Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.

  1.  
    Cross posted form the "Banality" section as more appropriate here:


    Does anyone have links to documentation on the Tek 7000 series scope used for the punchline demo? Specifically for documentation on the Math functions and how they can be combined with channel values and other conditionals / expressions to produce a calculated trace.

    I've downloaded a fair amount of documentation from the Tek site itself and registered for help and pointers on the forum, but to no avail.

    The documentation I've seen refers to the process for adding Math functions in passing, and I've also seen a document which takes *31 pages* just to list the functions available, but with no broader explanation / examples on the expressions possible.

    In particular how conditionals work in expressions is of interest.


    One explanation given for the -ve slope on the energy integral in the traces is that the expression being integrated may change depending on the sign of a measured value (+voltage and -voltage for instance).

    I can't remember who suggested this (so I can't credit) but I thought it was worth looking at.

    In standard Excel or OpenOffice world it's trivial to construct this sort of thing with IIF() and similar, but I can't get any traction on how this is achieved in Tek scope Math.

    None of the functions looked like conditionals at first blush, but maybe it's more direct in the form of some sort of ternary syntax?

    Any ideas?
    • CommentAuthorpoynt99
    • CommentTimeFeb 17th 2010 edited
     
    Indeed that scope is not doing any complicated math on that trace other than straight-forward integration.

    I've already shown this with my posts at overunityresearch.com, and I hope to replicate the wave form more faithfully soon using a modeled core with the inductor.

    In short, yes the voltage across the coil reverses and so then does the v*i power calculation. The scope is faithfully displaying what is happening in the circuit, but it's not what many believe it to be.

    I also expect to prove that the free-wheeling diode Steorn is using is either defective, not electrically connected, or just simply not doing the intended job, i.e. it is open-circuited wrt the IK spike.

    .99

  2.  
    @0.99

    Ok, I missed that - not on my list of regular reading.

    A bit of fiddling at this end shows that if you have an ABS() function (a certainty), you can construct a couple of expressions that would include / exclude other terms based on sign in the absence of explicit conditionals.

    But if it's all unnecessary, I won't bother digging at the rest much.
  3.  
    i think the spike might just due to the stray inductance of the long unshielded leads
    if the differntial probe is placed close to the diode it would probably not show the spike but if connected close to the coils that is at the other end of the long leads i would expect to see a significant spike of at least a hundred or so nanoseconds and possbly even a microsecond

    also the diode might be a rather slow rectifier diode rather than a fast recovery one
    from what i understand the ssr has a vey high switching speed so with all that stray inductance i would expect that sort of spikes to show up
    • CommentAuthorpoynt99
    • CommentTimeFeb 17th 2010
     
    The differential probe leads are placed right next to the diode.

    The inductance contributed by the long coil buss wire will not pose a problem or introduce skewing of the wave form or how the free-wheeling diode responds to the positive and negative coil voltage. The wire inductance simply becomes part of the overall total inductance being switched by the SSR, and neither the diode nor the DIFF-probe are affected by this.

    Even a relatively slow 1N4007 rectifier diode is plenty fast enough to catch the flyback in most switched-inductor applications, so this is not the problem in this case. Even if the diode was a tiny bit slow it would still capture 90% of the flyback and we would not see the large negative dive in that blue integrated power trace.

    .99
  4.  
    i wasnt paying enough atention and didnt realise you were inferring the spike from the integral trace
    that definitely contains way too much energy
    i was thinking about another trace i ve seen recently which had a very thin spike but i cant find out which it was

    i can also never seem to figure out which traces correspond to which experiment
    one that i saw definitely had the diff probe attached quite far from where the diode could have been so it seems i was associateing the trace with a different setup

    that said a 4007 will allow an easily noticeable spike on recovery
    to see it you would normally need a rather high sampling rate or if usin a crt scope it would be easily visible if you turn the brightness way up
    note that i say easily noticeable but that doesnt mean it contains any significant amount of energy certainly not enough to explain that step in the integral
    even more so given that the integral shows there is both a positive as well as a negative spike