According to Pete - Filtering


Just a heads-up: Friday January 11, 2013 is our favorite day of the year, Inventory Day. That means we’ll all be busy counting every IC, wire, and switch we have in stock, so orders won’t be going out that day. Our apologies for the inconvenience!

In today’s edition of “According to Pete,” Pete goes over the basics of filtering. This includes discussions about terms and definitions, the use of a Bode plot, and other basic in’s and out’s of filtering electronic signals.

Vimeo version can be found here

If you have any comments or questions, please feel free to leave them in the comments section below. Thanks for watching and we’ll be back soon with another edition of “According to Pete.”


Comments 28 comments

  • I can’t tell you guys just how apprehensive I was about doing this video. As the first few commentors state, this really isn’t a simple topic (I didn’t even touch phase!). But I hope I gave some tips for playing with basic filters, and I really hope I’ve done more good than harm (currently unconvinced of this).

    The video also shows that the holidays have been cruel and I’ve put on a few pounds…

    • As a 3rd year EE student who has always struggled with understanding the basic concepts of filters, this video helped me immensely. You always are able to fit more information in your videos than I often learn through a few hours of class. Thanks for reminding me of some core concepts that I didn’t catch on to the first time!

    • Great video. I studied this stuff about 20ish years ago in college and understood it really well. When I went off to a real job, I was working on a circuit that had a problem with electrical noise causing the gate of a mosfet to turn on partially, which led to failure of the part due to power dissipation issues.

      The lead engineer was working on a fix and asked me to test his ‘fix’, which was to put a bigger mosfet into the circuit. Well, in case you don’t know, bigger mosfets have bigger coupling capacitances between their terminals. I knew from the get go it wouldn’t work, but this circuit was his baby and he wouldn’t listen to me.

      Finally went to the head of the department and handed them a spice simulation of the effect of increasing the capacitances by way of this fix. I also recommended adding decoupling capacitance to the power rails of the circuit to get rid of the noise. He listened, agreed with me, then did nothing about it.

      About a month and a half later, I was talking to another engineer and found out he was assigned to work on fixing the noise on the power rails.

      Anyway - the video really brought back some memories. I don’t think I’d complain if you wanted to go more in depth with it, even though its a huge subject…

    • You said not to ask for more detail, so I’m going to ask for more detail! I would love to hear more about this subject. I’m sure there are some out there who want to work with things like analog audio circuits and share this interest.

  • Just a heads-up: Friday January 11, 2013 is our favorite day of the year, Inventory Day.

    This is true, at least for values of “favorite” which encompass panic attacks and heartfelt loathing.

    • Funny, the same is true of some of my favorite people.

    • Why have an inventory day? I know they use computers to keep track of their inventory. Answer: Because people who actually know how computers work do a ROFLMAO when they hear the phrase “Computers are Infallible.”

      • At best, that’s an oversimplification of the problem. At worst, it’s a gross mischaracterization. Unless access to the physical quantities is restricted to a robot inside of a sealed vault, they don’t technically “use computers to keep track of their inventory”. They use humans to track inventory; that data is digitized and subsequently manipulated as parallel to reality as possible. Since access to the physical quantities is not strictly regulated by the computer, the physical quantities and the inventory system’s assertion of those quantities can disagree. The fallibility of the computer is not as much of a concern as is the fallibility of the humans that are interfacing with the system.

        • Oddly enough, our inventory is inside a sealed vault, and touched only by robots. We humans will all be hitting the snowbeach at A-Basin tomorrow while the computers do their thing.

      • well when computers fail people tend to notice and complaint and there is a change that bugs get fixed, more often people fail and the IT guys may or not notice (It wasn’t me mentality), and then someone makes a decision to use a db field for something it wasn’t designed for or purposely miss classifies a item so the profits go to a certain department and causes a LOT of problems with some programs “suddenly” miss behaving, but that may just be me complaining again that the most sold CD in the store is actually a bag and that it messed up our sync system and top sold lists (somehow apparently is my fault). But more on topic, I definitely need to see more of these clips as there is always something new to learn in each one.

        • I always loved inventory… Weigh the box of resistors to tell how many are inside the box was the easy part. Locating 1000 boxes of resistors in the TP supply closet was not so much fun. Why they were in the TP closet, who knows.

  • I wish you were my lecturer in school. Bloody hate filters as a result! You explained that very well!

  • Great job Pete! This filled in quite a few holes that my AC theory class didn’t cover, yes please do more episodes about filtering.

  • Can you post the product number for the ice pick?

  • Please explain in more detail. Please show practical examples (for visualization). Please cover audio filters (such as the specifics of a 10 band EQ).

  • Is it just me or are there 422 unnecessary cuts in that 15 minute video? I lost the last 10 minutes or so just keeping track of how many times you cut the segment.

    Or, have you figured out short-distance human teleportation and are not sharing it with us? 0.0

    • I would really like to see more unnecessary cuts, or perhaps a cut counter in the corner. It’s like visual dubstep.

  • If an example is too much to ask for as an “According to Pete” can it come as tutorial? I would love to see it and have a couple of sensors signals that have been mucked up by noise. HELP!

  • Better than my college education.

  • Nice work. A great jumping off point for this topic.

  • Great video, but too short for the complexity of the subject

    Could you give a few book references? (an also post the links of the videos you talk about)

    I studied this stuff about 10 years ago and I actually remember nearly nothing :S

  • Great job Pete! I just “discovered” Sparkfun a few months ago as I began to re-ignite my hobby of micro controllers and electronics. I have been extremely impressed by the products and services provided by Sparkfun including your videos. Being an educator (my real job) I can tell that you have a passion for sharing knowledge; while the concept of filters is complex your video did what it was suppose to “do” which to get people fired up about learning more! Again great job and I always look forward to your videos. Thanks for the link/suggestions as I’m working on a project to measure weight using a load cell which has been plagued by noise issues.

    Keep up the great work-Brian

  • Nice video! This stuff tends to leave my brain too fast :)

  • Unrelated to According to Pete–If Friday is inventory day, will there be a new products post?

  • Nice video. Explaining analog filters in 15 minutes does not sound like an easy task! I learned them in school, but I didn’t really ‘get it’ until I worked with some real world examples.

    My vote would be to set up a couple real circuits and test equipment and show what you can do with a simple RC or Sallen Key. You seem to have a love for audio, so maybe an example showing how to remove 60 Hz hum or removing digital noise from a sensor output. I would get more value out of seeing those filters in action and understanding your component choices instead of going over the equations to configure a Bessel or Chebyshev roll-off.

    DSP would be cool too, but if you can explain DSP in 15 minutes, then you should be nominated for some sort of award!

  • Pete, great intro to filters. Since analog is becoming a thing of the past, why not do some DSP next? As an example, a hamming window.

    To everyone else. Filters can get EXTREMELY complex real quick (as Pete demonstrated). To fully understand filters, you need to know all the plots he mentioned and the Laplace domain. So, take his advice and go to the Youtube channel he mentioned to learn.

  • Good introduction to filters, but there’s more to even the simple one-pole circuit shown. The larger the value of capacitance desired (even at 1uF), the harder it is to find and use a nearly ideal actual capacitor.

    An ideal capacitor is unaffected by temperature, voltage bias or vibration. The “COG” ceramic types are excellent, but almost unavailable above about .01uF. Next best are “X7R” or “X5R”, which have moderate temperature sensitivity of capacitance, but severe variation with applied voltage and so must be voltage-rated many times that of the application.

    One advantage of the opamp filter is the ability to use small caps and large resistors and still have low resistance signal-level drive for subsequent circuits. The trade-off can be pass-band noise from the large resistors. Whole books have been written…


This Week

This Month

Heartbleed

Happy Arduino Day!

SXSW Create 2014