astable circuit "squeal" issue

[bluzeyonecat]

Hey guys. So I made an astable circuit to use as a tone generator for some pedal ideas. I made it out of a 555 ic. I've done this many time but something unusual happened today.

After I built it on the breadboard it started out fine. Then it began rising in pitch on its own to the point of this screaming horror. Bad cap I thought. So switched caps and it did it again! Tie pin 5 to and .01 uf cap to ground I thought to stabilize it. Did it again! Grant it, I was using a smaller portable breadboard to build it outside my regular work space but its never given me any problems before.
Anyone have any thoughts whats going on? Thank you for your time.

[Brisance]

I'd bet on stray capacitance from the breadboard

[GibsonGM]

I'd bet on stray capacitance from the breadboard

A good possibility.  Does it stop or the pitch change if you wiggle wires and parts around?    You said you've made this before...same model number chip, and same caps?  You sure your cap values are right?  Are you powering from a battery or power supply?

See if placing a .1uF cap from pin 8 to pin 1 does anything for you (in addition to what you have on pin 5).   Sometimes they need better decoupling - you may even need at add a 1u cap as well as the .1u (in parallel with it).     I use one on most 555 builds to avoid 'spurious oscillation'.  They can be sensitive to coupling to the power supply...long power leads can make this worse.

[bluzeyonecat]

Cool. Thanks guys. Later that night I made it again and it was fine. I was powering it with about 6v the second time. 9v the first. Thats a cool trick with a cap on the power supply. Have to remember that. Any others tips for stability in a circuit? Would really like things solid when I transfer to pcb.
I also recently learned from the King Soloman build about hooking the middle lead on a pot to one of the outside leads. My question is, where do you send the other outside lead? I presume to ground? Thank you for your time.

[GibsonGM]

Awesome, bluze.   Altho the supply voltage shouldn't matter as long as you're within the proper range for the 555 (?)

Decoupling the power supply is a major deal in electronics in general, but even more so for circuits that ask for rapid current draw like an oscillator.  The sudden demand for that current often leads to ticking...so one more trick for something like an EasyVibe, with an LFO, is to place a 100R resistor in series with the power supply, followed by ~1000uF of capacitance to ground.  This creates a reservoir to 'feed' those spikes, effectively decoupling.   

What you experienced may have been feedback....wire runs/lead lengths can allow signals to couple into places they don't belong (input to output stuff), and non bypassed power supplies also 'react badly' with oscillators.  The 555 is pretty good about this as long as you put in the caps I told you about....the LM386 amp chip is notoriously bad for it.   Think of that phenomenon as signal jumping from wire to wire (via stray capacitance) and shooting around in a loop.  It goes thru the power supply, and the decoupling cap(s) allow an easy path to ground for this unwanted signal to take rather than to keep running around the loop....sorta    There is more to it, but for our purposes close enough.  Use the tricks and things work out fine! 

>>Pot as variable resistor, or rheostat....say you connected pins 1 + 2 together...you are sending pin 3 to ground or what have you.   The pin 1 + 2 are really just one connection, and you send that to whatever voltage you are working with.   You connect the 'unused' end lug to the wiper just in case the wiper contact becomes dodgy over time (and it probably will eventually).  That would result in really nasty noise....the 'unused' lug assures connectivity in this case.  Good practice to follow.   

Try it with a battery and LED or something to get used to using them, but remember: you MUST have a limiting resistor in series with the pot for 'significant voltages' or you run the risk of burning out the pot due to exceeding its power rating.   9V battery, you turn pot down to 200 ohms, you get 45mA....9v*45mA=.4W, and most pots are rated for .25W!   The pot burns up.       So I'd suggest always having a 390 ohm resistor in series with the pot if you're messing around at 9V....and that will also protect the LED if you are using one.   Many schematics don't have a limiting resistor, as the one who designed the circuit knows you don't need it (small signals and bias voltages often are small enough to cause no harm).   

Here is how they wired a pot this way to control a small motor to show what's going on:
http://www.allaboutcircuits.com/vol_6/chpt_3/7.html

[PRR]

> it began rising in pitch on its own

Electrolytic timing cap?

If installed reverse-polarity, it will be nominal value at first. Then as the reverse polarity decays the oxide insulation, capacitance and leakage will change, probably in the direction of higher pitch.

[GibsonGM]

Great point, Paul.  ^^

[bluzeyonecat]

You guys are awesome and this place is the best resource I've found to learn. Thank you very much.
So what is the functioning difference between electrolytic and ceramic capacitors?

[Brisance]

electrolytes are polarised and have other problems, but are the best in size vs psyhical size, other types are not polarised and should be preferred , if possible by size.

[bluzeyonecat]

I see. So they do function the same. Was just wondering mainly because some schems. call for them.
This maybe a little off topic but does anyone have any good schematic ideas for making a low pass filter using a 741 op amp (only one in my arsenal right now)? My objective is to plug my synth into it and use it to do sweeps. Or dial a patch tone. Thanks again for all your help.

[antonis]

If you need a simple low-pass filter (1st order) you don't even need a schem..

Just connect the non inverting (+) input (pin3) to a cap (other side to GND) and to a resistor (other side to signal)..
(Cut-off frequency (6dB/oct) is determined by the formula: f = 1/6.28*R*C where R->Ohms, C->Farads and f->Hz..)

If you don't want any amplification, just wire the out (pin6) to the inverting (-) input (pin2) otherwise connect pin2 (-) to pin6 via a resistor (R1) and to GND via another resistor (R2)..
(Voltage gain is determined by the ratio of the resistors R1/R2..)

P.S.
If you interchange resistor and capacitor you create a high pass filter and if you cascade it with a low pass setting (free resistor's side to pin6) you create a band pass filter...


edit: the more I read the above the more I think that you certainly need a schem...

[Brisance]

I see. So they do function the same.

Yes, but electrolytics have a shelf life and some other undesirable characteristics as well as you have to be careful that you connect them the right way or they will fail.

[GibsonGM]

Electrolytics exist MAINLY because they are cheap to make in values over 1uF.  It's hard to find other types of caps in values over that.

Most ppl use poly-type caps for values <1u.  Ceramic for values <1n.  And electrolytics above that.  So, if you CAN, you might try not to use them, but you'll find you are 'funneled' to them.   I don't think they're so bad, if you use them properly.   The non-polarized type is better than the polarized, but sometimes you just use what you have.   And like PRR said, you need to think about how you're putting them in - they can be sensitive about orientation and you can get odd effects (like the squeal perhaps!).      Over time, they do dry out and require replacement....10 to 40 years, it seems

Bluzey - you need THIS!  Goes with Antonis' description up there:   
http://www.ti.com/lit/ml/sloa091/sloa091.pdf   

Download, keep, read, and even if it's still mysterious now, in time it will make GREAT sense.   Look up the filters, page 12, LPF, HPF....the formulas are right there to determine where they cutoff.   When you see "||"  it means  "in parallel with"...like, R1||R2  is R1 in parallel with R2.     I know you'll get some TL071's and 72's soon, instead of the 741, ha ha...but it WILL work for now. 

If you get some simulation software like LT Spice, Tina etc., you can view how this all works graphically; well worth it, IMO.   

[bluzeyonecat]

Woah! Huge score! Grazie! 

[GibsonGM]

You're welcome

[PRR]

> what is the functioning difference between electrolytic and ceramic capacitors?

Capacitor is two conductors separated by an insulator.

You can use "whatever" for the insulator. In your kitchen you might use aluminum foil and wax paper. This used to be very common, because thin paper was available and wax or oil keeps dampness out of the paper and fills the holes.

You want the insulator as thin as possible, but not so thin it breaks-down when you put voltage across it.

Various plastic "films" (like Saran Wrap) have replaced paper/oil caps in most small applications. (Oil caps still have uses in the trash-can size caps used in utility power distribution, and some hi-fi folks say they have "smooth sound".)

Glass works good for very high voltages. A modified glass is used for the super-thin insulator in a MOSFET.

Mica (a flaky rock) works very well into radio frequencies.

Ceramic technology has become pervasive in values from 3pFd to around 1uFd.

All of these depend on reducing "large stuff" (a vat of plastic or ceramic) down to "very thin". There's limits how thin you can roll stuff and it still holds together. But we want really-really thin insulation for maximum capacitance for the size/cost.

The other way to make an insulator is to "grow" it. Take fresh Aluminum and leave it out in the rain. It grows white crusty stuff, Aluminum Oxide. If you grow it very carefully with the right chemistry and some electricity ("electrolytic"), you can grow a super-thin of excellent insulation, exactly as thick as you need to avoid break-down at a given voltage. The super-thin insulator means you get LOTS of capacitance. So much that it is hardly worth making an electrolytic capacitor smaller than 1uFd, but you can build them for hundred, thousands, even a million uFd, at costs far lower than film caps.

But there is one big "problem" (and several smaller problems) with electrolytic capacitors. They grow the film with electricity. If you apply reverse polarity, it "un-grows". The insulator re-dissolves and the capacitor becomes an electric short. This rules-out electrolytics for some applications (although clever design is bringing electrolytic benefits into uses like motor-run capacitors). In most "electronics" this polarity sensitivity is something we can deal with. We often have a known DC polarity across our capacitors. When used to filter DC power we know which side is + and -. When used on a '555 timer we know (or should know) which end of the timing cap is more positive. Put the cap that polarity, and it may work for decades. E-caps 50 years old are sometimes still doing their jobs.

[bluzeyonecat]

Thank you fellas very much! Between explaining and the schems. it really helped alot.
Very interesting that simply switching the cap and resistor changes a low pass to a high. One small question with the schem. above, If I wanted to control the amount going through, would I put a pot on the first resistor by the Vin? Or in series with it? Thank you all very much.

[GibsonGM]

We're going off topic, so maybe if we go further we should start a new one, like "Opamp filters for newbies?" or something?   So others can know what we're talking about in here, and maybe join in

But for your question, Bluze....if you go thru the pdf I linked to, you'll see that in basic opamp operation, you have your gain set by "R1 and R2" - using ANTONIS SCHEMAS ABOVE (not the pdf, they may list different resistor numbers)....they're the feedback resistors.   The gain is equal to 1+ R1/R2....so, if R1 is 100k, and R2 is 10K, you get a gain of 11.    A 1V input will get you an 11V output, as long as you have a power supply able to take you there (15V or so).      If not, you clip (distort in ugly way here).   This isn't the whole story, but much of it (impedance worries come later...).   Look at stuff like Tube Screamer or Dist  + to see this in action.

And no, you can't go 'rail to rail' with most opamps (you can buy ones that do), so you need some room for the opamp to operate, too...so you'll clip earlier than you think with say a TL071.

To control the level, most folks would make R1 a pot so as to be variable.  A 100k pot can then be full on for gain 11, or turned down to 50k for gain of 6, and so on.  Feel free to add a "high" resistor, and then try the "low" one, to find the range you want to operate over - then replace with a pot.   

Adding a 'volume or level pot' to the front end of a circuit often just creates background noise...so you tend to see them more on the output side after the signal has been normally amplified above the noise floor.


A big part of design is to set out your parameters, and know what you're going to work with: "My input signal may be from .3VAC to 1.5VAC" and so on.   Then design around that....power supply is 9V....so don't set up the gain to be more than 3 or so to avoid clipping.    A good way to deal with variance like that is to make your 1st stage a buffer...feed the filter....and possibly even buffer the output too.     And if you have wild variance in input, you might make a "low" and "high" setting, to switch R.  Or just a pot, and live with needing to turn it down for hot signals (I rarely do much more than that since i'm not selling anything, these are just for me and friends who can live with that, and we're dealing with little variance, really).

Keep it simple, play around, and you'll see these things come together pretty soon, then they'll make sense!     

Re: switch R/C order and get HPF to LPF - it always helped me to think of "sending highs thru circuit" or "sending highs to ground" - made it more intuitive....

[bluzeyonecat]

Thank you Gibson! Really helpful info. I did download the pdf by texas instruments. If been going through it slowly grabbing up as much as can. A new thread is a good idea. I'll set one up. Thank again for your insight.

[bluzeyonecat]

To get this thread back on track concerning astable circuits, I have the 555 basics figured out but why is my 556 so grouchy!? Lol!
I get that it is two 555s but whenever I use one side of it, EPIC FAILURE! I made the Atari Punk Console (sort of) out of two 555s and it worked so good, it drove my wife nuts! Lol! What am I missing with the 556? Trying to do the same build with it. Ive checked and rechecked my wiring, used same value components. Any help is much appreciated.