| Author |
Message |
firsttimedesigning
Joined: 04 Jul 2007
Posts: 97
|
 11 Dec 2007 4:03 Bandgap Reference |
|
|
|
| So I am trying generate the bias voltage of an op-amp. So far the only circuit I know that can be used to generate the bias voltage is bandgap reference. But bandgap reference can only generate a voltage that is around 1.4V. I need a bias voltage that is higher than 1.4V...What circuit should I use?
|
|
| Back to top |
|
 |
renwl
Joined: 26 Apr 2004
Posts: 455
Helped: 23
Location: shanghai,china
|
| 11 Dec 2007 8:04 Bandgap Reference |
|
|
|
you can use LDO.
the output voltage will be :
vout=(1+r1/r2)*vreg.
so you can get the voltage by adjusting the r1 and
r2.
|
|
| Back to top |
|
|
firsttimedesigning
Joined: 04 Jul 2007
Posts: 97
|
| 11 Dec 2007 10:00 Re: Bandgap Reference |
|
|
|
| yeah but when i fabricate the chip, the resistance will vary within 20%...a 4k resistor might only have 3.2k resistance. the result wont be very accurate...i need something that can generate a very accurate bias voltage...something that only have error within 0.1%...
|
|
| Back to top |
|
|
renwl
Joined: 26 Apr 2004
Posts: 455
Helped: 23
Location: shanghai,china
|
| 11 Dec 2007 11:05 Bandgap Reference |
|
|
|
you are wrong.
r1/r2 means the accuracy depends on the relative accuracy of the resistors.
so you can get the result you want.
|
|
| Back to top |
|
|
firsttimedesigning
Joined: 04 Jul 2007
Posts: 97
|
| 11 Dec 2007 11:34 Re: Bandgap Reference |
|
|
|
i dont get it...let's say i want r1/r2 to be 2....i set r1 to be 4k and r2 to be 2k...
but the actual resistance of r1 is 3.2k and r2 is 2.4k....then the ratio is not 2...rather it is 1.3333....how is that going to give me the result that i want?
|
|
| Back to top |
|
|
renwl
Joined: 26 Apr 2004
Posts: 455
Helped: 23
Location: shanghai,china
|
| 11 Dec 2007 14:18 Bandgap Reference |
|
|
|
if you implement the resistor in silicon IC.
the varation of the resistance will be the same.
if r1 increase 20%, the r2 will increase 20% too.
the relative accuracy of resistor can give 0.1% requirement if you design and layout it carefully.
|
|
| Back to top |
|
|
safwatonline
Joined: 19 Nov 2005
Posts: 1347
Helped: 164
Location: EGYPT
|
| 11 Dec 2007 15:28 Bandgap Reference |
|
|
|
| use current mirror to copy the current from the band gap to another resistor
|
|
| Back to top |
|
|
gpwu
Joined: 08 Dec 2007
Posts: 48
Helped: 4
|
| 12 Dec 2007 0:54 Re: Bandgap Reference |
|
|
|
You can match 2 resistors better than 1% if you use common-centroid layout.
Please "reference the art of analog layout" book, chap.7 about common-centroid layout.
|
|
| Back to top |
|
|
firsttimedesigning
Joined: 04 Jul 2007
Posts: 97
|
| 12 Dec 2007 3:17 Re: Bandgap Reference |
|
|
|
| thanks for replying...i got it now..ok i will use LDO....thx a lot
|
|
| Back to top |
|
|
qutang
Joined: 25 Oct 2004
Posts: 315
Helped: 6
|
| 13 Dec 2007 7:26 Bandgap Reference |
|
|
|
| why u are using a bandgap bias the opamp?
|
|
| Back to top |
|
|
horsehill
Joined: 07 Dec 2007
Posts: 4
|
| 13 Dec 2007 8:18 Bandgap Reference |
|
|
|
| the bandgap circuit can get 2.5v.
|
|
| Back to top |
|
|
firsttimedesigning
Joined: 04 Jul 2007
Posts: 97
|
| 13 Dec 2007 13:37 Re: Bandgap Reference |
|
|
|
thx for replying...
to Qutang...bandgap reference is the only circuit that i know...are there temperature and power supply independent circuits that i can use to bias the op-amp?
|
|
| Back to top |
|
|
firsttimedesigning
Joined: 04 Jul 2007
Posts: 97
|
| 14 Dec 2007 10:02 Re: Bandgap Reference |
|
|
|
| can any explain how to design a LDO? i just dont get it....
|
|
| Back to top |
|
|
lladnar23
Joined: 20 Mar 2006
Posts: 211
Helped: 31
|
| 15 Dec 2007 5:59 Bandgap Reference |
|
|
|
| Why don't you just gain up the bandgap voltage? It's commonly done. Check out the brokaw circuit.
|
|
| Back to top |
|
|
lijianheng
Joined: 29 Sep 2006
Posts: 130
Helped: 10
|
| 15 Dec 2007 6:55 Re: Bandgap Reference |
|
|
|
| use a close loop buffer or charge pump if it has a clock.
|
|
| Back to top |
|
|
jnuhope
Joined: 10 Apr 2007
Posts: 50
Helped: 4
|
| 19 Dec 2007 4:18 Bandgap Reference |
|
|
|
| If for biasing opamp, you don't need "absolute value" of bias voltages, current mirror is a better way to generate your bias voltages for opamps. Razavi's book Ch.6 and Ch.9 have detail explainings.
|
|
| Back to top |
|
|
drabos
Joined: 03 Jan 2006
Posts: 431
Helped: 26
|
| 19 Dec 2007 7:49 Bandgap Reference |
|
|
|
| what about beta multiplier
|
|
| Back to top |
|
|
hung_wai_ming@hotmail.com
Joined: 05 Jan 2004
Posts: 364
Helped: 39
|
| 20 Dec 2007 15:34 Re: Bandgap Reference |
|
|
|
Why need LDO ?
It doesn't make any sense to use LDO as bias boltage generator,
as you know, LDO need large cap compensation.
I think you mis-understood what the LDO is
Bandgap, with an op-amp and feedback resistor divider will have a gain >1, so
you can get any voltage by resistor divider ratio.
|
|
| Back to top |
|
|
jnuhope
Joined: 10 Apr 2007
Posts: 50
Helped: 4
|
| 22 Dec 2007 23:09 Re: Bandgap Reference |
|
|
|
| firsttimedesigning wrote: |
| So I am trying generate the bias voltage of an op-amp. So far the only circuit I know that can be used to generate the bias voltage is bandgap reference. But bandgap reference can only generate a voltage that is around 1.4V. I need a bias voltage that is higher than 1.4V...What circuit should I use? |
If you are trying to bias opamp, you don't need bandgap. Just use current mirrors. Otherwise, how are you going to generate the bias for the opamp used in bandgap?
|
|
| Back to top |
|
|
chrisccq
Joined: 12 Nov 2006
Posts: 5
|
| 26 Dec 2007 5:18 Bandgap Reference |
|
|
|
| use the resistor network!
|
|
| Back to top |
|
|
jimmyshu
Joined: 08 May 2007
Posts: 14
|
| 27 Dec 2007 5:42 Bandgap Reference |
|
|
|
I agree with jnuhope.
Generally, opamp is biased by current not voltage, to get a constant Gm. So you need to design a current source independent with temperature and power supply.
What do you want to bias by voltage reference? The gate of current source MOS?
|
|
| Back to top |
|
|
pradeep.bhardwaj23
Joined: 18 Dec 2007
Posts: 6
Helped: 1
|
| 28 Dec 2007 13:26 Re: Bandgap Reference |
|
|
|
| please tell me also
|
|
| Back to top |
|
|
firry
Joined: 03 Dec 2005
Posts: 24
|
| 28 Dec 2007 20:50 Bandgap Reference |
|
|
|
| Bias the opamp by current is better than bias it by voltage. when you layout, you should consider the parastic of the metal and the voltage drop on it.
|
|
| Back to top |
|
|
turtleden
Joined: 15 Sep 2006
Posts: 32
|
| 02 Jan 2008 8:37 Re: Bandgap Reference |
|
|
|
| chrisccq wrote: |
| use the resistor network! |
yes, it's the easiest way
|
|
| Back to top |
|
|
linkfox
Joined: 18 Dec 2007
Posts: 43
Helped: 2
|
| 02 Jan 2008 8:48 Re: Bandgap Reference |
|
|
|
| jimmyshu wrote: |
I agree with jnuhope.
Generally, opamp is biased by current not voltage, to get a constant Gm. So you need to design a current source independent with temperature and power supply.
What do you want to bias by voltage reference? The gate of current source MOS? |
yes, one method is using beta-multiplier which has been mentioned earlier
|
|
| Back to top |
|
|
qqic
Joined: 25 Aug 2006
Posts: 75
|
| 03 Jan 2008 1:24 Bandgap Reference |
|
|
|
| use an opamp and resistor to boost it
|
|
| Back to top |
|
|
ashish_chauhan
Joined: 02 Sep 2007
Posts: 229
Helped: 37
|
| 03 Jan 2008 16:52 Re: Bandgap Reference |
|
|
|
If you are trying to bias opamp, you don't need bandgap. Just use current mirrors. Otherwise, how are you going to generate the bias for the opamp used in bandgap?
if biasing of bandgap opamp is considered ... you can use self biased opamp structure.
|
|
| Back to top |
|
|
seanyang
Joined: 12 Dec 2006
Posts: 28
Helped: 5
|
| 04 Jan 2008 8:46 Re: Bandgap Reference |
|
|
|
All right, just using the architecture like LDO form.
But must sure no cuurent drive from out pin.
Or you will have stability issue.
|
|
| Back to top |
|
|
amarendrap14
Joined: 17 May 2007
Posts: 12
|
| 05 Jan 2008 5:31 Re: Bandgap Reference |
|
|
|
generally bandgap also provide current reference (which can be 5-8%) accurate..
so best way is to bias through current(biasing is simpler and good control over power dissipation).. . also accuracy requirements for current is generally less compared voltage..
LDO are used for high current capability design (power carrying)..better use opamp-R combination if you really need voltage biasing(expect some 1-2 %) degradation due to this..
|
|
| Back to top |
|
|
northeast1
Joined: 21 Jun 2007
Posts: 150
Helped: 4
|
| 14 Jan 2008 23:49 Re: Bandgap Reference |
|
|
|
| you can use 1.4v for reference to generator higher reference voltage.
|
|
| Back to top |
|
|
