40106 Oscillator Questions

-minus- · Joined: Oct 26, 2008 Posts: 787 Audio files: 13

I'm starting to try out different capacitor values on the 40106 IC. It seems that creating oscillators is the heartbeat of a Lunetta system. Using 1uF caps I am getting very slow on/of pulses at the lowest pot setting. 33nF increases the slowest setting significantly! I'm wondering if it might be worth building say three banks of six oscillators for various tasks: LOW, MEDIUM, HIGH. The slow ones would be useful for clocks. Or is this overkill? 18 oscillators! Perhaps a switch on each oscillator is a better solution to select one of three capacitor values.

I also see people using the 4093 as an oscillator. What's the deal with this? Is it the same as a 40106 but with an extra pin on each oscillator input to send a logic on/off to in order to activate (gate? is that the correct term?). This could be fed from a 4040 for example...righty right fellow Lunetta Droogs?

Also what is this I hear about cross talk? Is it wise to use only four of the six oscillators on any one 40106? How much of an issue is this?

If anyone has recommendations regarding the use of capacitors with this IC it would be much appreciated.

Thanks! Happy Lunetta-ing!

electri-fire

electri-fire

-minus- · Joined: Oct 26, 2008 Posts: 787 Audio files: 13

Ah, of course! Yes, using the 4040 would give me those divisions of half etc. That way I could run the oscillators at a high speed and tap into slower rates from the 4040.

Four oscillators doesn't sound like much to me, but when you consider the divisions you can make, it probably is a bit over the top thinking I need 18!

Just wondering if you used 4 because of cross talk issues between oscillators which I am starting to read about?

Thanks for the quick reply!

....And a second response before I replied to the first one! Well, it's one thing at a time for me... something I never quite mastered! I'll no doubt get onto all that S&M naked oscillator stuff you mentioned! Laughing

Have a fun weekend away!

electri-fire

electri-fire

Come to think of it, how about giving each of the two NAND inputs a capacitor and a dedicated potmeter from the common output? Or just one pot with a small resistor in series on one of the inputs to make a small offset? Might be interesting.

Draal · Posted: Fri Jun 25, 2010 6:57 pm Post subject:

I think I am going to use electri-fire's method of hardwiring my 40106's to a couple of 4040's. I'm a minimalist at heart; both in art and my own lunetta project, so if I can achieve my sonic goals with less clutter the better. These 2 chips were made for each other.

RF · Posted: Fri Jun 25, 2010 7:44 pm Post subject:

Hardwiring them is handcuffing yourself...in my opinion..
I'd say, add some outboard dividers or counters etc if you want slower divisions - and patch them in when you need them. It can be useful 'cause then the divisions are related to each other.... Also, a single SPDT switch can offer three different cap values to a single oscillator.
I've used all six on a chip with no problems - except sometimes they seem to want to synch a bit when the freq's get close...but that might be considered a feature, not a problem...right? Smile

All that said - do what you feel like doing - no rules here Smile

bruce

Rykhaard · Joined: Sep 02, 2007 Posts: 1290 Location: Canada

Crosstalk ... imagine being in a conversation with your girlfriend/wife/mate and their mother, with both of them being the VERY talkative type. When things arise in discussions that you'd like to comment on, it could be hard to get your own word in, edge wise. As most of the time, 1 of them wont stop talking and possibly the other or the same, will overide what you're trying to say, to get one or both of their attentions.

Surprised

Now - apply that theory / sicheeEHshun, to a bunch of oscillators, as trying to oscillate at the same time. There COULD be, only so much 'current' available for all of them to run FROM. What happens when a couple of the oscillators are babbling away, and another one tries to 'speak'?
Why - there may not be enough current for him TO 'speak' away, causing him to not quite be 'heard' properly ...

Or in other words ... in order for him to just operate, he's pulling the current that he normally requires from a the power source supplying all of them. THIS, can start taking away current from the others who are already babbling away ... causing ripples in your power supply's current! THIS, will then reflect back across your ENTIRE SYSTEM ... and into your ears. Surprised

This is where I've mentioned the requirements of the buffer caps, on each of the chips, in the other thread - to try and keep as much current available for each of the chips, as SOON as it's needed.

Now - there can be other crosstalk troubles happening within the system, that are past my limited theory knowledge, for me to be able to properly deduce / etc. But this simple application of buffer caps, can help, quite a bit. Smile

As to the girlfriend / wife / mate and their mother? Surprised

I gave up, years ago. Surprised

-minus- · Joined: Oct 26, 2008 Posts: 787 Audio files: 13

Ha ha! ...the girlfriend / wife / mate and their mother: That's really hilarious!

So I guess you are a single oscillator built on a 555 now. I love the way you describe electronic theory Rich! You should write a book on it. I think it would be in the format of a collection of short stories or fables. Even those who had no interest in the subject could pick up your book and get a good read/laugh out of it! Thanks for the education, once again.

Rykhaard · Joined: Sep 02, 2007 Posts: 1290 Location: Canada

-minus- · Joined: Oct 26, 2008 Posts: 787 Audio files: 13

Experimenting... I've put together 6 oscillators on one 40106. Im using 5.8K (couldn't find 4.7K as my house is a Lunetta-mess) resistors and 500K B pots. I had 1M pots on it but they were audio taper and the linear ones seem better for this application. The caps I am testing are varied. I have a surplus of switches I acquired really cheaply so I thought I'd do a HIGH/MEDIUM/LOW oscillator module with 6 selectable switches, one for each oscillator. I'm trying 2.2uF caps which give a very slow tick to a growl. 39nF goes up into the inaudible when fully dialed... you know, the shit dogs hate! (Hmmm... might have to put out an album for dogs one day.... or for people who hate dogs) Anyway... There seems to be an overlap here. Not sure if the medium oscillation speed is necessary as it kind of gets covered with the other two settings. I shall report back findings when I have something together... and something constructive to say!

-minus- · Joined: Oct 26, 2008 Posts: 787 Audio files: 13

RF · Posted: Sun Jun 27, 2010 5:18 am Post subject:

"spdt center off" is the one you want...

For caps in parallel = the total capacitance is the sum of the individual capacitors... so in the center off position a single caps value is used...in each "on" position it's the sum of the two....

Sorry for the confusion.... Embarassed

bruce

Rykhaard · Joined: Sep 02, 2007 Posts: 1290 Location: Canada

Just in case, for further explanation - (Bruce had burbled out the exact same way that I had thought, of doing it.)

Have the output from the 40106 (for that type of oscillator), coming through it's pot and extra resistor ...

(1k minimum, for an 'at least, resistance' - otherwise, the oscillator could stall, at zero resistance)

... coming to the input of the 40106, that has the capacitor to Ground. At this point, attach a wire from there, to the center lug of the ON/OFF/ON switch.

Now - attach another value of capacitor, to the top or the bottom lug of the SPDT Switch, with the other end of that capacitor, to Ground.

Finally - attach another value of capacitor, to the last lug of the SPDT switch, with the other leg of THAT capacitor, to Ground.

As to the values, you'd mentioned being pleased with the 2.2uF and the 39nF (0.039uF).

The values that you could use, to always have a difference may be:

0.039uF
0.068uF
0.1uF

These values together would give your switch:

UP = 0.107uF
CENTER = 0.068uF
DOWN = 0.168uF

All 3 caps would never be added together this way. I'm thinking that an even wider value of them all, may be better ...

0.15uF
0.039uF
1uF

This would give you

UP = 0.189uF
CENTER = 0.039uF
DOWN = 1.039uF

There. That would cover a fairly decent range. The largest of the caps, you could go up to 2.2uF to end up with 2.239uF for that low clicky / growl end of things, with the center selection, being the one that passes human hearing. Smile

(The dogs also, may not notice anything past that as the sounds made will only go as high in frequency as that which your speaker(s) is/are capable of. Wink

There's another way to do it by combining capacitor values in serial and parallel hookup, but we don't need to get into that math, for this. Smile

-minus- · Joined: Oct 26, 2008 Posts: 787 Audio files: 13

THANKS HEAPS RF AND RYKHAARD!!!

It makes sense now! I was really stumped on this problem. I almost thought I should just go with one cap value and divide an oscillation with a 4040 if need be.

I shall play around with the cap values and see what works best. Those switches I seem to have a load of are DP3T ON/CENTRE OFF/ON...I may just use them... This is the time I should learn this theory of series and parallel capacitors I think.

I shall post results soon. Once again, thanks for your help! This is the great thing about this forum... being educated! And it's a lot cheaper than University! Smile

...now where's that giant bag of capacitors....

-minus- · Joined: Oct 26, 2008 Posts: 787 Audio files: 13

OK.... I went with these three capacitor values: 22nF, 180nF, and 4.7uF. I had soldered in 2.2uF caps but replaced them with the 4.7uF ones to get an even slower clock. Must be clicking along at one click (Hertz?) a second or less at its lowest setting. The switches on each oscillator are REALLY useful! I have this hooked up to the Melody Generator and with a click of a switch it changes everything drastically!! With 6 switches there is tons to play around with. I'll post a photo and sounds soon. Thanks for your help electro-people!