> Doesn't any unpaired layer (mechanical or otherwise) preclude the
> implementation of a true flip board function in the future? The main
> concerns about flip board have been what to do with layers especially
> that don't have a logical flipped layer, so doesn't adding more layers
> have no pairing add to the confusion. Perhaps this is just my limited use
> mechanical layers for assembly purposes ... what else are they used for
> wouldn't deserve pairing ?
> Douglas McDonald

I saw this posting at the time, intended to answer it, and am doing so now.

Objects which reside on unpaired layers would remain on the same layers
following an inversion (or partial inversion) of a PCB file, but like *all*
items undergoing inversion, they would effectively be mirrored compared with
previously (e.g. the mirrored state of all Strings toggle, and an arc whose
start and end angles were previously (say) 0 deg and 90 deg (respectively)
would change to 90 deg and 180 deg (respectively, and assuming a vertical
inverting axis)).

In some cases it would be preferable for some text items (on such layers) to
remain in an unmirrored state, and if those items had been amongst those
inverted, it would indeed then be necessary to follow the inverting
procedure with other procedures which restored such items (and perhaps other
objects on the same layers) into a desirable state. Exactly what following
procedures should be undertaken at the time would depend upon the
circumstances, but a typical scenario would involve deselecting everything,
selecting all items on *some* of the layers of an unpaired nature (but not,
for instance, selecting items on the KeepOut layer, or whichever Mechanical
layer has been assigned to define the PCB's border), mirroring all selected
items, moving those items to an appropriate location, then de-selecting
everything yet again.

Having to undergo such additional procedures could be regarded as a drawback
of using an inverting procedure. That said, inversion would still be a
useful capability to have provided, and an appropriate level of diligence
would minimise the associated drawbacks; one example could be to "cut"
various objects from a PCB file, then invert everything that is remaining,
then to "paste" the previously "cut" objects back into the PCB file again.
In other cases, when only *some* items in a PCB file are to be inverted (and
assuming that such partial inversion was actually possible), the user could
arrange to ensure that only those items which are desired to be inverted are
in a selected state when invoking the inverting procedure.

When I suggested (relatively recently) the provision of yet more layers, I
envisaged that these would *all* be of a paired nature, and with those
pairing relationships being of a *fixed* nature; I suggested that as an
*alternative* to my longer-standing suggestion of providing an
user-configurable pairing feature for the *existing* Mechanical layers,
because it occurred to me (over an extended period of time) that having what
in effect would be a variable type of pairing for those layers had the
potential to cause problems (which Ian Wilson has since commented on as

I did also suggest as yet another (and *distinct*) alternative the
possibility of Protel emulating Autocad and PCAD in regard of users being
able to define their layers as required, along with being able to pair such
layers to one another. If that was implemented, and if users so wanted, they
could define new layers that were solely of a paired nature. However, I
still think that if such a feature was to be provided, then users should
also be able to define layers of an *unpaired* nature, because users would
typically want some items to *not* change layers following the inversion of
a PCB file, even though followup action would often still be required
following an inversion procedure (to restore such text items to an
unmirrored state, for instance).

Using a Mechanical layer to define the PCB's border is one example of a
layer that users would not want to pair. (While it is not unheard of to use
the KeepOut layer for that purpose, the true purpose of the KeepOut layer is
to restrain where copper tracks can be placed, so while the border of the
PCB falls in that category, there can also be *other* areas within a PCB
where, for whatever reason, tracks should not be placed; in such cases, the
KeepOut layer is not *just* defining the PCB's border. And in any event, the
KeepOut layer, as implemented, is another example of a layer of an
inherently unpaired nature.) Another example is using a Mechanical layer to
hold text to document various aspects of a PCB, and/or to contain a
'.Layer_Name' Special String, for the purpose of identifying different
Gerber files and/or printouts. (If an entire PCB file was inverted, the
contents of the KeepOut layer and the Mechanical layer used to define the
PCB's border would typically *not* be touched afterwards, but other unpaired
Mechanical layers, such as those used to document the PCB and/or identify
the remaining layers, typically would need followup action.)

In the event that inverting procedures were provided (either by Altium or by
users), nobody would have to use them. However, in spite of the associated
complications that have just been discussed, I am picking that many users
would still welcome their provision. And as should have been made clear by
now, inverting a PCB is inherently safer than mirroring this, because
components then swap sides rather than having their (currently unregistered)
mirrored state toggled. (So any PCBs which are manufactured would be more
likely to be "assemblable", because there would be less likelihood of these
incorporating mirrored footprints.)

Geoff Harland.
E-Mail Disclaimer
The Information in this e-mail is confidential and may be legally
privileged. It is intended solely for the addressee. Access to this
e-mail by anyone else is unauthorised. If you are not the intended
recipient, any disclosure, copying, distribution or any action taken
or omitted to be taken in reliance on it, is prohibited and may be
unlawful. Any opinions or advice contained in this e-mail are
confidential and not for public display.

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
* Contact the list manager:
* Forum Guidelines Rules:
* http://www.techservinc.com/protelusers/forumrules.html
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Reply via email to