Problems about PGA(Programmable Gain Amplifier) design whose BW can be adjustable

Hi, there~
i am just doing some design about a PGA(Programmable Gain Amplifier) which should has a adjustable Band-Width.

i have read these references below

1)Flexible Baseband Analog Circuits for Software-Defined Radio Front-Ends. [J] IEEE Journal of Solid-State Circuits, Volume: 42 , Issue: 7, 2007: 1501 - 1512
2)Flexible baseband Low-Pass Filter and Variable Gain Amplifier for Software Defined Radio Front End. [J] Solid-State Circuits Conference, 2006. ESSCIRC 2006. Proceedings of the 32nd European , 2006 : 203 - 206
3)A low power flexible PGA with DC offset cancellation for software defined radio front-ends. [C] Communications and Information Technologies (ISCIT), 2010 International Symposium, 2010 : 49 - 52

i have already complete a design about a SOA (mentioned in the paper listed above, a common two stage Miller compensation opamp)

when it is open loop in the simulation
GBW=82MHz, load capacitor= 300f, Rc= 12K, Cc= 100f, open loop ac gain= 73.21dB, PM=81.04

now the problem i am facing is that when i put the 7 same soa in parallel as a big opamp and put it into a closed negative feedback,

Rs= 10k, Rf= 11.2k, load capacitor= 100f,

the THD is very terrible is higher than -40dB(Sorry , i feel so upset and forgot to write down the value)

so anyone know how to deal with the THD problem? Thanks a lot~

load resistor= Rs,

Putting several opAmps in parallel with a common feedback (nor with individual feedback) usually won't work, as each amp will work a little bit different (at least in reality), so the outputs just fight against each other (output contention). If you need the drive capability, just use one opAmp and add an appropriate driver.

u r so nice~thank u for your warm heart~;-)