DS1102E EMI debug

[handaza]

Hi,

I am facing an EMI problem with a Microcontroller PCB, I decided to order oscilloscope Rigol DS1102E to debug it.

my question to the experts  is can this model handle this job? and what setup shall I do to make it ready?

[han]

I suggest using Spectrum analyzer + near field probe

[handaza]

I have very tight budget, can the FFT function help?

[han]

If the budget is very tight and accuracy is not importance, you can make your own near field probe(lot of guide in internet), and use cheap SDR with lower band extension.
Not for measuring trough (just debugging)

[cdev]

Youtube user "SM5BSZ" has some excellent Youtube videos on using something like an RTLSDR to locate and eliminate EMI, look for his four "sensor" videos.

[Mechatrommer]

I am facing an EMI problem with a Microcontroller PCB, I decided to order oscilloscope Rigol DS1102E to debug it.

why E series? why not DS1054 Z series and mod it to 100MHz? about the same price... they can do the job to some limited extend, but still they are not the right tool for this job... esp if EMI of concern is far beyond the hobbiest grade 100MHz BW, ymmv.
http://www.emcsociety.org/wp/wp-content/uploads/2012/11/Debugging-EMI-problems-with-an-Oscilloscope_June2014.pdf

[handaza]

please Mechatrommer tell me why I shall select DS1054 Z not DS1102E?

I shall tell the problem, I am controlling a 3 phase motor, the 3 phase motor is controlled by Large contactors , the contactors are controlled by relays on the PCB, this is my 2nd PCB design, in the first design the microcontroller was freezing on any motor operation, in the 2nd design I cared for the ground signal return paths and separated Relays ground from digital ground, the microcontroller is now fine, the problem is with LCD and 24C64 i2c eeprom, the LCD display garbage when the contactors operate, and the 24C64 eeprom is not responding until I completely disconnect power from PCB at all and reconnect it.

another trick I did , the 5v output from 7805 regulator is splitted to two paths, one for microcontroller and other for remaining ICs, each path has series shotkey diode and after the diode a 1000uf capacitor.

I thought the oscilloscope shall help me to debug this problem.

kindly tell me what shall I do as this project is very important.

[cdev]

Maybe the energizing and collapsing magnetic field is inducting high voltage transients. Also so called back EMF is often a problem with motors as they can also act as generators?

Actually, there is an area where a new board category would be good. Motor control, Motion control, "mechanatronics" and also hardware that is wedded to electronics, for example, motors themselves, steppers, actuators, for example.

[crispy_tofu]

why I shall select DS1054 Z not DS1102E?

4 channels, can be hacked to 100MHz, better screen, higher memory depth, newer (so better support), overall better value. 

[handaza]

OK I will buy ds1054z and crack the bandwidth.
Can you tell me how to use the oscilloscope to debug this problem ?
Also is there any modification I shall do to the design to fix the problem?

[Mechatrommer]

Can you tell me how to use the oscilloscope to debug this problem ?

probe the mcu power rail. if you see any spike below Vmin (or usually below 0 in case of motor transient), thats your pointer...

Also is there any modification I shall do to the design to fix the problem?

have you beefed up voltage transient/voltage suppress/fast acting diode everywhere to avoid transient/backfire reaching the mcu?

why I shall select DS1054 Z not DS1102E?

other than tofu say... i simply say... E series are now obsolete discontinued...