Trick Question

[Frank Bemelman]

Can I pass on my prize to the first person who can explain this scope
trace?

No problem!

http://www3.sympatico.ca/borism/Misc/PDTRACE.jpg

Setup - 10mm unbiased silicon photo diode connected to the vertical input
(1M) through 3' of coax and facing the CRT thus looking at its own output.
All the lights in the room were out

Scope Tek 465 at 5mV/div vert 5ms / div hor.

Explain big and small wiggles in the trace.

I'm going to win my own transistors back Scope is free
running. As the dot moves closer to the photo diode, more
light falls on it, thus the trace rises. The jaggies are
caused by the graticule, which decreases the amount of light
when the dot passes behind it.

 

[Walter Harley]

Can I pass on my prize to the first person who can explain this scope
trace?

http://www3.sympatico.ca/borism/Misc/PDTRACE.jpg

Setup - 10mm unbiased silicon photo diode connected to the vertical input
(1M) through 3' of coax and facing the CRT thus looking at its own output.
All the lights in the room were out

Scope Tek 465 at 5mV/div vert 5ms / div hor.

Explain big and small wiggles in the trace.

Beautiful! I presume the small wiggles are due to the vertical grid lines
obscuring the light, and the big curve is because of how close the trace is
to the photodetector. And then we get these flat parts, where the trace
would "like" to go up, but if it does it gets obscured by a horizontal grid
line, so there's negative feedback keeping it constant just below the line.
(Some ringing in the coax, too, looks like.)

 

[Boris Mohar]

No problem!


I'm going to win my own transistors back Scope is free
running. As the dot moves closer to the photo diode, more
light falls on it, thus the trace rises. The jaggies are
caused by the graticule, which decreases the amount of light
when the dot passes behind it.

That solves my problem of what to do with all those transistors.

You can also "push" down on the trace with your finger between the screen
and the PD.




Regards,

Boris Mohar

Got Knock? - see:
Viatrack Printed Circuit Designs http://www3.sympatico.ca/borism/

void _-void-_ in the obvious place

 

[Ben Bradley]

In sci.electronics.design, "Walter Harley"

Beautiful! I presume the small wiggles are due to the vertical grid lines
obscuring the light, and the big curve is because of how close the trace is
to the photodetector. And then we get these flat parts, where the trace
would "like" to go up, but if it does it gets obscured by a horizontal grid
line, so there's negative feedback keeping it constant just below the line.
(Some ringing in the coax, too, looks like.)

I don't see anything I would attribute to ringing, especially at
5ms per horizontal division.
The trace is most interesting between two and three cm from the
left edge. It's mostly horizontal, but drops down a little bit as it
goes under each minor tickmark, but finally breaks through and almost
hits the next higher horizontal gridline but then hits the three-cm
vetical line and dips down for a few hundred uS before going up to the
next horizontal line.

 

[Guy Macon]

You can also "push" down on the trace with your finger between the screen
and the PD.

For years I have used a photodetector and a scope as an arbitrary
waveform generator. I cut a sheet of construction paper to the
waveform I want and have the photodetector act as a feedback loop
to keep the trace on the edge of the paper.

BTW, Here is a computer that does useful work and which you
can't measure with an oscilloscope:
http://www.research.ibm.com/resources/news/20011219_quantum.shtml

 

[Dave Cole]

What *real* circuit can't be measured with an oscilloscope ?

1st prize : a bag of 200 small pnp transistors.
2nd prize : a bag of 100 small pnp transistors.
3rd prize : a bag of 50 small pnp transistors.

A hydraulic circuit.
Dave Cole

 

[Guy Macon]

A hydraulic circuit.
Dave Cole

*Excellent* answer! All those years working at Parker
Hannifin hydraulics dic=vision and I missed this answer.
D'OH!

 

[N. Thornton]

I can measure those with an oscillosope. Just run a couple of
30AWG wire wrap wires from the light outside of the case, and
hook the scope probe up.

then youd be measuring the light, rather than measuring the receiving circuit.

Here is a computer that does useful work and which you can't
measure with an oscilloscope:
http://www.research.ibm.com/resources/news/20011219_quantum.shtml

true, though its not an electronic circuit.

Regards, NT

 

[Ian Buckner]

"later in the day" is a concept that doesn't apply well to a
newsgroup where users are in many different time zones...

I'm only in one timezone.

Regards
Ian

 

[Goran Larsson]

1st prize: Goran Larsson

Weird. Never expected that people, other than me, would try to
look inside an 'scope using the 'scope itself (did that in 1981
when adding more trig modes and trig view to a 'scope).

Winners can collect their prizes by emailing their
address, watch the 'x' and 'invalid' in the reply address,

I have no use for 200 small pnp transistors, and the postage for
priority airmail from .nl to .se probably cost more than they are
worth. Can they be given to a local charity instead?

 

[Frank Bemelman]

Weird. Never expected that people, other than me, would try to
look inside an 'scope using the 'scope itself (did that in 1981
when adding more trig modes and trig view to a 'scope).


I have no use for 200 small pnp transistors, and the postage for
priority airmail from .nl to .se probably cost more than they are
worth. Can they be given to a local charity instead?

Maybe I should start a new contest-> design an audio amplifier
with only bc556b's in it, or something like that

 

[Walter Harley]

Maybe I should start a new contest-> design an audio amplifier
with only bc556b's in it, or something like that

Can we at least have a power source of some sort? ;-)

Although, hmm, if we cut the top off some maybe we could use the
photoelectric effect.

Certainly, you can make capacitors out of transistors (not that you need
'em, we can be direct-coupled with a bipolar supply); diodes are easy; a
resistor is just two diodes back to back in parallel, although the
resistance is rather nonlinear... hmm, maybe with enough BC556's this is
do-able.

 

[Guy Macon]

resistor is just two diodes back to back in parallel, although the
resistance is rather nonlinear...

If that's true, a single diode is also just a nonlinear resistor.

 

[Philip A. Marshall]

true, though its not an electronic circuit.

initially the OP didn't specify that it was an electric circuit.

 

[Guy Macon]

initially the OP didn't specify that it was an electric circuit.

I can make an electronic circuit that stops working when a probe is
applied that has over 0.01pF of probe capacitance or less than 1
GigOhm of probe resistance. At microwave frequencies. Try measuring
*that* with an oscilloscope!

There are electromagnetic circuits that will disable any CRT-based
scope nearby with strom magnetic fields, but an LCD scope would be
able to do the job.

In addition, many of the things we measure with oscilloscopes are
actually measured with specialized oscilloscope probes.

 

[John Woodgate]

I read in sci.electronics.design that Frank Bemelman

Maybe I should start a new contest-> design an audio amplifier with only
bc556b's in it, or something like that

Quite possible, but to do it with only PNPs is in the nature of 'cruel
and unusual punishment'.

 

[Wouter van Ooijen]

Maybe I should start a new contest-> design an audio amplifier

with only bc556b's in it, or something like that

But that is much too easy (unless you mean the T's only no passives?).
I would say: only 555's and passives! And the 555's are from a junk
yard, so 10% might be disfunctional.




Wouter van Ooijen

-- ------------------------------------
http://www.voti.nl
PICmicro chips, programmers, consulting

 

[Robert C Monsen]

BINGO! I thought it would take longer for someone to
come up with this answer. Congratulations!

Actually, the act of measuring a circuit with an always ocilloscope modifies
the operation of the circuit under test. The oscilloscope itself is simply a
more severe case of the same problem (it CAN be measured, you simply get
bogus results.) Consequently, no circuit can be measured perfectly with an
oscilloscope...

Regards,
Bob Monsen