Xref - Layer 0

[Domz]

Is it necessary to put the XREF drawing into layer 0?
Whats the advantage of this process compared from putting the XREF into another common layer?

FILE/DRAWING SET-UP:
ROOT 01.DWG - ROOT DRAWING (CONTAINS OBJECTS AND BLOCKS)STRUCTURAL
ROOT 02.DWG - ROOT DRAWING (CONTAINS OBJECTS AND BLOCKS)MEP
ROOT 03.DWG - ROOT DRAWING (CONTAINS OBJECTS AND BLOCKS)ARCHITECTURAL

COMP 01.DWG - COMPOSITION DRAWING (CONTAINS OBJECT, BLOCKS AND ROOT DRAWINGS-INSERTED AS XREF)
COMP 02.DWG - COMPOSITION DRAWING (CONTAINS OBJECT, BLOCKS AND ROOT DRAWINGS-INSERTED AS XREF)
COMP 02.DWG - COMPOSITION DRAWING (CONTAINS OBJECT, BLOCKS AND ROOT DRAWINGS-INSERTED AS XREF)

FINAL 01.DWG - FINAL DRAWING WHICH CONTAINS OBJECT, BLOCK AND COMP. DRAWINGS-INSERTED AS XREF.

[Syd_Derbyshire]

XREF's (or Blocks) all in Layer 0:

Visibility is controlled by Freeze/Thaw (preferred), or On/Off by selection of
layers comprising XREF or Block. Controlling layers in latter will probably
affect the parent drawing as well. Controlling layers in XREF can be time
consuming both in Model and Paper Space.

XREF's (or Blocks) all in dedicated layers:

Visibility (global) is controlled as above except that the selection is simply one layer per XREF or Block.

[Paul Aubin]

The issue is really one of consistency and management. theoretically, it does not matter. The biggest issue is that if someone freezes the layer of insertion, then the entire XREF disappears. So, from a management point of view, you can mitigate the potentially negative effects by requiring a consistent layer for insertion. Some argue for layer 0, others a custom layer. layer 0 has special behavior when nested within a block or XREF, so some prefer to use a custom layer for XREF insertion.

I think the most important issue is choosing a procedure and doing it consistently. In the long run, consistently inserting onthe same layer every time will solve many more issues than the specific layer chosen will.

[Cap'n Mac]

Well, actually, I really prefer to plant XREFs on Layer XREF. If the XREFs are complicated, I prefer to give them each a layer, so that they can be truned on and off with Freeze/Thaw as needed.
That leaves 0 free for Block & similar development.
But, that's me; others differ.

[Domz]

thanks for the inputs!!!
hmm...i guess both are the accepted process...
i always thought that putting the layer of insertion to 0 is the ideal one (for xref)
then again i have to study Cap'n Mac inputs coz i didnt used this method...
and stress the issue of consistency with full force...
ive been working for 2 drawing for 3days now and used a global base plan that contains loads of xref...this gave me a hard time controlling the visibility of some objects....
and frankly...i dont know what to do...considering 20 cad draftsmen and no cad manager....maybe im tired no and needed a few hours of sleep...

[Domz]

Hello!
just finished looking at the drawing files and heres what i found out with regard to the files ive been working into:


one of the problem im pointing out is the control of visibility particularly on using layfrz on express tools.

my initial thoughts were the inconsitency on layer of insertion with the xref and block.
what you think about this?

[Paul Aubin]

I don't really see an incosistency. Your diagram looks very typical - recommended in fact. All objects in blocks shoudl typically be on layer 0 unless you have a good reason not to. In tradiditonal AutoCAD, best practice is typically to have a dedicated layer for insertion of XREFs. Some call this simply XREF like Cap-n-Mac or others have a LISP or something that creates unique layers. I use AutoCAD Architecture (ADT) and so for years, I have not been XREFing manually, but rather letting ADT do the XREFing for me. This is why in the original post I said I could go either way. To me if you have a routine doing the XREFs for you, (navigator for me, LISP for others, still others may have something else...) then let it do its job. But if you are managing the XREFs manually without the assitance of any automation tool, then I would at the least create an XREF layer and use that for insertion. Bear in mind that regardless of the layer, you can always turn off the layers in the XREF independently and you can always unload the entire XREF without detaching. You don't therefore need a special layer for each XREF, but some folks still prefer it. Finally, express tools should reach into the XREF and freeze the nested layer automatically, not the layer the XREF itself is inserted on. (Unless of course they are the same like layer 0) and this is in fact the reason many folks do not like to insert XREFs on layer 0...

[Cap'n Mac]

Yeah, like PA says, that's a pretty clean object tree, overall. It might have some funky visibility issues--but at a certain level of complexity, many things do just that.

That's not at all as bad as the file I just got from a civil consultant. It has an xref of our previous base drawing we sent them, plus four overlapping Blocks of just parts of our drawing overlain on parts of the surveyor's drawing. No two of those objects are inserted on any other layer than what ever was current when the parts were added. Oh, and there's three layers of layer-dependant Blocks across this mess, too. This makes for exciting visibility controls (even more excitment for editing the information).

That's life behind the CRT (or flatscreen in the hip offices).