Re: Fet scope probe design recommendations

From: Mike Monett (gqtacfwfjfmx_at_spammotel.com)
Date: 09/17/04 

Date: Thu, 16 Sep 2004 21:55:16 -0400

Paul Burridge wrote:
>
> On Thu, 16 Sep 2004 17:13:27 -0400, Mike Monett <no@spam.com> wrote:
>
> >John specified "fast". For him, "fast" starts at about 10GHz, and you
> >need a 50 ohm environment to work at these frequencies. In fact, pretty
> >much anything above 1GHz should be 50 ohm.
>
> Okay, fine. Like VNAs you mean? But doesn't that limit that type of
> scope to RF work only and even then at points in the board where the
> characteristic impedance reverts to 50 ohms? I mean, I can't see it
> being much use to check signals from any hi-Z impedance source!

The problem is to acquire the signal with minimum loading and enough
bandwidth to represent the waveform, and transport the signal to the
measuring instrument (VNA, spectrum analyzer, scope, etc.)

The solution is to use a resistive probe as we have been discussing, a
standard 10MEG 10X probe, or a FET or bipolar active probe. At low and medium
frequencies, the choice depends on the desired bandwidth and circuit
impedance. At high frequencies, the circuit impedances are low so a resistive
probe is a good choice.

Getting the signal to the instrument is another problem. For a 10X probe, a
low-capacitance cable is needed. This means a very thin center conductor
which is fragile and has higher loss at high frequencies.

For 50 ohms, a short length of RG-174 is ok up to about 1GHz, but as others
have mentioned, the losses start to round the signal edges at higher
frequencies. This becomes noticeable with the EclinPs Plus with risetimes of
160pS or so.

So you need a better cable, such as RG-58, which is harder to handle. As John
mentioned, the problem gets worse the higher you go, which may be the reason
scopes top out at 70GHz.

A partial solution to the transport problem is to convert the signal right at
the source to minimize the cable losses. I believe both Tek and HP offer
remote pods for their samplers. Also, Picosecond Pulse offers 25GHz and
100GHz sampler modules which can be mounted close to the signal to minimize
loss:

  http://www.picosecond.com/products/category.asp?pd_id=17

Mike

Relevant Pages

  • Re: FPGA on-die LVDS termination issues
    ... You can't just probe in the middle of the trace and expect to see on your 'scope what the receiver sees. ... each pin has 10pF loading on it no matter what the IOB is set up for. ... There are a bunch of FET's connected to the signal to drive signals out if required by the configuration loaded into the device. ...
    (comp.arch.fpga)
  • Re: Mains pick-up on I/O pins
    ... I am trying to program an RC-100 demo board, ... My probes have been calibrated; the probe and scope have adequate ... still not seeing the signals I need (scope doesn't pick up anything but ...
    (comp.arch.fpga)
  • Re: Hiss on a scope?
    ... place the probe? ... from your gain hole, I'm guessing. ... Scope with no signal in, record the reading, then ... Or maybe record some test signals on a CD and use with an old Walkman. ...
    (alt.guitar.amps)
  • Re: Fet scope probe design recommendations
    ... >looking at low voltage signals up to about 100MHz. ... >mounted as close to the probe tip as possible, ... >the connection to the scope I was going to use the cable and compensation ...
    (sci.electronics.design)
  • Re: TV for oscilloscope
    ... >> trigger(a sawtooth) but I'm not sure at what currents and voltages I ... > Remember that a real scope has blanking. ... > then just putting the right DC voltage in will blank the video driver. ... to figure out how to amplify the signals properly. ...
    (sci.electronics.design)