Happy new PWM circuit

[electricteeth]

So simple yet it works... ...very well! Anything i plug into it comes out sounding like a gameboy. love it. enjoy. (ps i know that there are alot of pwm circuits out there but this is my own take.)

[Earthscum]

Might want to make mention of the chip used, transistor, etc.

[iccaros]

plus what are you using to generate the original 300hz?

[gritz]

Might want to make mention of the chip used, transistor, etc.

Looks like a 555 to me. And a vanilla npn.

[Earthscum]

Might want to make mention of the chip used, transistor, etc.

Looks like a 555 to me. And a vanilla npn.

Aye, but not everyone knows it's a 555. Also, does gain play into this? did he try different transistors, or is it a good "plug N Play" design? Also, WHICH 555? If you recognize it as a 555, you should also know that CN's sometimes don't play the same way that other 555's do in some circuits. That's all I was getting at. Looks simple and all, but I kinda want to know more about the circuit (and his experiences designing it, if he wants to share that  )

[electricteeth]

Regular 555. Vanilla transistor. I figured it was obvious. sorry. 300hz is just low voltage ac signal. falstad simulation. The transistor also has a cap from collector to base in order to cut the amount of frequencies being amplified. Gain plays into the circuit yes. The transistor has to have a high enough output to make the triggering happen. If you cut the .22uf from ground it makes for a yucky hi pitched fuzz. This isn't really an original circuit as much as it is two simple circuits working together. can't take credit for any design work except for deciding to run a booster into a monostable circuit. Since it is a monostable circuit, a more squared off output at the collector of the transistor will possibly work more smoothly (clipping diode?). Anyway, pretty simple tricks with radio shack stuff.

[earthtonesaudio]

A couple issues: the trigger pin has no DC bias.  This works in the sim but is unreliable in real life.  Also, it will hurt the IC to put the control voltage directly to 9V, I forget at the moment how much current is safe for that pin, but it's in the data sheet.

I like 555 noisemakers.  Hope it turns out awesome!

[electricteeth]

sounds fine to me. I have the board built but i'm waiting for zee enclosure in the mail. Looking at the data now, the transistor should be adjusted to put out more than 1/3 vcc  (over 3v) when on i believe. Pin 5 it says on the data sheet can receive the full Vcc. I may be misreading it though. I am relatively new to this.

[Earthscum]

It's what is called "open collector", so basically what is happening when you rotate your pot to the V+ side, you will be running voltage straight across a transistor to ground. A 1k resistor between the end of the pot and V+ will keep you from dumping a bunch of current unnecessarily. As far as what that internal transistor can handle, that should be in the datasheet. You will want to look for current rating, not volts. A 1k with the entire 9V across it is only dumping .9mA, much better than whatever your battery can dump across the wire resistance and whatever the transistor allows. I tried digging through the datasheet but didn't really come up with much, either. I really don't know a whole lot about these chips.

[PRR]

> what are you using to generate the original 300hz?

I think that's his guitar. Either he drones 300Hz all night, or the actual frequency is unimportant and 300Hz is just for-example.

I have this feeling the input amplifier needs more gain, higher impedance, and maybe a more-certain bias, to give reliable triggering in real life. Might just go to a '741 or similar opamp set up non-inverting, gain of 10-100, plus band-filtering to dump bass and smooth highs for more musical-like triggering.

On an original 555, pulling pin 5 Ctl to Vcc does no harm. Ctl is the upper tap on the resistor divider plus a base in the threshold comparator. Take it all the way up, it stops working as-expected, but no harm is done.

On an original 555, leaving Trig without a DC path "should" be the same as a pull-up. I agree this is bad habit; also a resistor here is VERY useful for defining trigger pulse width.

Even the second-source 555s had some differences from the original. Now that we are down to 7th-generation re-makes, probably best to define everything and be prepared to tinker.

[DavenPaget]

Paul , "he's" a she .

[electricteeth]

Again, it sounds fine. Alot of times its hard for you guys to believe that it sounds fine when not using conventional stuff or playing by the rules. Its also hard to believe with no sound clips. Once i get my enclosure together i will record some soundclips.

[Earthscum]

It's not an argument of what it sounds like, by any means. People are just giving you suggestions, that's all. If you don't want to use them, that's fine. I'm finding it interesting, though. Don't get too stuck on "It works just fine". Sure, it sounds great, but I think people are pointing out more longevity and stability points here.

I have a couple circuits that I thought were ultra awesome, until I started putting them in different situations. Flaws became readily apparent, lol. One is an SCR based gated square wave shaper. Works GREAT on it's own, but does not play nice without a solid buffer on either side (unless you're running passive pickups straight in, the way I designed it).

So, speaking on my behalf, no offense meant, and definitely not knocking your circuit. I'm gonna give it a shot, but all I have is a CN and a 556. I gotta figure out what to do with the other half of the 556 (do I ground a couple pins, leave everything floating?, or...) or wait til my Tayda order gets here... ugh, I hate waiting on parts.

[DavenPaget]

It's not an argument of what it sounds like, by any means. People are just giving you suggestions, that's all. If you don't want to use them, that's fine. I'm finding it interesting, though. Don't get too stuck on "It works just fine". Sure, it sounds great, but I think people are pointing out more longevity and stability points here.

I have a couple circuits that I thought were ultra awesome, until I started putting them in different situations. Flaws became readily apparent, lol. One is an SCR based gated square wave shaper. Works GREAT on it's own, but does not play nice without a solid buffer on either side (unless you're running passive pickups straight in, the way I designed it).

So, speaking on my behalf, no offense meant, and definitely not knocking your circuit. I'm gonna give it a shot, but all I have is a CN and a 556. I gotta figure out what to do with the other half of the 556 (do I ground a couple pins, leave everything floating?, or...) or wait til my Tayda order gets here... ugh, I hate waiting on parts.

NE555CN or LM555CN ?

[Earthscum]

Apparrently MIA555CN, lol... it was a RS one, so I guess that doesn't narrow it down much.

[Earthscum]

DOH! Found it... not CN (those were the 741's, and I can't remember why that made a difference, but different subject)

They are TLC555CP.  I just bummed myself out though... found NE55... ugh, 32. I got all excited thinking I may have forgot about one or something.

See, told ya I don't know much about these, lol. I think the last project I tried out was one of Rob H's. That was about a year ago. I just remember a bunch of us found out the CMOS 555 needs different impedances to work correctly, and some things it apparently doesn't work on. I think ticking was the big problem with them? Anyways...

ETA: BOOYA!!! While digging for a transistor, I found an NE555P... sweet. I guess I DID forget about one!

[electricteeth]

So, speaking on my behalf, no offense meant, and definitely not knocking your circuit. I'm gonna give it a shot, but all I have is a CN and a 556. I gotta figure out what to do with the other half of the 556 (do I ground a couple pins, leave everything floating?, or...) or wait til my Tayda order gets here... ugh, I hate waiting on parts.

I found the monostable circuit in a forrest m mims book, there is another circuit for a frequency divider that uses a 555 so you could use the other end of the 556 for that and have an octave down pulse as well. maybe? Also, You are right, i definitely want to put more time into this circuit to give it longer life. I think i will forgo the transistor for an opamp boost stage as well!

[DavenPaget]

DOH! Found it... not CN (those were the 741's, and I can't remember why that made a difference, but different subject)

They are TLC555CP.  I just bummed myself out though... found NE55... ugh, 32. I got all excited thinking I may have forgot about one or something.

See, told ya I don't know much about these, lol. I think the last project I tried out was one of Rob H's. That was about a year ago. I just remember a bunch of us found out the CMOS 555 needs different impedances to work correctly, and some things it apparently doesn't work on. I think ticking was the big problem with them? Anyways...

ETA: BOOYA!!! While digging for a transistor, I found an NE555P... sweet. I guess I DID forget about one!

Yeah ... don't use the TLC555 for this ... save it for stuff that WANTS a ICM7555/TLC555 to work properly without annoying artifacts .
IT does good to separate your CMOS Timers, Hex Inverters from your opamps and your normal timers and decade counters etc that you think you won't be using ...
I separate mine using cardboard sheets because there are 3 guides to put plastic sheets ... plastic sheets ? No thanks , i'll make my own .

[Earthscum]

So, I was scoping the output of this and noticed that the pulse does vary a bit by frequency, but I am using a bass. That's kinda neat. I don't get any variation with Escebedo's PWM (well, negligible, anyways). But, I did have to bump to an MPSA13 (darlington) to get a decent triggering out of it. I ran 470k C-B and 100k B-E (gnd). With the 10k C resistor, put me about 6V. I noticed, also, that the sensitivity does go down as the pulse is shortened. I think this is one of the effects caused by the issues that Paul mentioned.

I think the OPA would be a good idea. More gain, and a bit more controllable. I'm digging on the varied width, though. It looks to be fairly stable. On bass, this makes a difference... like how it's harder to wah a bass. Other PWM's are pretty solid in the pulse width, so G string sounds great, but E and B sound sooooo thin that you can hardly make a distinct note of it in a full song setting. Or, B would have a decent sounding width, but about the G string and up it's pretty much square 50/50 duty cycle. I can't get this to quite go to a square wave (no gripes there, lol).

[DavenPaget]

So, I was scoping the output of this and noticed that the pulse does vary a bit by frequency, but I am using a bass. That's kinda neat. I don't get any variation with Escebedo's PWM (well, negligible, anyways). But, I did have to bump to an MPSA13 (darlington) to get a decent triggering out of it. I ran 470k C-B and 100k B-E (gnd). With the 10k C resistor, put me about 6V. I noticed, also, that the sensitivity does go down as the pulse is shortened. I think this is one of the effects caused by the issues that Paul mentioned.

I think the OPA would be a good idea. More gain, and a bit more controllable. I'm digging on the varied width, though. It looks to be fairly stable. On bass, this makes a difference... like how it's harder to wah a bass. Other PWM's are pretty solid in the pulse width, so G string sounds great, but E and B sound sooooo thin that you can hardly make a distinct note of it in a full song setting. Or, B would have a decent sounding width, but about the G string and up it's pretty much square 50/50 duty cycle. I can't get this to quite go to a square wave (no gripes there, lol).

Still wondering why commercial bass effects are so few and are mostly the jellybean ones , overdrive , chorus , eq that's all .
This PWM needs to be re-tweaked for bass . Well i could try but i don't play bass or own a bass .