Why all the noise!

Ask any questions relating to the PicoScope hardware or kit contents here.
If you have any questions prior to purchasing the kit post them here.
Post Reply
Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1126
Joined: Sun Aug 25, 2013 7:22 am

Why all the noise!

Post by Steve Smith »

“Noise” or to give it one of its official titles EMI (Electromagnetic Interference) is a topic that is often raised during Training, Forums, Facebook discussions and during scope use in general.

Unfortunately the environment we work within is no picnic and lends itself to such intrusion into our measurements.

Let’s think that through for a minute:

Environmental noise from EMI.
This can be in the form of ignition systems, chargers, invertors, lifts, welders, overhead cables, fluorescent lighting, workshop mains grounding etc. A point to note here is that the EMI generated by these devices will be picked up by unscreened wiring in the engine bay and so measured by PicoScope

The scope power & ground
can both be subjected to high levels of noise from poor Laptop/PC power supplies, invertors, USB hubs and inferior USB cables. (Always use the Pico Blue USB cable)
If you are using either the PicoScope 4225 or the 4425 you have the option to remove noise via the USB line by grounding the “Set Screw” at the rear of the scope to the vehicle chassis ground under test.
4225 4425 USB additional ground
4225 4425 USB additional ground
The test leads themselves can present noise to the scope depending on their routing and grounding. Where possible (and we have to live in the real world) try to reference all test leads to a clean ground point (ideally battery negative terminal).
If using the PicoScope 4225 or 4425 you have the option to use sensor grounds which are somewhat shielded by the PCM

The filtering tutorial below explains this technique further and is well worth a read from a "noise" point of view and the tools/features at our disposal to minimise the impact of noise
https://www.picoauto.com/library/training/filtering

Here is an extract below:
Both the PicoScope itself and the test leads provided are well screened to avoid the pick-up of noise, the same however cannot be said of the vehicle wiring harness. In the average vehicle, few signals are screened (to reduce costs) and sensitive signal wires are often run next to cables carrying signals that radiate noise such as those used for injectors.

I should add here that PicoScope does not generate noise, it does however capture noise and will display exactly what is presented via the test leads. The relevant Safety and EMC approvals can be viewed here under “Detailed Specifications” https://www.picoauto.com/products/autom ... ifications

Reading this all back it’s a wonder how we capture anything without noise and I guess we can say you don’t. Whatever is presented to the Pico test leads will be displayed on screen and we know some signals are more vulnerable than others.

Lambda, Airflow, ABS and MAP sensors are all particularly vulnerable and may require filtering in order to view finite detail. Once again if you have the PicoScope 4225 or 4425 the Bandwidth Limit feature is a rapid solution to remove a considerable percentage of unwanted noise. Again see https://www.picoauto.com/library/training/filtering

Oscilloscopes such as the older PicoScope 3423 3223 with a 20MS/s sampling rate will struggle to pick up high frequency noise (or measure high frequency signals such as FlexRay) as the majority of high frequency noise is will be inadvertently filtered

A PicoScope 4425 4225 has a 20MHz bandwidth and a 400 MS/s sampling rate so will capture and display high frequency noise that was always there but also hidden. Think of the faster oscilloscope like a high resolution photo - there is more detail that could not be seen before.

A typical example is alternator ripple - high frequency noise (often 1V or more) can be seen by faster oscilloscopes but is invisible to slower oscilloscopes and multimeters. The more recent trend of moving the battery away from the engine bay also leads to higher noise levels than seen in the past.

Noise is present and it should be captured, measured and evaluated for any relevance to our customer’s complaint. (Catch it dirty)

High frequency ignition spikes into CAN networks or PCM ground wiring are just two possible scenarios where the “noise” is in fact responsible for the fault. In order to capture these events we need high bandwidth, fast sample rates and sufficient memory to present evidence to our customers and PicoScope delivers.

I hope this helps and please feedback with any concerns you have surrounding noise and filtering techniques

Take care…….Steve

Post Reply