| Author |
Message |
caosl
Joined: 26 Mar 2006
Posts: 212
Helped: 25
|
 03 Aug 2006 5:28 Trimming of bandgap reference |
|
|
|
|
hi,all
I have a doubt about trimming of bandgap reference which is showed in attached schematic.The R2 and R3 should be trimmed at the same time or only to trim R2 is enough.
For trimming convenience, i tried to combine the M1&M2, but the perfermances were difficult to tradeoff nice.
By the way, the opamp's bias current was controlled by itself.Because of starup circuit exists, this is not a problem except for minimum startup power supply voltage.
|
|
| Back to top |
|
 |
gevy
Joined: 17 Nov 2004
Posts: 377
Helped: 41
Location: Russia
|
| 03 Aug 2006 8:03 Re: Trimming of bandgap reference |
|
|
|
|
| To trim R2 is enough as output resistance of transistors lot more R2 and R3.
|
|
| Back to top |
|
|
caosl
Joined: 26 Mar 2006
Posts: 212
Helped: 25
|
| 04 Aug 2006 6:58 Re: Trimming of bandgap reference |
|
|
|
|
hi,gevy
Thanks for your reply.I felt that only to trim R2 is adequate,but the mismatch of R2 and R3 will has some impact on the performance of the circuit.Now,how extent it is going to.
|
|
| Back to top |
|
|
DenisMark
Joined: 16 Sep 2005
Posts: 314
Helped: 64
Location: Russia
|
| 04 Aug 2006 9:35 Trimming of bandgap reference |
|
|
|
|
| If u connect upper contacts R2 & R3 together and connect them to Vref thought R4 than by trimming R4 u can change Vref more precisely. One more advantage u have got that the value of R4 is smaler than R1,R2,R3.
|
|
| Back to top |
|
|
caosl
Joined: 26 Mar 2006
Posts: 212
Helped: 25
|
| 04 Aug 2006 12:00 Re: Trimming of bandgap reference |
|
|
|
|
hi,DenisMark
The structure i have tried already,which showed in the picture below.Indeed,this structure can save area occupied by the resistor,but the performances i simulated was not better than the first one.What's the reason give birth to this phenomena.
|
|
| Back to top |
|
|
DenisMark
Joined: 16 Sep 2005
Posts: 314
Helped: 64
Location: Russia
|
| 04 Aug 2006 12:56 Trimming of bandgap reference |
|
|
|
|
"but the performances i simulated was not better"
What parameters do u mean?
|
|
| Back to top |
|
|
caosl
Joined: 26 Mar 2006
Posts: 212
Helped: 25
|
| 05 Aug 2006 3:55 Re: Trimming of bandgap reference |
|
|
|
|
| startup up power supply voltage,static line regulation(dc psrr),transient line regulation(high frequency psrr).If the power supply voltage increase,the psrr descend greatly(90dB at 5V--->60dB at 6V).
|
|
| Back to top |
|
|
DenisMark
Joined: 16 Sep 2005
Posts: 314
Helped: 64
Location: Russia
|
| 07 Aug 2006 8:20 Trimming of bandgap reference |
|
|
|
|
| It's seem to me one resistor isn't a reason for such "prefomance" degradation. Do u understand that such circuit only situable for capacitance load or very high resistive load? May be u use this circuit for lager resistive load or switched load? Use buffer for this case.
|
|
| Back to top |
|
|
andy2000a
Joined: 18 Jul 2001
Posts: 756
Helped: 7
|
| 08 Aug 2006 8:08 Re: Trimming of bandgap reference |
|
|
|
|
I think caosl design is better , becuase 2 BJT current is "match" when you trim resistor ..
if only trim R2 , will cause another BJT current mismatch
|
|
| Back to top |
|
|
arsenal
Joined: 17 Oct 2004
Posts: 135
Helped: 11
|
| 08 Aug 2006 22:49 Re: Trimming of bandgap reference |
|
|
|
|
| caosl wrote: |
| startup up power supply voltage,static line regulation(dc psrr),transient line regulation(high frequency psrr).If the power supply voltage increase,the psrr descend greatly(90dB at 5V--->60dB at 6V). |
there must be sth wrong in the operating point of ur circuit, otherwise it will be good enough. and this circuit can utilize selfbias and consume little power such as several uA if ur starttime is not constrained.
|
|
| Back to top |
|
|
caosl
Joined: 26 Mar 2006
Posts: 212
Helped: 25
|
| 09 Aug 2006 4:10 Re: Trimming of bandgap reference |
|
|
|
|
Thanks all people's reply.The circuit i designed have reached my goal already.The architecture adopted is showed in pic2 for the sake of trimming convenience. The psrr >85dB at all corners after a pre-regulation circuit is used.The current dissipation <5uA.But there a drawback exists,namely,the oscilate is a little too much related to power pulse.Although add a cap between opamp's output and power line can alleviate it,the startup time is increase,particularly startup voltage is raise sharply.Someone has the other method to solve this problem??Thank U very much!
Best Regards!
|
|
| Back to top |
|
|
edwintsu
Joined: 24 Mar 2005
Posts: 84
Helped: 7
|
| 04 Jun 2007 4:39 Re: Trimming of bandgap reference |
|
|
|
|
I want to make bidirectional trimming.
Typically trim R3 for higher voltage.
 Can i trim the resistance R1 for lower voltage? 
If not, why?
Thanks!
|
|
| Back to top |
|
|
layes2
Joined: 03 Dec 2004
Posts: 346
Helped: 5
|
| 17 Jun 2007 4:44 Trimming of bandgap reference |
|
|
|
|
| trim current may better than res
|
|
| Back to top |
|
|
