Ferrite filter question

Reecently I helped a neighbor connect a Chinese made CNC plasma table
to an American made plasma cutter. After I figured out that the
"partial pressure" the Chinese manual mentioned was in reference to a
built in jumper selectable voltage divider I was able to get the two
machines talking to each other and everything works now. But the
Chinese manual shows a filter on a sense wire and I don't know
exactly what to do about it. The sense wire is connected to the copper
shield on the plasma torch and to a terminal in the CNC control. The
machine uses the shield to sense the work. The machine moves the
plasma torch down until the shield contacts the work and then moves
away a set amount and energizes the plasma torch. I think, but am not
sure, that the control can now sense the distance between the work and
the torch tip, so that if the work is warped and rises or falls the
torch will follow the contour of the work. In any case, the connection
diagram for this sense shows it being wrapped a few times through a
ring. And somehow the manual alludes to this ring as being some sort
of filter. Since the neighbor didn't have any type of ferrite rings I
just wrapped a few turns of the sense wire through a few insulated
washers just to see if I could get the machine to work. And it did.
Well, I dug a ferrite ring out of a bad switching power supply and
removed the coiled wire from it. I'm hoping I can just duplicate the
washers trick with the ferrite ring but I'm thinking that I need
advice on how many turns of wire through the ring I should use. I
know, there are many different types of ferrite and I have no idea
what kind mine is. And there must be a bunch of other variables as
well. But if I can at least not ruin anything it would be great.
Thanks,
Eric

 

[George Herold]

On Monday, January 29, 2018 at 1:15:10 PM UTC-5, et...@whidbey.com wrote:

Reecently I helped a neighbor connect a Chinese made CNC plasma table
to an American made plasma cutter. After I figured out that the
"partial pressure" the Chinese manual mentioned was in reference to a
built in jumper selectable voltage divider I was able to get the two
machines talking to each other and everything works now. But the
Chinese manual shows a filter on a sense wire and I don't know
exactly what to do about it. The sense wire is connected to the copper
shield on the plasma torch and to a terminal in the CNC control. The
machine uses the shield to sense the work. The machine moves the
plasma torch down until the shield contacts the work and then moves
away a set amount and energizes the plasma torch. I think, but am not
sure, that the control can now sense the distance between the work and
the torch tip, so that if the work is warped and rises or falls the
torch will follow the contour of the work. In any case, the connection
diagram for this sense shows it being wrapped a few times through a
ring. And somehow the manual alludes to this ring as being some sort
of filter. Since the neighbor didn't have any type of ferrite rings I
just wrapped a few turns of the sense wire through a few insulated
washers just to see if I could get the machine to work. And it did.
Well, I dug a ferrite ring out of a bad switching power supply and
removed the coiled wire from it. I'm hoping I can just duplicate the
washers trick with the ferrite ring but I'm thinking that I need
advice on how many turns of wire through the ring I should use. I
know, there are many different types of ferrite and I have no idea
what kind mine is. And there must be a bunch of other variables as
well. But if I can at least not ruin anything it would be great.
Thanks,
Eric

Cute trick using some steel washers. If the signal line is coax, or
twisted pair, or something else with a shield, then you are making a
common mode choke. If a single wire then it's just a choke. Regardless of
which it's kinda hard to say how many turns you'll need. The impedance
of the choke is working against some other impedance in the circuit. I
typically do what you did.. wrap a few turns and see if it works. If
you've got a 'scope, you might be able to look at the signal line and see
the noise and reduction... but all sorts of caveat's here about floating
AC loads and HV. You don't want to blow out your 'scope.

George H.

 

[whit3rd]

On Monday, January 29, 2018 at 10:15:10 AM UTC-8, et...@whidbey.com wrote:

Reecently I helped a neighbor connect a Chinese made CNC plasma table
...manual shows a filter on a sense wire and I don't know
exactly what to do about it.... the connection
diagram for this sense shows it being wrapped a few times through a
ring.

Yep, that's a common-mode filter, usually a lossy ferrite bead,
sometimes a split bead and plastic housing that clamps over the
cable. Washers might work, but (one hopes) a ferrite would
be slightly better because steel can magnetize and cores for
this purpose ought not to do that.

Since the purpose of such a filter is to deal with unwanted interference, there's
no real calculation to selecting the core and number of turns.
What works, is starting with the smallest core and least turns, doing trials, and when
the bad symptoms die down, use the next larger size or one extra turn.

If you have an emissions test lab available (I did), watch the emissions measured,
and omit the 'next larger size' step.

 

[amdx]

On 1/29/2018 12:21 PM, etpm@whidbey.com wrote:

Reecently I helped a neighbor connect a Chinese made CNC plasma table
to an American made plasma cutter. After I figured out that the
"partial pressure" the Chinese manual mentioned was in reference to a
built in jumper selectable voltage divider I was able to get the two
machines talking to each other and everything works now. But the
Chinese manual shows a filter on a sense wire and I don't know
exactly what to do about it. The sense wire is connected to the copper
shield on the plasma torch and to a terminal in the CNC control. The
machine uses the shield to sense the work. The machine moves the
plasma torch down until the shield contacts the work and then moves
away a set amount and energizes the plasma torch. I think, but am not
sure, that the control can now sense the distance between the work and
the torch tip, so that if the work is warped and rises or falls the
torch will follow the contour of the work. In any case, the connection
diagram for this sense shows it being wrapped a few times through a
ring. And somehow the manual alludes to this ring as being some sort
of filter. Since the neighbor didn't have any type of ferrite rings I
just wrapped a few turns of the sense wire through a few insulated
washers just to see if I could get the machine to work. And it did.
Well, I dug a ferrite ring out of a bad switching power supply and
removed the coiled wire from it. I'm hoping I can just duplicate the
washers trick with the ferrite ring but I'm thinking that I need
advice on how many turns of wire through the ring I should use. I
know, there are many different types of ferrite and I have no idea
what kind mine is. And there must be a bunch of other variables as
well. But if I can at least not ruin anything it would be great.
Thanks,
Eric

You might put your blurb on rec.crafts.metalworking
Ignore the bickering that is presently happening, new name popped
in and giving one guy a hard time.
There are some very bright guys there.
Mikek

 

[Helmut Wabnig]

On Mon, 29 Jan 2018 10:21:01 -0800, etpm@whidbey.com wrote:

Reecently I helped a neighbor connect a Chinese made CNC plasma table
to an American made plasma cutter. After I figured out that the
"partial pressure" the Chinese manual mentioned was in reference to a
built in jumper selectable voltage divider I was able to get the two
machines talking to each other and everything works now. But the
Chinese manual shows a filter on a sense wire and I don't know
exactly what to do about it. The sense wire is connected to the copper
shield on the plasma torch and to a terminal in the CNC control. The
machine uses the shield to sense the work. The machine moves the
plasma torch down until the shield contacts the work and then moves
away a set amount and energizes the plasma torch. I think, but am not
sure, that the control can now sense the distance between the work and
the torch tip, so that if the work is warped and rises or falls the
torch will follow the contour of the work. In any case, the connection
diagram for this sense shows it being wrapped a few times through a
ring. And somehow the manual alludes to this ring as being some sort
of filter. Since the neighbor didn't have any type of ferrite rings I
just wrapped a few turns of the sense wire through a few insulated
washers just to see if I could get the machine to work. And it did.
Well, I dug a ferrite ring out of a bad switching power supply and
removed the coiled wire from it. I'm hoping I can just duplicate the
washers trick with the ferrite ring but I'm thinking that I need
advice on how many turns of wire through the ring I should use. I
know, there are many different types of ferrite and I have no idea
what kind mine is. And there must be a bunch of other variables as
well. But if I can at least not ruin anything it would be great.
Thanks,
Eric

You did not mention any problems WITHOUT the choke.
So it might work anyway and your ring does nothing to it.


A choke is usually wound with five to ten turns
around a ferrite.
Its performance can be measured, but that was not your question.

w.