New to Lunetta with some questions

[Steveg]

Just a thought ... those bolts are insulated from the panel aren't they?

[dadinfinitum]

[Steveg]

No, all good!

I just missed where you said they were 3d printed when I re-read it.

[dadinfinitum]

I'm finishing up a simple 40106 osc bank, and have some plans for future pieces, but I have some general questions (I do feel like I've absorbed a good bit).

1) Reading this great post, is using a current limiting resistor always a good idea on an output?

2) I learned on my own (before reading about it, hah) when a good use for a pulldown resistor would be - I had a switch for controlling an oscillator, but when I'd turn it off, it'd sometimes be high, sometimes be low (makes sense now). When are other good uses?

3) How often do you want to use an op amp (like TL074) on outputs? I see it in a number of schematics, but is this just for listening? If I have a mixer at the end that uses an op amp, do I need to worry about it?

I also have some ideas and am unsure if some builds already exist. Namely, I'm looking for a PWM and a chorus effect (ideally separate pieces, so I can understand them).

Thank you again everyone for your help and ideas. I'm loving this!

[dk]

There are certainly more knowledgeable people here than me, but:

1) Generally, yes. Since a lot of the circuits here are "CMOS chips wired directly to the panel", you could theoretically get away with not having them, but tossing on a 1K resistor on the output can save you from replacing chips whenever you accidentally patch something inappropriately (ie output to output), which inevitably happens to everyone at some point or another.

2) That's more or less all they're used for. You can also think of it as the pulldown presetting an input. Some inputs you might want high or low when nothing is plugged in... pullup or pulldown resistors make it happen. Such is the case where you use a chip that will only output signal when an enable pin is held high, etc.

3) It depends on the circuit. Circuits with digital ouputs (either high or low, for simplicity) don't need them. In a bigger, more complex system (ie not a Lunetta), you often see them as comparators on the inputs to condition the signal for the digital chips following them, but that's not a problem you'll run into much here. For circuits that spit out analog voltages such as non-square/pulse waveforms, CV, etc., they're a bit more necessary. Most CMOS chips will neither accept nor output non-digital signals, so op-amps are better suited for this role. The common exception to this are the 4049 and 4069 inverter chips, which can be set up to behave like opamps (although they tend to get warm and use a lot of current to do so).

PWM - Are you looking to modulate digital signals or analog? For digital, something along these lines is common. Specifically, U2a, U2c, and the cap and resistors between them. For analog, the simplest is to use an op-amp as a comparator. One input sets the threshold, the other is where you put your signal in. Whenever the signal is above the opamp will turn on.. when it's below, it'll turn off. A more in depth read can be found here.

[Steveg]

Most CMOS chips have similar power limitations so if most of your inputs are CMOS you can potentially leave the current limiting resistor off of the output and only add it to inputs where it is needed. HOWEVER, if you do that you must be careful to never connect two outputs together! Generally it is safer to put a current limiting resistor on every output that goes to a patch bolt just in case.

A CMOS chip draws most current when it switches state. If your input is floating it can oscillate across the on/off boundary in response to any stray electrical field. Keeping inputs tied to either a proper output, or a pulldown (or pullup) resistor keeps the chip stable, especially for inputs like CE (chip enable).

Most Lunetta circuits do not need buffers. However if you are feeding some circuit that needs more current than the CMOS chip can supply then buffer your CMOS output (or the input of the following circuit) with an op-amp. However for Lunetta usage maybe a single transistor wired as an emitter follower would be simpler. Otherwise the only op-amps will be your mixer and maybe a downstream filter or effects circuit.

There are PWM oscillators in these pages. Chorus effects usually use bucket brigade delay chips. If you want one you won't find the circuits on these pages.
It is possible that you could make a rough square wave chorus effect using a few different variably clocked shift registers. Don't expect hi-fi!

[dadinfinitum]

Wow, dk and steveg, thanks so much for the info. Again, learning all the time.

So, the takeaways, generally speaking, are:
1) current limiting resistors keep things safe and are good to have
2) pulldown/pullup resistors are primarily for default states
3) opamps before an output are good/necessary for analog outputs, not necessarily digital outputs

(all of this with the usual disclaimer of read the schematic/datasheets, and try it out if you're unsure)

[PHOBoS]

my 2% for what it's worth

[blue hell]

And as a side note: pull-up / down resistors on inputs can be zero ohms.

[dadinfinitum]

[PHOBoS]

yes, you can just power it from +12V and GND. What you will have to watch out for is negative voltages on the inputs which you can get rid of with a single diode.
You can also add another diode or use a zener for that matter to protect inputs from overvoltage.

As for outputs, if you power a CMOS chip with +12V then the output voltage will be close to that already. If however you use a lower voltage and need to boost it
you can use an opamp configured as a comparator which you power from the voltage you want the output to be.

[dadinfinitum]

[PHOBoS]

sure, take your pick.

The left one is the most basic one which will only protect against negative voltages not overvoltage.
It works fine if your whole system uses the same voltage but not if for example your CMOS circuit is powered with +5V
and you connect +12V (or anything >+5V) to an input. There is also a voltage drop caused by the diode but it's only about
0.6V for silicium and you can reduce it to about 0.3V if you use a schottky diode.

The middle one does protect against both negative and overvoltage, I would recommend using schottky diodes for this method.
note that this is often already integrated into CMOS chips but that's only for very low currents and to protect against electrostatic
discharge.

The right one also protects against both with the help of a single zenerdiode but can require a bit more current to function properly.

---

CMOS input protection.gif
Description:
Filesize:	20.54 KB
Viewed:	187 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---