Power Supply Noise

I have noticed that people think that when you say 12 volts it actually means 13.7 to 14.7 volts This is only because car batteries run 13.7 to 14.7 volts.
Well 12 volts means 12 volts, is seems that some think they can crank the voltage up to the max the supply will allow and get " MORE POWER " ('Time the Tool man Taylor')

When you try to run a 12 volt 3D Printer from a 12v power supply at 13.7 to 14.7 volts you will note ripply noise will go up because you are running at the voltage to the max,
"NO HEAD VOLTAGE" when you have the voltage at max every time your system need a little more it throws a noise spike, with no more voltage to give you get noise spikes.

The same will happen no matter what voltage your supply was designed, when you try to get more voltage than the supply can deliver you get into into noisy voltage.
So Please keep your supplies adjusted to the voltage and current limits which they were designed to operate.

A noise free 3D printer is a happier 3D Printer.

---

Computer Programmer / Electronics Technician

also consider that voltage under load might be a little bit higher as well.
it depends on what type of power supply used as to the voltage. anyway a good power supply will keep voltage constant. but noise can be introduced in a few inches. if a ripple is seen across an o-scope or voltage tester then finding out why supply is slow to regulate it. long wires?, wires too small for current? capacitors?
so if long power cables it is probably best to have another regulator. this however is built into the logic section that runs off of 5 volts.
i'm not 100% sure the 12v and above requires that much control. how does this effect the current regulated drivers for motors? (current is set as voltage changes depending on motor resistance and speed)
or the temp regulation of the heater or the bed? these device have different methods of regulation, and the end result should be about the same.



i would be more concerned if the ripple was at 5v regulation.

i will note that Arduino recommends max voltage for board regulator input is 12v (it can go up to 20v, but recommend no more than 12v). so if the design uses this regulator there could be overheating and shut down of electronics if voltage is constantly pushed beyond this without a cooling fan.

Edited 2 time(s). Last edit at 09/19/2018 12:07PM by jamesdanielv.

This is true if you don't have ripple on the 12 or 24 volt supply and you do on the 5V DC-DC simple switching circuit this opens a whole new set of problems.
Troubleshooting the Step Down 5V DC-DC simple switching circuit or Buck converter is sometimes quit difficult to troubleshoot.

As in this LM2596 simple switching circuit 12-30vdc input 5Vdc output



And here is the biggest have seen lots of 3d Printer controller schematics and for some reason DC-DC simple switching circuit or Buck converter is never included on the schematics.
This circuit is one I found as a stand alone, the one on my 3D Printer controller has multiple inductors and capacitors and without a schematic can be difficult to troubleshoot.
Unless you manually trace out all the componets and reverse engineering the circuit.

---

Computer Programmer / Electronics Technician

here is the schematic of the Arduino board
[www.arduino.cc]
and of the Arduino mega 2560
[www.arduino.cc]

both use the NCP1117ST50T3G regulator
here is a doc that references this regulator
[www.onsemi.com]

Again i would only be concerned if 5v has a detectable ripple, it can cause temp sensor variance if it were greater than 2% or 0.1v, but the effects would be minimal, digital logic would be unaffected unless swings were greater than 2.6v (although i would wonder why this would be happening)

does the designs you reference use the internal regulator or a separate 5v connection that is self regulated?

if using the internal regulation of the board, then is the ripple at acceptable levels at higher voltages? if so then i dont think using more voltage is a big concern as long as electronics are still in spec, and heatsinks and fans are placed on the board to remove extra heat.

the extra noise you are talking about might be from the additional current draw from devices such as nozzle heater and heated bed, these have internal resistance that are set for a certain voltage. raising the voltage increases current draw, and if the power supply, or the wires to the circuit are not capable of handling the current, voltage will drop and since current can not flow properly in wires, there is nothing regulation of voltage can do at this point. this would most likely happen on the 12v rails, and not the internal 5v regulation of the board.


but yes i agree with the ending of the first statement, make sure your supply can handle the current if it is run at a higher voltage. most supplies can.

after looking into more detail the LM2596 regulates to ±4% voltage. im not sure if this is over time or range, or its average tolerance. still it is high enough that i would not use it for board voltage, unless it is >6.4v and internally regulated by board (6.4 gives wiggle room, and assumes NCP1117 1.2v drop). several things to verify.

Edited 7 time(s). Last edit at 09/19/2018 02:59PM by jamesdanielv.

Quote
Roberts_Clif
And here is the biggest have seen lots of 3d Printer controller schematics and for some reason DC-DC simple switching circuit or Buck converter is never included on the schematics.

You've never looked at a Duet schematic then. Duets have included buck regulators on-board ever since the first model was released in 2013.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
jamesdanielv
the LM2596 regulates to ±4% voltage. i am not sure if this is over time or range, or its average tolerance. still it is high enough that i would not use it for board voltage, unless it is >6.4v and internally regulated by board (6.4 gives wiggle room, and assumes NCP1117 1.2v drop). several things to verify.

The LM2596 is on my controller, they warrantied it for me and stated I did not have to return the defective board.
Now I know you don't know me, but it there is something that is broken I have to fix it or throw it away.

This controller will work sometimes for 10 minutes others 4 hours before it goes off into the Twilight Zone, and this is on the same sliced ".gcode".
Replacement controller card works perfect without error - Well there is still human error but I DO Not Count as a printer error.

---

Computer Programmer / Electronics Technician