atmegaschematics
I made a schematic with an ATmega328, a LT1763, and a HC-05 module. I was wondering if I made any mistakes on the schematic. It also has a 4 pin female header for a MPU-6050 module. It might end up being a mini quadcopter someday, but for now I made the schematic as a first step and to learn more about better circuit design. Here is my circuit, you can click for a larger view:
R1 and D1 will not limit the voltage to 20V. The resistor would need to be in series with the +24V supply rail for this to happen.
In any case, why do you want to limit the voltage to 20V if you are limiting the current through the resistors? The current limiting circuit will adjust the voltage across the resistors to the same value either way.
ZA and RA just limit the gate voltage to 5V (or not - I didn't look at the values, Steven is right there), in case you want to interface with a microcontroller according to the Instructable. The only effect this will have on the current limiting is setting an upper limit to the MOSFET turn on (which may be a good thing, but not necessary)
The circuit will limit the current through the LEDs just fine like this.
To show what happens with the circuit over a range of supply voltages, here is a quick circuit I created, without the unnecessary bits:
Simulation:
I set the current for around 20mA using 0.7V / 0.02A = 35Ω.
The X axis is V+ (the supply voltage) being swept from 10V to 30V. The green trace is the voltage across the 3 LEDs, and the blue trace is the current through them.
You can see how once the voltage rises high above the LEDs combined Vf, any further increase has very little effect on the voltage across them or current through them (less than 1mA variance over 20V)
I have answers for all your stated questions:
The first way would be to use a ICSP USB programmer to program the Atmega instead of serial programming, Arduino-style. This is more robust, faster, and avoids the problem entirely, and optimizes for BOM cost for the actual device (at the expense of requiring a few dollars of programmer hardware if you need to actually program it.) This can be implemented even on an Arduino board.
The second way is to run the Atmega328p on the internal RC oscillator, not using a crystal, and use port B instead of port D for the DDS output. This leaves the serial port entirely un-touched.
The third way is to write PORTB and PORTD one after the other, and use PORTB for the two bits that the serial port uses. This will generate a two-cycle glitch at about 4-8 Mhz. Given that you will need a low-pass filter anyway to generate any kind of smooth waveform, you simply need to lower the cut-off frequency of that filter to eliminate these glitches. To properly write the two ports in sequence, load the values into registers, disable interrupts, write the two ports, and re-enable interrupts.
Finally, you could use a small-signal diode (like 1N4148) instead of R17.
Best Answer
Looks good at a quick look through. Nothing obviously bad.
BSS138 are especially wimpy MOSFETS. Ok for playing but do not do justice to the rest of the circuit. Not up to quad copter use, but, you know that.
DTR is intended to reset the processor I assume.
To pass the Olin schematic test you'd want (amongst other things)
To move text strings off lines (eg PS3 PS4 by MOSFETs and rotate them to horizontal where sensibly possible. PWM3 label is overwritten by MOSFET name etc. VCC label in eg FTDI box turn 90 degrees.
Have connectors from FETs vertically above FETs rather than bent at 90 degrees. Vbatt could be above or to one side.
And similar
Aim is to not 'twist brain' to follow obvious flow.
Some people will read diagrams with small points as above and find them less than comfortable but bearable. Others will essentially refuse to look at them. The latter reaction is somewhat extreme but reflects an underlying semi-reality - a really tidy diagram with attention to layout, text and component orientation, consistent labelling and more SUGGESTS an underlying technical competence, and a messy diagram tends to suggest the onbious. Neither of these impressions may be correct, but they are often a useful starting guide. Also, arguably, a tidy diagram reflects respect for those who you expect to use it (and the validity and relevance this has been argued at length :-) ) and is more likely to allow you to spot technical errors.
NB: DO NOT ever take my diagrams as examples of how to do things well. I am a "classic messy" who attempts to force the world's mold on my output with varying success. Judgements based on normal standards may mislead one wrt the value of a classic messy's offerings - and may not :-). Equally a chronic super-tidy (who can in their own way be just as bad as a messy) may produce superb copy but not always match it with underlying competence. FWIW.