Moderators: Dave Moll, Andrew Davie, Steve Anderson
ppppenguin wrote:Now you know why I don't fully trust frequency meters:-) It's not the basic accuracy** which may or may not be good enough for a given job but mistriggering which you seem to be suffering here.
What probe(s) were you suing with the scope? They make a big difference at higher frequencies.
**Basic accuracy even with a rubbish meter is hardly likely to be worse than 100ppm or 0.01%.
Steve Anderson wrote:Well, that basically seems to be working OK, but you do have some ringing/overshoot which could be down to ground/0V runs and/or scope probe problems.
You should be using a x10 probe to reduce the capacitive loading on the circuit. Otherwise the 4040 appears to be doing its job. So I think you can assume that the two LCD 'scopes' are useless at anything above high audio frequencies, or at least the frequency counter part is. No surprise really.
Having got that out of the way what is the frequency at pin 1 of the 4040? Is it still 6.6kHz?
Is the Freq. pot centred? What is that frequency at either end of the pot track, clockwise and anti-clockwise?
The labeling of the pins is just down to each manufactures whim. In the two shown Q0=QA, Q1=QB and so on, there's no other difference, just the naming.
Steve A.
ppppenguin wrote:Just a thought. Do you have decoupling caps across the power pins of each chip? You really need 100n on each chip for HC. Inadequate decoupling can cause all sorts of grief. Likewise a poor 0V rail. If you can't have a gruound plane then a solid bit of wire is ideal. Make it grid of wire, round the edge of the board and connections across. The more solid the ground, the more predicatble the circuit.
If you don't have decent x10 probes then the DIY approach is a lenght of co-ax and a resistor.
Terminae at the scope end with a T piece and a resistor the same impedance as the co-ax. At the other end of the probe put a resistor that's 100x the value of the termination.
The voltage accuracy won't be brilliant and you'll need to crank up the gain of the scope but for logic neither of those usually matter. The extra loading on the circuit doesn't matter either in most cases. Don't forget to earth the braid to the circuit under test. A probe like this, even crudely made, is good to at least 1GHz. Far better than any x10 passive probe that I've seen and a lot more tolerant of long earth connection etc. And its cost is approximately zero.
Steve Anderson wrote:Progress indeed! Some of the scope traces look a little 'rough' but that could be down to the rather scattered layout of the project. Once it's tidied up a bit those may clean up.
I know I keep banging on about decoupling/bypassing and nice fat ground conductors or where possible ground planes, here you can see that it becomes essential when dealing with frequencies over 1000 times those usually found in NBTV circuits.
Steve A.
Return to The Tasmanian Devil VCR139....A for Andrew
Users browsing this forum: No registered users and 2 guests