Earthing equipment questions

[adrianh]

@Andy

I had a look at the machine just now. The machine does not have any earth whatsoever. When I turn the machine on an measure the voltage between the chassis and te installation earth it sits at 110VAC. I see that even on the motion controller housing.

It seems that the whole lot has been left to float for some reason.

Could it be a function of the massive transformer?
There is a green lead that comes from within the transformer core and is wired onto the body of the transformer - what is that all about?

[adrianh]

I just had a thought, maybe one of the output phases of the VFD is shorting to the chassis somewhere along its cable run.

[adrianh]

I checked the VFD and all its connections are OK.

I got it sorted.

I disconnected the VFD, transformer, coolant pump but the problem remained the same. Then I had a thought and it was to isolate the metal body of the motion controller from the rest of the machine....and suddenly everything seemed ok.

So, I put a proper earth in to the machine chassis. I then reconnected each component in turn. That all worked fine. I then installed padding between the motioon controller and the machine chassis and bolted it down with plastic bolts.

The machine is working happily and I no longer fear for my life.

I can't say that I understand what is going on with the motion controller but it may be a function of the USB cable connecting to the PC. I should have checked to see what it would do if I just disconnected the USB cable but I didn't think of it at the time.

I connected my TOPTRONIC ELCB / Polarity testers and everything seems good.

[AndyD]

First the earthing. The transformer looks like it's a standard step down, it doesn't appear to be an isolating transformer with a screen so the earth wire connecting to the laminations should go to the supply earth (CPC), it's just for safety if the transformer ever goes faulty.

From what I can see I'd say the chassis of the machine should definitely be connected to the CPC (the supply earth).

Secondly the elevated voltage that was giving the static/tingling sensation. The 110v you saw on your tester is possibly misleading, probably what I'd call a ghost voltage if you were using a digital tester. You can take it as an indication of earth leakage but nowhere near as bad as the voltage may suggest. If you did the same voltage test with an analogue tester or a digital tester with a LoZ range you'd see a much smaller voltage being indicated. The voltage you saw could be caused by functional leakage, especially so looking at the internal components. The VFD, the constant current driver/PSU and the 4-axis USB motion controller are all candidates for having standing functional leakage. As I've explained elsewhere, functional earth leakage currents are not considered fault currents as long as they meet certain criteria. If the machine is now supplied via an RCD and the chassis is bonded to the supply earth then there's no reason it shouldn't be safe.

[adrianh]

I agree with you about the ghost voltage. I find it interesting that the machine was never earthed at all, seems that the person who said that one should be careful of Chinese machine earthing was right.

[Justloadit]

Switch mode power supplies, have a tendency to create an induced voltage in metal surrounding the PSU. The high switching frequencies cause by the capacitive effects in the internal transformers are usually the cause of this. You will note this tingling feeling when you connect a laptop which is being charged to your desktop PC. The reason for this is that the laptop PSU does not have an earth, because it is supplied with a 2 wire chord. The secondary side negative usually becomes the earth for the laptop, and the electronics negative is usually 100odd Volts to earth. This is what you may be experiencing. This was more prevalent in the days that we used RS232 to transfer data between P.C.s, one could see the spark when the connectors were inserted, and the casings touched.

This is known as "touch current". Do a google search to get some articles on it.

Theoretically there should be no problem connecting earth to the chassis, however not knowing your machine, I would not say that it will not cause any failures.
If you are connecting the PC to the unit, then the PC usually creates the earth path via the comms cable. Once the PC is connected check to see if there is still this elevated voltage.

[AndyD]

I thought it might be because they left it optional for a second CPC to be run and connected to an exposed termination point but you say that wasn't the case with this machine so I'm completely clueless why it would have been left floating.

The ghost voltage thing is a common issue and it confuses experienced sparkies sometimes. If you ever come across an AC voltage that's an arbitrary value just retest using an old analogue tester with a needle if you have one, if not keep look out for one that's a recognized brand at the car boot sales/gumtree, you can pick them up for cheap and they have their uses.

Now you're on a roll you should check your other machines for safe earthing just for peace of mind. It's the kind of problem you don't notice until something goes wrong one day and you get a proper shock.

[adrianh]

Here is a scenario that I had that that I also couldn't resolve. I was working on a model train DCC controller. The controller was connected to the PC via a usb interface. I connected the negative scope probe clip to the one output rail of the USB interface and when I touched the other rail of the interface the contoller went up in smoke. Everything is really good quality and nothing has ever given any problems on their own. It was just this particular combination that fried the controller. The scope is a 60MHz DSO

[bones]

@bones - Thank you for your truly interesting take on the workings of a modern laser cutter. May I be so bold as to ask you where you obtained your information? I really need to brush up on my poor understanding of such matters.

got mine second hand that is how
they explained it to me and what
i can tell it works so i guess they
where right

[adrianh]

got mine second hand that is how
they explained it to me and what
i can tell it works so i guess they
where right

They were wrong

[Justloadit]

The ghost voltage thing is a common issue and it confuses experienced sparkies sometimes. If you ever come across an AC voltage that's an arbitrary value just retest using an old analogue tester with a needle if you have one, if not keep look out for one that's a recognized brand at the car boot sales/gumtree, you can pick them up for cheap and they have their uses.

In the case you do not have an analogue meter, use a 220V pilot light connected to your meter probes when you do the measurement. If there is a high voltage, then the light should glow, and your meter will show the voltage

[adrianh]

In the case you do not have an analogue meter, use a 220V pilot light connected to your meter probes when you do the measurement. If there is a high voltage, then the light should glow, and your meter will show the voltage

I'll remember this.