Electronics > Beginners

170KHz LC Oscillator vs Plain BNC Anomaly. Please Help.

(1/6) > >>

Mechatrommer:
Houston! or anybody out there... this is another sequel from High-precision L/C Inductance Capacitance meter Design

i was playing around with LC Oscillator using LM393. Original design from the above thread is using LM311, but i dont have. Long story short, i managed to get it oscillating at the comparator's output.
below is the circuit diagram. black colored is original design i copied from the thread, i add the red colored, 2 X 100KOhm so i can get half volt divider (6V from 12V supply) so i can feed node 1 and 2 (balanced between +ve and -ve) to my frequency counter and counted correctly (Hantek 3x25 frequency counter only count on zero crossing signal, afaik), node 1 goes to +ve, node 2 goes to -ve.

whats buggering me is dso reading at far below (bottom left and right), pls note i've indicate in red text how i connect my dso probe to the circuit (+ve probe, -ve probe) to get the reading. not so much annoying on the left, just a little bit of distortion. the most annoying is the bottom right (out of 4 graph figure), the amplitude goes down from 10Vpp to 5Vpp (half). it happened when i introduce BNC connector at node 1 and 2. the BNC is not connected yet to anything, its just hanging there, but it sucking the voltage down. the BNC is not terminated and continuity and resistive DMM check indicates its just open circuit (+ve and -ve wire).

i know we've discussed on BNC performance at various frequency in this thread How to Better Wire the BNC Connector and GHz issue, but come on, this is just 170KHz circuit, hardly touching MHz mark, how bad can a BNC connector be to a KHz circuit. or maybe there is something else causing this? fyi, the real bugger is the BNC cable (BNC3), it sucks the volt most (from air!), BNC1 and BNC2 only attenuate the signal a little bit.

Any explanation?

deephaven:
the 2 X 100K give you quite a high impedance of 50K which means at 170KHz you won't need much stray capacitance to affect the mesurement. Either make the 2 X 100K a lot smaller or put a capacitor (say 100n) across the bottom 100K to lower the impedance at 170KHz.

tnt:
That cable represents a capacitance, probably around 50-100 pF from the length.

The potential at 1 is probably always the same, what changes is the potential at 2.
The node 2 has a 50k impedance so with the cable capacitance (lets say 50p), it makes a 60 kHz low pass filter ...

Mechatrommer:
thanx guys. i'll try to digest (calculate) that later. further test, i connected it to freq counter. it only count correctly when i put the dso probe at the same node as bnc. when i move dso probe to 1-G (FC remains at 1-2), the FC does not count anymore, and the dso read back as the top left picture. 12V draw 6mA, and this thing is unstable as touching my finger on the +ve battery will change the graph from left to the right side of the graph below! sigh :(

Mechatrommer:

--- Quote from: deephaven on April 14, 2011, 10:23:42 am ---...Either make the 2 X 100K a lot smaller...

--- End quote ---
i'm afraid this solution will draw more current from the battery, makes it less energy efficient circuit. thanx for the 100nF advice.

Navigation

[0] Message Index

[#] Next page

There was an error while thanking
Thanking...
Go to full version