of Bandgap accuracy. Is it a 1% or a 10% bandgap? This actually depends on how you are going to use the Bandgap reference.& P/ o/ y( Q0 }. ^2 a" @
(2) Process variations would be another factor, this is what out of our control. You can "Monte-Carlo" your circuit across the Vcc and process corners to determine your trimming ranges, A8 @$ h7 v; S
(3) What you should control are the circuit components matching. Say the circuit posted by Mr Leonken01, matching of Q0, Q1; M0, M1, M2 and all the R are extremely important./ L3 W, K! _3 H, O3 U$ k: O
(4) Circuit structure also important, as the circuit mention above, it would be tough to match the current legs of M0, M1 and M2. Unless you need the low voltage Bandgap, I would not recommend this circuit5 n2 z8 V! y- M0 Y# C# \
(5) On top of all these, your trimming range should also cover the TC, second order effect# ^. Y4 P; k9 X
will show up across temperature: `6 g* z: b% V( m/ f8 y+ t1 \
X2 M C# S- U/ C" F0 R
then you decide your trimming range and PAD needed.' @* }- E- U7 i2 I
* B+ g0 d- n* d) q7 M4 d7 u5 ~
Hopefully these help. |
評分
-
查看全部評分
|