Hi Dave and Ravinder, The scope chassis is grounded even if floating as far as ESD is concerned. Amps of current can flow through the chassis capacitance to its surroundings. Most of the problems occur at the point of connection to the circuit.
The inductance of the coax shield connection to the EUT is dominant as far as errors go. The other problem with a coax is it makes a great antenna and can cause ESD problems in the circuit. How well a voltage measurement works depends on the enclosure design and using it if possible to terminate the coax shield. But in general, forget all the probes available from the major measurement companies. If a voltage measurement must be made, the Fischer BCP-2 passive differential probe probably works best as it can take lots of common mode without saturating. But, usually for a problem like this, I use loops to trace current flowing around. I also like to take a square loop, feed it from an EFT generator and probe individual paths and plane surfaces on the board looking for weak spots that produce the same symptom. It takes some practice at doing this but I have fixed more ESD problems than I can count this way. Applying the above as described on my website, I can generally get to the bottom of a problem like this in a day or two. Just think Ldi/dt (20 nH * 10 A/ns = 200 Volts!) and Cdv/dt (1 pF * 2kV/ns = 2 Amps!) while you are working. Currents flow and voltages are generated that defy normal troubleshooting intuition. Do "null" experiments on any measurement you do (such as shorting a probe while connected to the circuit during the ESD event to make sure the probe "noise" is lower than what you are trying to measure. Hope this helps. The problem with ESD debugging is not that it is rocket science, it isn't, but as with many things, it takes a lot of practice to get proficient. There should be someone in a company of yours size charged with becoming proficient in this type of troubleshooting. Trying to fix a problem like this for the first time just when it comes up requires a bit of luck. I have not seen many companies the cultivate internal experts, but they should. Doug [email protected] wrote: > Ravinder, > > This is a challenging measurement. The di/dt of the ESD pulse is in the > neighborhood of 15 A/ns. Figuring roughly 0.5 nH/cm for bonding the PCB > to a measurement plane and you have 8 volts of ESD signal per cm of > bonding wire. This will drive any active probe I know of to distraction. > If you can connect to the measurement node/common with 50 ohm coax you > have a chance. Some ferrite beads at the PCB end will help. A floating > battery operated oscilloscope might help, especially if it is enclosed > in a faraday cage. To see how things are doing you can connect the > coaxial hot and shield to the PCB common and see what signal the ESD > pulse induces into the measurement equipment. If this is smaller than > the logic signal you are looking for then it might work. But I would try > to avoid this approach and go for an indirect measurement. > > This is one of those troubleshooting exercises that are great if you can > solve it without an oscilloscope measurement - if you can deduce what is > happening by how the circuit behaves in response to the ESD event. > > This fast di/dt event can induce signals into the digital lines that > will trigger things. Some lines you can decouple with capacitors. These > would be reset lines, power rails, and any other line that will work > with a slow signal. > > There are tricky ways around these ESD issues such as having the circuit > sense that there is a problem and having it reset or reboot or something > along those lines. Another solution I have seen is to have the circuit > blindly resend the data every few microseconds so that if there is a > disruption due to an ESD event no one notices. > > Dave Cuthbert > Micron Technology, Inc. > > -----Original Message----- > From: [email protected] [mailto:[email protected]] On Behalf Of > [email protected] > Sent: Monday, February 27, 2006 5:48 PM > To: [email protected]; [email protected] > Subject: RE: Measurement of logic signal during ESD test > > A great place to start is at Doug Smith's High Frequency Measurements > Web > Page: http://emcesd.com. Doug is also a member of this list. > > Regards, > Stu Benner > EMConsulting > www.emcwizard.com > > > -----Original Message----- > From: [email protected] [mailto:[email protected]] On Behalf Of > [email protected] > Sent: Monday, February 27, 2006 5:03 PM > To: [email protected] > Subject: Measurement of logic signal during ESD test > > > > Hi EMI Gurus, > > I need to troubleshoot a logic failure which occurs when 16 kV ESD > pulses > are applied to the PCB. Is there a reliable way to attach the > oscilloscope > probe to the failing device while ESD gun is being fired elsewhere on > the > PCB. Will it help if I use a differential probe. > > Thanks for the help. > > Regards, > > Ravinder Ajmani > Server PCB Development > Hitachi Global Storage Technologies > -- ___ _ Doug Smith \ / ) P.O. Box 1457 ========= Los Gatos, CA 95031-1457 _ / \ / \ _ TEL/FAX: 408-356-4186/358-3799 / /\ \ ] / /\ \ Mobile: 408-858-4528 | q-----( ) | o | Email: [email protected] \ _ / ] \ _ / Website: http://www.dsmith.org - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. Website: http://www.ieee-pses.org/ To post a message to the list, send your e-mail to [email protected] Instructions: http://listserv.ieee.org/request/user-guide.html List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas [email protected] Mike Cantwell [email protected] For policy questions, send mail to: Jim Bacher: [email protected] David Heald: [email protected] All emc-pstc postings are archived and searchable on the web at: http://www.ieeecommunities.org/emc-pstc
