Re: Lost steps
Posted by
Mariss Freimanis
on 2000-08-22 17:17:03 UTC
What may be overlooked here is the intermittent nature of the problem
originally reported.
If you don't want to have a system plagued with noise problems, do
the ground wiring right from the beginning. Right means use the
concept of a "holy point ground". This means ALL ground wires are
returned to a single point; power supply ground, motor drive ground,
parallel port ground, in short if it has a ground wire, take it to
that point! If shields are used, take them there too.
Resist the temptation to "daisy chain" ground wires. It can be
tempting to run a 3" wire from one device to an adjacent one where
otherwise it would be a 3 foot wire run back to the "holy point" You
won't be doing yourself a favor by taking the shortcut.
The whole point is to keep a level playing field. Voltages are always
relative. That is why it takes 2 leads to measure them. For most of
us that means "relative to ground". But what happens if one device's
ground has +/- 2 volts of high-frequency noise on it relative to
another device's ground? It will be added with the signal voltage and
no bypass cap will help there.
A single point ground "levels the field" so every device has the same
point of reference. It really saves a lot of grief.
Mariss
originally reported.
If you don't want to have a system plagued with noise problems, do
the ground wiring right from the beginning. Right means use the
concept of a "holy point ground". This means ALL ground wires are
returned to a single point; power supply ground, motor drive ground,
parallel port ground, in short if it has a ground wire, take it to
that point! If shields are used, take them there too.
Resist the temptation to "daisy chain" ground wires. It can be
tempting to run a 3" wire from one device to an adjacent one where
otherwise it would be a 3 foot wire run back to the "holy point" You
won't be doing yourself a favor by taking the shortcut.
The whole point is to keep a level playing field. Voltages are always
relative. That is why it takes 2 leads to measure them. For most of
us that means "relative to ground". But what happens if one device's
ground has +/- 2 volts of high-frequency noise on it relative to
another device's ground? It will be added with the signal voltage and
no bypass cap will help there.
A single point ground "levels the field" so every device has the same
point of reference. It really saves a lot of grief.
Mariss
> OK, what Peter is talking about here is putting small valuecapacitors
> across the signal inputs (one from signal ground to step in, and onewiring.
> from signal ground to direction in) right at the stepper driver
> terminals.
> I would recommend 100 pF (Pico Farads) as a start. This will remove
> extremely short duration noise pulses that are picked up by the
>
> Jon
Discussion Thread
Tim Goldstein
2000-08-14 10:55:55 UTC
RE: [CAD_CAM_EDM_DRO] Re: Lost steps
Joe Vicars
2000-08-15 14:33:28 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
Ozzie@h...
2000-08-20 13:41:24 UTC
Re: Lost steps
Mariss Freimanis
2000-08-20 15:57:28 UTC
Re: Lost steps
JanRwl@A...
2000-08-20 17:10:24 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
Ozzie@h...
2000-08-20 20:04:20 UTC
Re: Lost steps
Ozzie@h...
2000-08-20 20:23:39 UTC
Re: Lost steps
Mariss Freimanis
2000-08-20 23:57:26 UTC
Re: Lost steps
Jon Elson
2000-08-21 16:08:15 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
ptengin@a...
2000-08-21 17:18:36 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
Ozzie@h...
2000-08-21 20:02:49 UTC
Re: Lost steps
dave engvall
2000-08-21 20:24:57 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
JanRwl@A...
2000-08-21 21:04:54 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
Ozzie@h...
2000-08-21 21:47:02 UTC
Re: Lost steps
Ozzie@h...
2000-08-21 22:00:05 UTC
Re: Lost steps
Jon Elson
2000-08-21 23:22:44 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
ptengin@a...
2000-08-21 23:47:53 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
Jon Elson
2000-08-22 13:31:34 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
ptengin@a...
2000-08-22 15:44:18 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
Mariss Freimanis
2000-08-22 17:17:03 UTC
Re: Lost steps
wanliker@a...
2000-08-22 17:42:41 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
Mariss Freimanis
2000-08-22 18:37:23 UTC
Re: Lost steps
JanRwl@A...
2000-08-22 21:36:53 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
Jon Elson
2000-08-22 22:16:56 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
Ian Wright
2000-08-23 04:19:53 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
Ozzie@h...
2000-08-23 11:49:05 UTC
Re: Lost steps
Tim Goldstein
2000-08-23 13:19:04 UTC
RE: [CAD_CAM_EDM_DRO] Re: Lost steps
hans
2000-08-23 13:49:25 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
JanRwl@A...
2000-08-23 15:04:55 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps
Ian Wright
2000-08-24 01:08:01 UTC
Re: [CAD_CAM_EDM_DRO] Re: Lost steps