According to Pete - August 1st, 2011


It's that time again - the first Monday of the month (yeah - it's August now) and "According to Pete" is in the house. Literally - this is a special edition of "According to Pete" that takes place in Pete's garage. It's kinda like MTV's Cribs, "The Geek Edition." Check it out:

There you have it - a video exploring the wonderful world of audio technology. Once again, leave any questions you have in the comments here, on Youtube, or email feedback@sparkfun.com! We hope you found this edition of "According to Pete" useful - see you next month! Cheers!


Comments 52 comments

  • I really wish this could be every other monday or something

  • For those of us self taught nerds… Do you have a website or book for finding all of those oddball equations that seem to jump out of your head? or where to learn about switching vs linear power supplies? (currently Google is my friend, but sometimes not very helpful, or too helpful).

    • Oddball? Hmm… well, no, but your point is taken. This sounds like the topic of another (future) episode, though. I have a handful of equations that I use for off-the-cuff circuit analysis (if you can call it that), one of them being C = 1/(2pifXc) to get capacitance given frequency and capacitive reactance (Xc, in ohms). Another is L = 2pifXl to get inductance given frequency and inductive reactance (Xl, in ohms - somebody’s going to start beating me with phasors pretty soon). A little algebra lets you get the term you want for whatever’s given…
      The different supplies… yeah, I hesitate to ballpark that one. Switching supplies can be sorta complicated when you get up close and personal with them, and I couldn’t even manage to multiply 7*9 right in the video. But basically, switching supplies turn on and off really fast to give you something of an average voltage out, then you filter the bejesus out of them. Noisy. Linear supplies…

      • Linear supplies give you exactly what you want out of them using a network of active components (internal to the regulator), and maybe an external resistor network if it’s adjustable. But they have to dissipate (Vin - Vout)*(total current). So they’re quiet, but inefficient. Does that help at all?

        • Yes, a lot! I think a “handy” equation page may be useful to other newbs like me for a first stop before going to the google/wikipedia/WAG (wild guess) equations.
          Oh, and thank you and all spark fun for many hours of well enjoyed “work”

  • years ago i built an bass amp with the LM12, this was an amazing piece of silicon. paired with an ca3140 preamp stage. this was an awesome, and minimal component high power amp.
    anyways i’d love to see you project finished.

  • qzjake,
    I’m gonna say google is the number one learning tool for… anything. SOme teachers of mine used to debate what is more important : Math or reading skills,,, the answer was READING, because with that you can teach yourself anything. Before the internet we used to actually make road trips to the library of Congress to learn stuff! It’s so much easier now, I have a mechanical engineering degree but other than the Radio shack Forest Mimms books EVERYTHING else I’ve learned about electronics has come from google searches, take the time to learn that skill it will open all doors ;-)

  • Nice Workbenches! I just built a pair of them a while back, and they look almost the same as yours!

  • Hmm… the “According to Pete” icon should be updated – it currently shows some forehead hair that Pete actually does not have.

    • Hey, it’s growing back. One accident with the trimmer…

      • It’s the giant eye balls on the logo that disturb me, I keep thinking I’m about to be the subject of some kind of mind control experiment.
        Wait, what was I saying?
        All hail the Hypno-Pete!

  • Hi Pete probably my calculator is wrong, but it says 663,1 uF not 6631 uF…

    • Cripes, I think you’re right. I’m off one decimal. That makes it much cheaper… but I think I still like it bridge-tied better.

  • Three questions:
    1. Since the center voltage is 7.5v, wouldn’t it just be 15v peak-to-peak? Will the output of the inverting amp ever be below 0v?
    2. I assume that the output of the inverting and non-inverting amp won’t be EXACTLY the same (7.5v) with no input. I suspect there will be a minor voltage difference between them. Do we need to be concerned about that, and if so, isn’t a capacitor the solution?
    3. Why a 63v rating for the cap? That seems like a lot of headroom for 15v, unless there’s something about the mechanics of the speaker than can cause voltage spikes above 15v…

      1. The part is specifically designed so that you can do the bridge, so the output offset won’t be anything to worry about. If it were… we’d have picked another part and/or redesigned the circuit. Technically speaking, a cap would be the solution. But for anything other than a toy noise maker, I don’t like coupling speakers that way.
      2. I’m going from 30Vpp, and a 2X rule of thumb. It may be overkill, but I wouldn’t want to ever have to look at it again, know what I mean?
    • I’ll answer 1. since I think I can.

      Since the center voltage is 7.5v, wouldn’t it just be 15v peak-to-peak? Will the output of the inverting amp ever be below 0v?
      When input = max, out1 = max (15V), out2 = min (0V) (effectively +15V). When input=min, out1=min, out2=max (effectively -15V - (+15 but inverted)). Thus from the speaker POV the Peak-to-peak is 30V.

      • Correct. Basically, the speakers are ‘centered’ (physically) at 0V. Applying 15V bias in one direction is different than applying 15V bias in the other direction. So you wind up with 30V worth of dynamic range, even though you can only generate 15V worth of bias in either direction. One thing I think Pete left out that is causing some confusion is that, if I’m understanding correctly, the input signal is still 0-15V, so the amp assumes a 7.5V center (well, half of supply voltage I imagine) and essentially you get a 2X gain through the amp. So applying +7.5V relative to center on the input gets you 15V on the output, and applying -7.5V relative to center on the input gets you -15V on the output.

        • The input signal is still on that scale, but it’s less than about a volt pp. And that’s capacitor (AC) coupled, so we don’t care what the input to the STA540 is sitting at, DC wise. It’s much easier to AC couple an input than an output because inputs have much more impedance (which is to say AC resistance) associated with them. Each individual amp circuit (there being 4 total) gives you a gain of 10X. Tying them in a bridge gets you 20X.

          • Well, if the pp gain is 20 you do care that it’s always below 1.5V_pp or you’ll get clipping. To double check, you’re basically outputting an ~1.5V_pp signal with a 7.5V DC offset correct? Otherwise how does the amp know what to center off of, since the AC signal input is always >0V?
            I’m sure it’s in the next part of your tutorial, but what are you using for your DAC? Or is this just an amp and not a full receiver?

            • A receiver is defined as an amplifier with a built-in pre-amp (essentially volume control), AND a tuner.
              An integrated amp would be just an amp and a pre-amp.

            • Well, there’s no DAC. It’s just an amplifier so I can plug in whatever I like. And the part expects something on the order of a line level input (or less), which Wikipedia gives as 0.894Vpp. Given you supply voltage and gain, you can back out what the input can and can’t be. Maybe some of this will make it into the next installment…

  • This video makes us want to build a lav mic.

  • First time watching this, I love this guy!! :)
    Sam

  • altought I think audio projects sux, maybe I would like it in the future or something… I mean, tutorials and stuff like this are never much, it’s always good to have more around…
    the thing I didn’t like (at least for me, an idiot morrom) is the fact that you constantly saying stuff that I don’t understand, like “linear something”, I don’t know what is a “linear something” and why it’s better than “not linear something” for certain applications… but it’s probably me just being dumb :S
    that’s why I never played with analog stuff… it’s always cheap to buy something that will do watever you want and/or making your own do not pays the price… Also it normally comes down to some magic IC that does all the hardwork… so you (me) actually get inproductive by doing stuff with caps/transistors/etc.
    Of course together with the fact that I normally hate uncessary math… I like programming, so uC rox =)

    • I’m probably not one to speak… but watching videos like this give an awesome demonstration of concepts. Despite the fact that you(we) dislike audio he really covers a lot of cool concepts in a project format.
      The reason I like it so much is that I just read a lot on electronics and don’t take any formal classes, so watching him reinforces those ideas…
      The great thing about what he is teaching is that you can go look up what you don’t know. I would definitely know because I’m a newbie. :P
      Plus building circuits yourself is much more rewarding than a storebought solution. I think it was great!

      • I know… I mean, everything with him is better than learning at university or something like that…
        Collin from make also likes audio projects and such, it’s nice because these guys show it in a nice way… collin most of times shows diferent stuff in small projects, like tips for smd soldering, tips for prefboard soldering, etc…
        I believe this one from pete is going to look very nice at the end xD… I mean, I’m imagining he is going to have some 2 - 3 ft tall Vu meters or something… and will look cool at his garage :p
        I mean, the fact I don’t like making audio projects (just because I’m fine listening music at my headphones…. like I’m doing while I type) doesn’t take the “niceness” of it =)

  • Awesome stuff. I also must have a good stereo system playing when I’m in my work area. I have a VU type display as well - a VGA monitor displaying 7 bands of 40 bars each, using a PIC16F886 and a MSGEQ7 7 band equalizer chip. Looking forward to seeing what you come up with!

  • built the work bench as well? I hope so.

  • Some how I just cant let go the fact that Pete looks just like the villan in “the italian job”. Lol!

  • Can i use my paypal standard to pay money for the stuff i want to buy from here???

  • Out of curiosity, what is the frame on the garage floor from?

    • Yes, that looks like the backbone of a hot rod :-D

      • It’s a 60’s era Triumph Spitfire, mark 1 or 2. The motor and frame say it’s a ‘62, body says it’s a '68. That’s another one of my projects…

        • http://www.sportscarcraftsmen.com/
          If you haven’t seen Paul’s shop, it’s the best resto shop I’ve been in.Arvada.

          • I have not and that’s awesome! I was looking for a local place to get stuff. Wonder if the can hook me up with a six cylinder.

        • About 5 years ago I test drove a ‘63 Spitfire for a friend who couldn’t drive a manual gearbox. Classic British feel to it. Particularly enjoyed the suspension geometry that caused the wheels to squeal around every fast, hard corner. Would have bought it myself if I would have had the means.

          • Boy that brings back memories from the late 60’s I helped my father rebuild a ‘63 Spitfire he had acquired for a 'bargain’ price. It had been run through a barbed wire fence and the engine was blown. I had to handle all the wiring as he was color blind. I’ve come a long ways since then, of course so has electronics.

        • Oh, cool!

  • 9v x 7A = 63W
    You’re working too hard Pete!
    :-)

    • Derp. That’s all I can say.

    • Well, 9 x 7 IS 63…

    • And I’m nit-picking, but aren’t ATX supplies noisy switchers?
      http://www.youtube.com/watch?v=ksfGJuWoT5w&feature=player_embedded#t=392s

      • Eh, it’s not really nit-picking, but yeah, they are VERY noisy. No one that cares about audio would use one as a supply, for garage or otherwise.
        PS - Nice video Pete ;-)

        • Yeah, what Robert said. I can never seem to be consistent with my shortcuts. I may do things differently at different times of the day, or depending on how much coffee I’ve drank… But in a pinch, I might use an ATX. But I’d try to filter it pretty good.
          And thanks, el Bob.

          • Hmm, class D amps are basically switchers themselves, right? Would love to see an SF audio amp board with one of the TI class D’s (and an AtP video on the basics!)

          • Fun Project. I must have tunes in all work spaces.
            I do suggest you think about taming the echo chamber (garage) with some wall and ceiling treatment. If not an ATX power supply will do just fine you’ll never hear the difference :-D
            Looking forward to the next installment.


This Week

This Month

Heartbleed

Happy Arduino Day!