2 quick questions

Hello everyone. I posted this in general, but I figured it belonged here instead. This is probably a really stupid question, but I'm working on my reprap, and I can't figure this out. I want to make sure, and the instructions don't say anything. The part that goes into the wall gets hooked up to L and N for the power supply correct? And which color is live and which is neutral? I have a blue wire and a purple wire. Thank you for any help.

Please don't post the same question in multiple sections.

Modern wiring should have three mains wires, one that runs to L(ive), one that runs to N(eutral) and one that runs to G(round) or a symbol including vertical line and three decreasing-size horizontal lines in it, with or without a circle around them (that one should also be run to the adjacent negative post).

As long as the wires are suitable for the application and you get them run from each terminal to the corresponding prong, the color is not all that critical, but those wires would concern me, as I find no single-phase AC color-coding standard that has a purple wire in it.

Are you saying connect supply earth to DC Negative there, IMBoring?

What's the thinking on that, may I ask?

I'm not disputing it, just asking.

-a

The glib answer would be that's what RepRapPro's instructions said to do, and it made sense to me since it would keep you from being the only path to earth in the event of a short to the case (which at 12V shouldn't be dangerous in most cases, but could certainly be unpleasant).

Upon further investigation, it appears there are a number of industry standards that dictate that approach, and some history of electronic gremlins and difficulty diagnosing shorts when it is not followed.

It wouldn't affect safety of mains shorts. Connecting any metalwork to earth makes sense for that.

The danger of 12v shock is negligible. To connect the psu -ve to earth would increase your chance of contact with it.
Heaters are permanent +ve with the negative switched (I'm sure. Could be wrong), so the chance of dangerous (fire starting) short circuit is greater with any metalwork like psu enclosure at -ve potential.

I've hear of comms faults when connecting puters to equipment with different ground potential. Earth loop faults or something.
My laptop is only on a 2 core, so that's a floating ground.

I don't know what you mean by; difficulty diagnosing shorts (without psu negative being connected to mains earth, we're talking.)

I'm not an expert on this. I'm happy to be proven wrong in the quest for truth but I'm not convinced it is a good idea,

-a

I don't see why this would have any impact on the likelihood of coming in contact with +12V. If the case shorts to +12V, you will be coming into contact with either a floating or earthed +12V, and you will form a part of a circuit to earth. The only question is whether there's a redundant current path to earth to reduce the current through the part of the circuit that includes you. Both my OEM PSUs are wired this way, and I've never gotten so much as a static charge off them.

A short in the machine wiring shouldn't make any difference either way. Whether it's floating or earthed, the PSU will keep converting mains and merrily pumping 12V to the circuit to the machine board unless a fuse is popped somewhere along the way.

Comms fault is one of the things I've seen discussed, but I'd fall on the other side of that discussion as well, since my laptop does have a 3-prong plug.

The person in question didn't go into a great deal of detail and I think more complex circuits were involved, but I could see extra circuits being energized through inconsistent floating grounds, which could reasonably be expected to complicate troubleshooting.

I'm no guru on the electronics side of things either. My background is aerospace/mechanical.

We build power electronic equipment at work. I have always connected SMPS outputs to ground as a matter of course. I have always understood that, unless there is a good reason, electronics should be referenced to ground. Without this, they become susceptible to noise.
A few years ago, we had some equipment we were testing, and over a few weeks, we would see unexplained failures in the electronics (about three times, an op-amp would fail). I found the ground connection had been left off the negative output of the SMPS supplying those electronics. Once the connection was installed, we had no more failures.

Ok-dokey.

I was trying for a definitive answer as there seems a lot of confusion as to what an 'Earth' does. People seem to think its some magic current path which 'soaks electricity and noise into the ground'.

Sorry I confused my argument by talking about switched negatives and permanent positive feeds but I can see no 'safety' benefit in connecting DC negative to ground.

Fair point about 'reference to ground' but my printers reference to ground is its own negative supply rail.

Like my laptop has it's own internal ground reference from the its DC negative. If I plug that into my printer, those 2 grounds connect and have the same 'reference' potential. I don't understand the benefit of an electrical connection between that and domestic supply earth.

I've never seen a laptop plug that had 3 prongs on the DC side.

I'm still not convinced.

-a

Exactly! The three wire cord on a laptop protects the high voltage stuff - typically the external power supply blob, and has nothing to do with the DC side of the supply. Switcher or linear supply, they are all isolated from the AC line, and all tying DC negative to line ground would accomplish would be to transfer more noise.

AC/powerline ground is for fault protection, period . . .

- Tim

Edited 1 time(s). Last edit at 05/03/2015 04:29PM by tadawson.

The laptop PSUs I have checked all connect the negative side of the output to mains ground. If they didn't, they would not need a 3 core mains lead, because they are invariably enclosed in plastic cases. I suspect the reason is that it is too expensive to make a switching power supply that provides a reasonable amount of power and also meets the required double insulation and leakage standards to not require the output to be grounded.

---

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].

But note that the UL rating is for the supply only. This is why so many low end products use wall warts - you don't need to certify the core product if the 120v power supply is external, you just buy an approved supply.

- Tim

I just checked three laptop power supplies I had on hand. Like dc42's experience, they all had continuity between the negative DC pin and the AC earth. The modern ones had no measurable resistance. The 1999 model checked as a megohm.

Well that's all very interesting but it still doesn't answer the question as to why anyone should connect their reprap power DC negative to ground.

I'm not saying you shouldn't. I'm still only asking why.


Actually. Back to safety. There is a significant issue with some printers. . .

My printer has a metal frame with the power supply inside it and 230v going into it.
Quite correctly, the metal parts are earthed to prevent mains voltage shock risk.

As I mentioned - reprap controllers switch the heaters with a negative feed (permanent positive on the heatbed and hotend.)

Which means; If that negative feed wire gets chaffed or pinched or somehow contacts the metal work, then the heater is switched on. (if DCneg is connected to ground)
Permanently and without control and you wouldn't even know about it (apart from the heat, obviously)

That's not safe.

On any kind of electrical machine or appliance, if a wire touches the chassis it should either do nothing or blow a fuse. It should not energize a component without you knowing it.
Especially not a heater (have you seen thejollygrimreapers video of a runaway hotend, dripping aluminium?!)

So for that reason alone I won't be connecting my -ve side to earth. I'll just have to live with the . . . the . . . what was it again?

If your printer has no earthed chassis or the PSU is external, that doesn't apply.


-a

Quote
alan richard
Well that's all very interesting but it still doesn't answer the question as to why anyone should connect their reprap power DC negative to ground.

I'm not saying you shouldn't. I'm still only asking why.
-a

It's a signal integrity issue with comms paths. Let's see if I can explain...

Let's imagine we have two bits of electronics powered from separate 12VDC power sources and that the power sources do not connect the DC negative to ground. Each is therefore free to float its rails some way off true ground provided its the DC negative and DC positive both float together and stay 12V apart. Each bit of electronics is happy in itself as it only sees the difference between its two rails, 12VDC. Relative to true ground however one might be (say) 0V & 12V and the other 20V & 32V (a bit extreme maybe but it's an example)

Now lets connect a comms cable between the two, it could be RS232 or LAN or anything else. The comms cable -ve will form a connection between the two DC negatives so we'll have 0V connected to 20V and current will flow. That current will be limited by the capability of the power supplies but it could be quite large as one power supply pulls the other's negative to match it's own. In practical situations the differences in potential are not normally enough to give a safety concern but we now have a comms cable which is unbalanced, it has more current flowing in its negative wire than in its signal wire, and this can lead to all sorts of very hard to track down comms failures.

Quote
alan richard
Actually. Back to safety. There is a significant issue with some printers. . .

My printer has a metal frame with the power supply inside it and 230v going into it.
Quite correctly, the metal parts are earthed to prevent mains voltage shock risk.

As I mentioned - reprap controllers switch the heaters with a negative feed (permanent positive on the heatbed and hotend.)

Which means; If that negative feed wire gets chaffed or pinched or somehow contacts the metal work, then the heater is switched on. (if DCneg is connected to ground)
Permanently and without control and you wouldn't even know about it (apart from the heat, obviously)

That's not safe.

On any kind of electrical machine or appliance, if a wire touches the chassis it should either do nothing or blow a fuse. It should not energize a component without you knowing it.
Especially not a heater (have you seen thejollygrimreapers video of a runaway hotend, dripping aluminium?!)

So for that reason alone I won't be connecting my -ve side to earth. I'll just have to live with the . . . the . . . what was it again?

If your printer has no earthed chassis or the PSU is external, that doesn't apply.

-a

You are right to worry about this, I had it happen to me when a single strand of the outer braid of the hot-end heater cable shorted to the control pin.

I guess it comes down to which problem you consider most in important/troublesome.

Thanks for a very clear answer. Good for reference.


-a

Quote
csambrook
Let's imagine we have two bits of electronics powered from separate 12VDC power sources and that the power sources do not connect the DC negative to ground. Each is therefore free to float its rails some way off true ground provided its the DC negative and DC positive both float together and stay 12V apart. Each bit of electronics is happy in itself as it only sees the difference between its two rails, 12VDC. Relative to true ground however one might be (say) 0V & 12V and the other 20V & 32V (a bit extreme maybe but it's an example)

Now lets connect a comms cable between the two, it could be RS232 or LAN or anything else. The comms cable -ve will form a connection between the two DC negatives so we'll have 0V connected to 20V and current will flow. That current will be limited by the capability of the power supplies but it could be quite large as one power supply pulls the other's negative to match it's own. In practical situations the differences in potential are not normally enough to give a safety concern but we now have a comms cable which is unbalanced, it has more current flowing in its negative wire than in its signal wire, and this can lead to all sorts of very hard to track down comms failures.

I'm sorry, that explanation is misleading.

If both pieces of electronics had 12VDC supplies that were truly floating, then there would be no problem with connecting a comms cable between them. Both 12V supplies would still be floating, but tied to float at the same level.

If one were floating and one were grounded, then again there would be no problem. The 12V supply that was previously floating would be tied to ground when the comms cable is connected.

If both were grounded, and the mains grounds they are connected to were truly at the same potential (i.e. ground), there would again be no problem.

Problems arise when one of the following is true:

1. Both supply outputs are grounded, but the grounds not always at the same potential. This can occur if then two devices are plugged into different power sockets, and other devices on the same mains circuit generate ground noise. In this case, the communications cable (e.g. USB cable) connecting the two devices will carry ground current due to the difference in ground potential. This is the common situation when a 3D printer is connected to a PC using a USB connection.

Possible solutions include:

(a) Use an alternative form of connection that doesn't connect the grounds, e.g. Ethernet. This is how I run my printers, except when debugging new firmware. An Ethernet connection does not join the two grounds.

(b) Power the printer and the PC from the same mains distribution block, not from different mains sockets. Don't power anything else that might create ground noise from the same distribution block.

(c) Put a ferrite bead around the USB cable. This will reduce any ground current caused by short-lived transient differences between the ground potentials. This is best used in conjunction with (b).

2. At most one supply output is grounded, and there is significant capacitive coupling in the PSU(s) providing the non-grounded output(s) between the mains input and the output. In this case, transients on the mains may get coupled through to the PSU outputs, so that the outputs are not truly floating at high frequencies. When a USB cable is connected between the devices, high frequency ground currents may flow.

Possible solutions include:

(a) Use an alternative form of connection that doesn't connect the grounds, e.g. Ethernet.

(b) Power at least the device with the non-grounded output via a surge suppressing mains distribution block. This assumes that the device does not itself generate mains transients.

(c) Put a ferrite bead around the USB cable. This will reduce any ground current caused by short-lived transient differences between the ground potentials.

---

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
dc42

Quote
csambrook
Let's imagine we have two bits of electronics powered from separate 12VDC power sources and that the power sources do not connect the DC negative to ground. Each is therefore free to float its rails some way off true ground provided its the DC negative and DC positive both float together and stay 12V apart. Each bit of electronics is happy in itself as it only sees the difference between its two rails, 12VDC. Relative to true ground however one might be (say) 0V & 12V and the other 20V & 32V (a bit extreme maybe but it's an example)

Now lets connect a comms cable between the two, it could be RS232 or LAN or anything else. The comms cable -ve will form a connection between the two DC negatives so we'll have 0V connected to 20V and current will flow. That current will be limited by the capability of the power supplies but it could be quite large as one power supply pulls the other's negative to match it's own. In practical situations the differences in potential are not normally enough to give a safety concern but we now have a comms cable which is unbalanced, it has more current flowing in its negative wire than in its signal wire, and this can lead to all sorts of very hard to track down comms failures.

I'm sorry, that explanation is misleading.

If both pieces of electronics had 12VDC supplies that were truly floating, then there would be no problem with connecting a comms cable between them. Both 12V supplies would still be floating, but tied to float at the same level.

If one were floating and one were grounded, then again there would be no problem. The 12V supply that was previously floating would be tied to ground when the comms cable is connected.

If both were grounded, and the mains grounds they are connected to were truly at the same potential (i.e. ground), there would again be no problem.

Problems arise when one of the following is true:

1. Both supply outputs are grounded, but the grounds not always at the same potential. This can occur if then two devices are plugged into different power sockets, and other devices on the same mains circuit generate ground noise. In this case, the communications cable (e.g. USB cable) connecting the two devices will carry ground current due to the difference in ground potential. This is the common situation when a 3D printer is connected to a PC using a USB connection.

Possible solutions include:

(a) Use an alternative form of connection that doesn't connect the grounds, e.g. Ethernet. This is how I run my printers, except when debugging new firmware. An Ethernet connection does not join the two grounds.

(b) Power the printer and the PC from the same mains distribution block, not from different mains sockets. Don't power anything else that might create ground noise from the same distribution block.

(c) Put a ferrite bead around the USB cable. This will reduce any ground current caused by short-lived transient differences between the ground potentials. This is best used in conjunction with (b).

2. At most one supply output is grounded, and there is significant capacitive coupling in the PSU(s) providing the non-grounded output(s) between the mains input and the output. In this case, transients on the mains may get coupled through to the PSU outputs, so that the outputs are not truly floating at high frequencies. When a USB cable is connected between the devices, high frequency ground currents may flow.

Possible solutions include:

(a) Use an alternative form of connection that doesn't connect the grounds, e.g. Ethernet.

(b) Power at least the device with the non-grounded output via a surge suppressing mains distribution block. This assumes that the device does not itself generate mains transients.

(c) Put a ferrite bead around the USB cable. This will reduce any ground current caused by short-lived transient differences between the ground potentials.

Thanks for the clarification dc42. You're right of course (on all counts!) but in practice I've met so many "floating" outputs which are not truly floating in the theoretical sense but rather drive the negative up to some small potential above ground that I now treat all floaters as a source of potential offsets. (As an aside the best was a metal cased laptop which I learned not to put on my lap when wearing shorts - it wasn't a problem normally but sometimes my leg would touch the metal of my chair which was "unpleasant". Thanks Sony).

alan_richard: I don't think that avoiding an earth connection on your printer should make you sleep any easier as the problem you describe with heater runaway only needs a connection back to the controller board's 0V rail not to external earth. On my RepRapPro Huxley they have taken great care to use beautifully screened and shielded cables to all of the heaters and motors and sensors, the result of which is that a short from the hot-end return wire to its cable screen has no trouble finding a path back to the controller 0V. With the inevitable exciting and smelly result.

Thanks for spoiling my sleep. Also for the thorough answers. I was more concerned with shorts to earth/ chassis.

Of course, with printer heaters being on the move all the time, the cable is especially vulnerable to mechanical damage and shorts if not properly managed. A warning for everyone there.

I would think, then, this must be an issue faced by a number of machinery manufacturers, as electronic power switching is normally done on the Negative, isn't it?

What is the usual industry answer to this? Double insulation/ segregation?



-a

Edited 2 time(s). Last edit at 05/07/2015 05:32AM by alan richard.