Combining grounds logic and HV

[simpson36]

I'm sure this has been asked before and I have done searches and read what I could find, but still I am confused a bit by the options.

I my setup, 5VDC and 12VDC are supplied by a standard PC power supply which is grounded to the control box chassis.

HV is a toridal linear unregulated PS with 72VDC output which is not grounded to the chassis.

Voltmeter says no continuity and no voltage between 72V + or 72V- and chassis.

I now need to supply a drive with 12V and HV that have a common ground.

What are the options and what is the best way to do this?

[Al_The_Man]

First make sure the AC secondary side of the toroid is not at ground, (probably not), but it pays to check if grounding the DC-ve side.
I use this method all the time with Advanced Motion drives, and in fact the AMC site has a Technical Section where they describe this.
The drive signal common and the DC power common should have a bonding connection to a common Earth ground plate, this is a point set up for all the connections that are set up to Earth potential, including the service ground.
The PC power supply common should also be bonded to this point.
Al.

[simpson36]

The AC secondary side of the toroid would be the wires coming from the toroid prior to the bridge? I will check these and see it either has continuity with earth ground.

Assumming these are not grounded, then it is OK to simply earth ground the negative side of the HVDC output?

I read an old post of yours on this subject and you mentioned using resistors to test to see it the voltage 'collapsed'. That is unfortunately a foreign language to me. Is this something I should do before I make a hard connection fro HVDC- to earth ground?

Sorry to be so paranoid, but I learned about 'isolation' the hard way and have been probably overly cautious ever since. This new drive is relatively expensive and I don't want to fry it right off the bat.

[Al_The_Man]

Originally Posted by simpson36 The AC secondary side of the toroid would be the wires coming from the toroid prior to the bridge?.

Yes

Originally Posted by simpson36 Assumming these are not grounded, then it is OK to simply earth ground the negative side of the HVDC output? .

Yes You should be OK, generally with a simple linear supply you can visually see if there is any present connection.
If you have already measured the voltage and continuity from the DC to chassis or Earth ground and had no reading, then it is safe to assume you can make the connection from DC common to Ground.
The resistor test was where a voltage appears when measured from DC to ground, this can often be leakage on a sensitive meter, if a 1k resistor is connected from the DC point to ground and the voltage is sustained then it may not be possible to connect common to ground without further inspection and correction.
Another example is if you happed to get a reading of say 70v to chassis ground and you placed a 120v lamp across this connection, if it was just leakage the lamp would not light and the voltage would collapse.
In this case it should be OK to bond the common to earth ground.
Al.

[simpson36]

You are . . the man!

Thanks for the explanation, I was not familiar with the term 'voltage collapse', but I completely get it now.

I would liken it to a charge held by a cap after the PS is switched off. It is there and reads on a meter, but with any path to ground, the voltage 'collapses' to nothing.

[Al_The_Man]

Years ago when the only VOM had a moving coil meter, this did not show up as the relatively low impedance meter loaded the circuit, the draw back was, if you were measuring a particularly high impedance circuit, the voltage reading could be misleading due to the meter unduly loading the circuit.
The digital VOM cured most of these loading problems, but brought others, where sometimes a strange reading has to be checked into a bit further.
Reading a (leakage) voltage on the output of an SSR for example that should be OFF confuses some.
Al.