5v Linear Regulator bypassing

So, for my CupCakeStrap I ordered all the components from mouser and digikey and hand made them. Now, being "smart," I ordered a different 5v Linear regulator... it regulates a negative current, and not a positive one. Whoops. I blew up one SMD2.3 with it (pretty sparks and smoke!). I happen to have a SMD1.2, so I'm still ok - I just attached a fan to it and it works fine for Z.

Now, I would like to get my machine running within the week, and I don't think I can order the regulators and have them soon enough. A few guys on the reprap IRC have shown me that the SMD2.3 does not actually need the 12v leveled to 5v, and jumping the 5v PSU supply to the regulator terminal allows them to work fine. I've done a lot of testing with it, seems to work fine, even for my oversize motors.

So, that hack works great for the SMD. Can I perform a similar hack for the Extruder Controller?

yes, however be very aware of electrical noise- twist the 5v supply wires, bypass with 10uF and 100nF and 10nF at both ends, run separately from any other wires, etc etc

one of the reasons that it's a good idea to have lots of 5v regulators everywhere is that it keeps the 5v supply fairly clean by filtering noise out. Whenever you run power over distance, it picks up noise which can cause all sorts of subtle problems, and some not so subtle ones too

---

-----------------------------------------------
Wooden Mendel
Teacup Firmware

Ok, I'm a novice and am not sure how exactly to implement that (luckily, I bought extra 50v 100uF capacitors), but I get the idea that Noise = very bad.
How bad is "very bad?"
Chip-killing bad?

I have four of these big 100uF capacitors on-hand, and possibly more of other types one I can get to my school's supplies. Wire braiding I can do! (5v and GND, but ignore the 12v and other GND, or all?)

it can be chip killing, but far more likely is mere chip misbehaviour, as if receiving random signals.

small capacitors are more important than big ones for noise, because big ones are too slow to shunt radio signals.
Any sudden changes in current consumption emit radio signals, so your microstepping drivers will be blasting the stuff everywhere.

Also, while chips take only a tiny amount of current /on average/, they take it in very very short slurps of up to an amp. big capacitors can't keep up with that either.

SO, you need a small capacitor at each end of your cable that can shunt radio signals- ceramic types are best for this.
A larger capacitor (1-10uF) can be handy for helping the small ones beyond the shortest transients, but the small ones are the critically important ones.

---

-----------------------------------------------
Wooden Mendel
Teacup Firmware

So, I should go find some small (sub-1uF) capacitors, and put them between PSU and each board.

Like this?

PSU
|
|--- cap - board 1 --- GND
|--- cap - board 2 --- GND
|--- cap - board 3 --- GND
|
+5v

Ok, nobody else bypass the regulator.
I have not attached any capacitors yet.

Capacitor 6 has exploded on my more troublesome board, and I am worried that this may be the cause. It also puts more stress on my senior project, due in one week.

The capacitors are in parallel to the boards otherwise the boards won't get any power.


PSU
|
|--- cap --- GND
|--- board 1 --- GND
|
|--- cap --- GND
|--- board 2 --- GND
|
|--- cap --- GND
|--- board 3 --- GND
|
+5v

---

---

Darwin clone, Gen 2 electronics, Arduino Duemilanove w/ AtMega328, 5D Firmware, Pinchwheel extruder
[www.codeerrors.com]