Loading ...
Sorry, an error occurred while loading the content.
 

Re: [VLF_Group] Re: E-field rx, front-end protection

Expand Messages
  • Paul
    ... Same experience. We need to make sure the device has enough open loop gain so that it is linear enough, when the loop is closed, to avoid distortion at
    Message 1 of 73 , Feb 20, 2013
      Wolf wrote:
      > I tried opamps instead of JFETs in the past, but I seemed
      > to have picked the wrong ones (low-power devices with
      > great noise performance, but too slow for handling MF and
      > HF signals before the low-impedance filtering stages).

      Same experience. We need to make sure the device has enough
      open loop gain so that it is linear enough, when the loop
      is closed, to avoid distortion at all frequencies which
      reach the input.

      prx-e1 at the moment doesn't meet that requirement.
      The AD823 is not quite fast enough, could do with 3 or 4
      times its bandwidth. The LT1028 is better but only suitable
      for loop receivers.

      Unfortunately wide bandwidth and low noise don't seem to
      come in the same package. It may be necessary to use a
      combination of wideband (100MHz+) bipolar opamp with a low
      noise discrete JFET in front, and the loop closed around both.

      Things get a little easier when moving away from the opamp
      differential pair at the input - 40% less noise straight
      away and we get a choice of JFET and control of its current,
      so there are other advantages to a hybrid.

      Will continue testing to determine the limits of the AD823
      front-end, then will try some JFET/opamp arrangements.

      Also will test some 1G bias resistors. There are some cheap
      ones, all thick film unfortunately, but worth some trials
      to measure the excess noise. If I replace the existing 100M
      with 1G the ELF noise will drop by about 60%.

      (Bootstrapping doesn't increase the noise, but nor does it
      reduce the noise, whereas moving to a real high-value bias
      resistor does reduce the effect of device current noise).

      --
      Paul Nicholson
      --
    • Paul
      Got a good price for the PCB fabrication from a local (Todmorden) manufacturer and I added up the parts cost, In UK pounds for the populated boards: prx-e1
      Message 73 of 73 , Mar 21 1:32 PM
        Got a good price for the PCB fabrication from a local
        (Todmorden) manufacturer and I added up the parts cost,

        In UK pounds for the populated boards:

        prx-e1 E-field front end: 23.67

        prx-t1 Filter, line driver, PSU: 52.73

        prx-r1 Line receiver: 24.42

        About 100 quid for the set, not including boxes and
        cables, etc, and doesn't include assembly and test.

        Well that's a bit cheaper than I thought it would come
        to. Will complete the breadboard testing this month
        and have a set of boards built up in April and then some
        serious testing.

        Further back in the pipeline is prx-h1 which is an
        H-field front-end developed from the loop receiver I
        currently use, and a -p1 isolated power supply.

        --
        Paul Nicholson
        --
      Your message has been successfully submitted and would be delivered to recipients shortly.