On 02:24 PM 9/09/2002 -0400, Richard Sumner said:
>Measuring spikes on the supply lines is not trivial. Where you put the 
>scope probe ground is everything. The best way to measure is with a 
>differential probe, placed across the power pins of the chip in question. 
>Anything less may not give you the true picture.

Off topic but maybe of some interest...and I am sure many of you have 
similar techniques.

A trick I learnt way back in 1986 when designing my first switchmode supply 
- make your own *low* impedance probe.  A low impedance probe is less 
susceptible to high speed transients nearby that cause false readings. A 
series cap (with suitable resonance characteristics - possibly made with 
multiple caps) and a shunt low impedance resistor.  The CRO probe connects 
across the resistor using the spring gnd clip that should be available with 
any high speed cro probe.

If a short spring gnd is not available make one using a paper clip bent and 
cut to make as short a gnd spring as possible.  Be careful if using the CRO 
probe as a former when you bend the paper clip - it can break them.  Choose 
a slightly larger drill bit and then crush the small spring you have made 
to make a tight fit onto the CRO GND sleeve.  You now have a spring to slip 
over the CRO GND sleeve (remove the scope condom to expose the GND sleeve). 
Bend and cut the other end of the paper clip to make a nice stiff probe 
point.  If you cut at an angle it will be a sharp point.

To choose the resistor you may want to try a range of values - there will 
be a "flat" spot where the low impedance probe is not so low as to load the 
cct but not so high as to give a false reading due to pickup.  For power 
supply work I sometimes go as low as 47 ohm (impedances are low so the low 
R is not adding any loading).  For decoupling tracing this may be a little 
low but 1k or 4k7 may be OK.

This is a useful trick, especially when you don't really trust the diff 
mode of your CRO or when your high speed active probes have died.  You have 
to think about the results as the low probe impedance can affect the 
results of course.

Sorry for those that the above is old news but it may be useful for some.


* Tracking #: 53D7B142AFF8B646AB9B2FCDF73D75C2DE43004C

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
* Contact the list manager:
* Forum Guidelines Rules:
* http://www.techservinc.com/protelusers/forumrules.html
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Reply via email to