In principle yes but the various units you mention may well not be happy with unregulated/smoother DC. I'd suggest that the 24V needs to be at least regulated. What actual power rails do the drives etc need? Stepper drivers often need a higher voltage for the output stages, hence the 60V I assume, and typically a 5V logic level rail.

There's a standard IC, the LM7824, which outputs 24V from a minimum input of 33V. Your 24V rectified and smoothed should just about reach this (24 x root 2 = 34V). I'm sure there are other ICs that will work with a lower input – in fact this one:

**LINK**

which goes down to 27V and can supply 2A max. 73 pence each from that supplier. You could provide one for each subsystem. They will need a bit of strip board for mounting, a heatsink and probably some decoupling – see the application notes.

Edited By John Haine on 19/05/2021 07:13:14

The risk of sharing is that, when a device is called on to pulse a powerful stepper motor, it might put enough noise on common wiring to discombobulate anything else in the system relying on clean power. Best practice is to separate power supplies, and to filter, shield and route wiring to minimise accidental coupling.

However there's a good chance a simple solution might work. Electronic designers are well aware of noise problems and often defend their circuits. Trouble is, the level of protection varies between nothing at all and prohibitively expensive. Looking at Printed Circuit Boards, it's common to find the designer provided for filtering but someone reduced cost by not fitting the components!

Good news. In my experience stepper motor drivers are well protected, but beware naive web forum reassurances based on small numbers of amateur results! My microcontroller operated rotary table uses the same power supply for everything apart from the Arduino being semi-isolated behind a second regulator. It works fine but although I got away with it, the result wasn't guaranteed. It was a risk. Nonetheless, given positive results on all my stepper bodges projects, I'd be inclined to try a single power supply, test thoroughly, and install the second power supply only if there's trouble.

'Test thoroughly' may be a problem: in the event of misbehaviour I have a certain amount of experience and a capable oscilloscope. Glitch errors are hard to debug without the right tools and 'common sense' is no substitute for well-equipped expertise. In which case it may be easier to throw money at it and apply 'best practice' from the get go.

Dave

Rod,

Connect the mains ground to the machine metal body – a good connection – thats your lifeguard.

Connect the VFD ground to the same mains ground connection with a copper wire connection – preferably not via the machine body

Then in your installation I would –

Connect the +5V supply ground to a 'common ground point' ( this can be the chassis, but be sure the connection is good and cannot work loose – I prefer it to be a tag board or junction block, etc)

Connect the 24VDC negative lead to that same common ground point.( this means your 24VDC supply is no longer isolated, but it is not needed in your application)

Then take a single 20 or 18guage wire from the 'common ground point', to the mains ground connection.

So now Mains ground, +5V ground and 24VDC ground are common.

I would leave the 60VDC supply isolated – ie, don't connect its -Ve lead to ground anywhere.

If you do want to connect it to ground ( it can reduce EMI, sometimes…) then connect it from as close as possible to the smoothing cap -Ve lead to the Common Ground Point.

Ground loops are avoided by the common ground point.

DO ground the switching power supplies -Ve leads to the Common point, and make a good between the chassis of the power supply to the machine body and/or the mains ground.

The MESA 7i76 card's 24VDC input -Ve lead should come from that common ground point.

Joe