Hi Fred, I can understand well the annoying questions Olin has asked you. Everytime when you are dealing with noises you need to know how much noise is accepted after the amplifier. A good help will be a scope with FFT in the range of your input signal. In this way you can see where is the noise component and if your fourth order filter is able to suppress it. If you don't know your signal bandwith comming from your transducer is very difficult to compute a right filter. An instrumentation amplifier (and thanks God it was invented and you don't need do design your own instrumentation amplifier, like other people here back in the 80) need a lot of care for the PCB design. But usually the noise is mixed with the transducer signal and sometime it can't be separated. Vasile On 1/17/09, fred jones wrote: > > Hi all, > I asked a few months back about amplifying and filtering a very small sig= nal sinewave. Since then I've done a lot of reading but there is a big dif= ference between book learning and experience. Please bear with me if I say= something stupid. I will be connecting a sensor made of a coil of wire th= at will be expected to output a sinewave somewhere between 20uV to 50uV. I= used an instrumentation amplifier for the 1st amp stage first at a gain of= 100 but I changed it to a gain of 2000 because the final output of the amp= lifier chain was cleaner. I am using a single supply of 10V so I put 1Mohm= resistors on each input leg, one to 10V and one to ground to bias it up to= 5V. The output of the first stage is difficult to see on my scope because= it is still relatively small but appears to be quite noisy. I think this = is the stage where I need great improvement. I feed the output to an LMH62= 44 via 1uF capacitor. This stage was set to a gain of 20 but I raised it = to 100 as the output looks a little cleaner. Next the signal is coupled to= a 4th order bandpass filter. I'm feeding 10uV in and the output of the fi= lter is too noisy. I am trying to attach a picture of the scope signal at = this point. I don't know exactly what this might be an indication of but I= hoped somebody here would recognize what this is an indication of and poin= t me to where my problem might be. I can furnish a jpg of the schematic. = As I increase the signal level the output gets cleaner but at 50uV it is st= ill not quite clean enough. I feel like if I can get a clean, stable outpu= t at 10uV input, then it would give some breathing room. I'm not sure what= to try next. It might be a deficiency in my circuits or a PCB layout issu= e, I just don't know. Any help is greatly appreciated. If there is an exp= ert who would like to look at this in detail, I would be willing to pay. > Thanks, > FJ > _________________________________________________________________ > Windows Live=99 Hotmail(R): Chat. Store. Share. Do more with mail. > http://windowslive.com/howitworks?ocid=3DTXT_TAGLM_WL_t1_hm_justgotbetter= _howitworks_012009 > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > > > -- = http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist