Products > Test Equipment
2.5 GHz 12 digit frequency meter - Silicon Chip Australia Dec 2007
DMartens:
hi ... assembled the above freq meter ... both PCBs were ordered directly from SC and parts from a mix of the usual parts suppliers ... after power up, the SC welcoming message shows ... and then (sometimes) "0.".
Using a NEO M8 GPS module, I generated a 1 MHz signal and fed it into CH1 of a scope and connected CH2 to TP2 which is the PIC crystal frequency divided by 8 ... I then adjusted the PIC crystal until the scope showed no (or minimal) phase shift between both 1 MHz signals ... I repeated this for the 32,768 Hz internal time base crystal ... afterwards I also verified both frequencies with another frequency meter I have and both were spot-on.
Last thing to adjust is the Offset Adjust for the input stage around the MC10116 ... the SC article says to adjust the 1K pot meter until the display shows "0" with no input signal connected ... which for me I am unable to do consistently ... sometimes it does, sometimes it doesn't.
By feeding that same GPS output signal of (let's say) 1MHz square wave into CHA of the freq meter, the display is showing frequencies anywhere between 2 and 4 MHz.
If I read the schematic correctly (I am not very familiar with ECL logic and also not with differential amplifiers), putting a scope on TP6, I should see a 1 MHz square wave but I get a 1 MHz square wave with lots of very short "hair" / noise pulses ... adjusting the 1K pot meter from full CCW to full CW I can minimise the noise but I am not able to fully remove it.
I am know the values on the schematic in article Pt1 were changed when Pt2 was published. I have installed the values as per PCB silkscreen).
Am I overdriving the input stage by applying the 3V3 GPS signal? Is it related to impedance mismatching? Am I picking up EMC since I have not mounted the PCBs in a housing yet? (I cannot test CH B yet as I am missing one SMD component which is on order).
Anyone who also has built this unit who encountered similar issues? Or anyone who can advise how I can troubleshoot?
Kleinstein:
An ECL input should get away with a much smaller signal (e.g. 100 mV should be plenty). They have the tendency to pic up EMI or oscillate if left open.
A lack of termination and thus signal reflections that cause ringing is possible with a relatively long cable and sharp edge square wave as from a CMOS output.
coromonadalix:
no offence, but you should share the schematic to help others to help you ....
some or many of us dont ses it until you go / search on sillicon ...
DMartens:
Good point and screenshot attached ... I had (wrongly) assumed that only people who have built this unit would be in a position to reply ... but that is of course a silly assumption from my end.
nctnico:
I'd start by probing each stage to check at which stage the signal starts to look odd and also measure the voltage between the input of IC 5b to see if there is enough voltage difference to create hysteresis.
Maybe this is a power supply issue.
Navigation
[0] Message Index
[#] Next page
Go to full version