Project Navigator Help

[bradg]

Hello,

As our company is attempting to transition to MEP 09, I'm trying to evaluate the PN to see if it is worth changing our workflow to accommodate it. I'm currently working on a project using the PN to create sheets, views, etc. that we would normally create using 1 drawing with several layout tabs, and have run into some issues/questions that I was hoping those of you more experienced with the PN could help me with. They're numbered with somewhat of a topic statement, so even if you don't have time to read through this massive post, if anyone can help with just 1 of the questions it'd be much appreciated.

Constructs/Views/Sheets:
1) Annotating Views?:I've gone through the tutorials using PN, and AutoCAD kept mentioning that you should draw your model in the construct and annotate it in the view. Does anyone know the reasoning behind this? Several of our engineers like to annotate as they draft, and having to draw a duct in one drawing, then save, switch to the view, reload, and add a diffuser tag seems like a lot of extra work. I know you can still add tags in the construct, but what is the advantage/why is it recommended to annotate in the view (or sheet)?

2) Updating layer colors: Is there a way to have layer colors that are changed in a Construct update in the View and Sheet? Recently, we were in a pinch needing to get something to plot correctly, and went to the construct and changed the layer color, under the presumption that it would also change the color in the sheet, and voila, our plotting problem would be solved. However, after several xref reloads, the colors never updated in the sheet or the view, and changing the colors in the view drawing didn't work either. Ultimately, we just changed the xref layer colors in the 2 sheets that we needed to print. (I'm now looking into using .stbs instead of .ctbs)

3) xRefs in Views: Is there a way to avoid having to re-add a drawing that is xref-ed in a construct (e.g. an architectural background) when you create a view based on that construct? Not too big of a deal, since luckily the xref gets placed at the same coordinates, but it's just a nuisance to have to re-clip the xref, change layer colors, etc.

4) Layers!: Anyone have a good way to use multiple model space views with different layers frozen/thawed? Having the Project property "Match Sheet View Layers to View" set to "Yes" is no good, since we'd like to have ductwork and piping in one construct/view, and then create a ductwork sheet and a piping sheet depending on which layers are hidden in the viewport. I've also tried freezing the layers in the view, and attempting to 'Save Layer State' by right clicking the model space view associated in that view, except every time I try to add an mspace view with a saved layer state to a sheet, I get an INTERNAL ERROR and AutoCAD crashes.

5) General Notes: Anyone know of a good way to handle General Notes? Should you place the note markers on the construct/view and then the actual notes on the sheets? Does AutoCAD happen to have a way to easily manage the notes?

6) Disappearing Equipment (*solved* see next reply): I've noticed that occasionally a sheet will not show certain equipment in the viewport (air diffusers, VAV boxes, etc.). It seems to be completely random which objects don't show up, as some diffusers will show, while others won't, and unloading/reloading the xref-ed construct and view doesn't cause them to re-appear. I got them to re-appear once, but (unfortunately) I can't remember what I did. Anyone else experiencing this?


General Project Navigator:
7) Working Remotely: For the particular project I'm testing the PN with, one of our engineers is working away from the office. All of the project files are stored on our server, and he can access the server through a VPN. However, the VPN is too slow to allow him to work on the drawings over the network, so he was told to download the drawings he is working on to his local machine, set the network version to 'read only' and then re-upload the drawing(s) when he is finished. This works somewhat well when using 1 drawing with multiple layouts for the sheets, but can become complicated when using the PN, since each sheet needs to have an associated construct and view downloaded as well, and making sure to re-upload the files back to the correct location can be troublesome. For now, he has a copy of the project folder on his machine, which he updates about once a night, and then uploads any drawings he changes back to the server. Does anyone know of a better way to do this? If AutoCAD doesn't have some sort of feature built in, then perhaps a utility program that eases the download/upload process, and keeps track of which version is newest? Or maybe even a way in AutoCAD to package the project so he can download one (hopefully smaller) file and then unpack it on his local machine?


If you made it through all 7 questions, thank you very much for your time. If you know of any possible answers to any of the questions, I'd really appreciate it.

Thanks in advance!
-Brad

[bradg]

Found a solution for #6:

When I drew the 3D equipment in the construct, I was basically tracing over a line & block drawing that the engineer created. For some reason, the blocks he used for the terminals were drawn above some of the MvParts terminals, and in the sheet it was causing the mvparts terminals not to show up, since the layer for the block terminals was frozen. It still strikes me as bizarre why some would be drawn above, while others weren't, and even moreso why freezing the layer in the viewport didn't allow the mvpart terminals to show through.

Oh well, nothing that a simple draworder in the construct couldn't fix.

[dzatto]

Well, for number 3, I would make my Xref background an element. then drag and drop it in the construct. That way, it's an attached Xref, not an overlayed Xref. Attached Xref's will follow the parent drawing, overlayed Xref's won't.

[Brockster]

Here is my take...

1. The Construct files can be considered as the base reference files. They can be shared between disciplines. Because of this, other discipline do not need the notes, dimension and general symbols of another. Discipline specific annotations belong in the discipline's View file. View files are never shared or referenced by the other disciplines. There are component notations that do belong in the construct. An example of this would be the ABS duct work. As the duct work tool is used for the duct layout, the duct size text automatically populates the duct run. I do feel component & device callouts do belong in the Construct. Another example would be room, door, window callouts, which need to be in the Construct so they can be viewed and referenced by other disciplines. A callout that is needed by other trades belong here. The annotations in the View file are notes, dimensions and general symbols.

2. I don’t know a fix between the Construct and the View, but there is a fix between the View and the Sheet file. In the Project properties dialog box, set the “Match Sheet View Layers to View” line to YES.

3. No and a little yes. When a Construct is dragged into a Construct, the PN makes it an overlay. The little yes is really an exception. There are times when you may need a composite plan. I’m on the architectural side of things and sometimes recommend using composite plan when many constructs make up a plan. In this case I manually add the reference and set it to attached. There just no way the engineers are going to know the name of every construct needed to create the plan. This composite simplifies their life.

4. Create

5. If it’s a sheet specific general note, it belongs in the View file. If it is spanned to several sheets, make the note an element and drag it onto your Sheet file. You could ever make a View Notes.dwg file of various notes, make views around each note group and drag them onto your Sheet file(s). Notes never are placed into Construct files.

6. Resolved…

7. My only comment here is it seems the Xref pathing would change from UNC paths to logical named paths. Ex \\usacb12\project\Camel Project\123456\Construct\M\Plan\01 floor.dwg to Q:\Camel Project\123456\Construct\M\Plan\01 floor.dwg

Well, there's my two cents worth...

[stelthorst]

Hi Brad,

I think brockstar has done an excellent job of explaining things but I'll add my 2c.

My take on the Construct/View/Sheet debate is this. A construct should contain only physical items (ie ducts, pipes, conduits etc) This is useful especially when coordinating in 3D as it creates a "clean" file to import into Navisworks. A view is similar to a viewport and should contain annotation and text. At first I thought the use of a view file was a complete waste of time but have come to realize it is an important step in creating a sheet set that will update and work together as a unit. The view allows the user to drag the viewport into a sheet and the PN understands what sheet it was used in and will update all drawings that reference this view. A sheet should contain the view, border and any other sheet related entities.
I believe VISRETAIN still works in a project environment (Try setting it to 0 in the view drawing modifying the construct and reopening the view) In addition, In the properties of a project (Go to project browser click on a project and right click and select project properties) you can set "Match Sheet View Layers to View" to Yes and the sheet should match the layer state and color of the view. The third option is to create layer state(s) and copy them to the sheet.
The PN is structured so that a construct will overlay another construct and attach an element. The reason for this is to keep the drawing "clean" and avoid circular references. When you create a new view you should be able to set the level and division and only those constructs will be available to add to the view. You might want to evaluate how you use the PN, there is no reason to clip a construct only the view (IMHO)
This is a tough one. In theory you should be able to create multiple model space views with various layer states and they should work when you drag them into the sheet. I, like you, have had problems getting consistent results with this method. The only suggestion I could make is to draw your pipework and ductwork in separate constructs and then only bring the construct you need into the view.
I create my general notes as an element and drag them into the sheet drawing. Don't know if it's the best method but it works for me.
Glad you found a solution
What some of my designers have done when working remotely is to copy all of the project files onto a zip drive with the same folder structure as our network. Then when they add their zip drive they make sure to give it the same drive designation as our network (in our case F: ). Doing this allows them to open PN files without having to re-path anything. Probably not the best/easiest way but it works. The only caveat to this method is we try to have only one designer assigned to a project so keeping track of the latest version/file is not an issue.Hope this helps,

[bradg]

Thank you all for the replies, I'll do my best to respond to each issue.

1) Thanks for clarifying constructs, sheets, and views. I'll run it by our engineers and see what they think. I have a feeling that we will end up doing something similar to Brockster's solution, since having to switch between constructs and views just to tag a piece of equipment seems like a big hassle. That said, here's a new question: Where should the schedules go? I'm only beginning to get more acquainted with how to make schedules, but it seems like it's easiest just to put them on the construct and not have to worry about linking to another file. But that's no good for keeping the constructs "clean", so I guess the question is, schedules on views or sheets?

2) Haven't gotten a chance to test out the solutions mentioned. Hopefully it won't be a problem, since I'm starting to get the hang of .stb plot settings instead of .ctb. However, I'm weary of using the 'Match Sheet Layers to View' for reasons covered in #4.

3) I think I'll give dzatto's solution for this a shot. I knew about the constructs being overlayed, and elements being attached, but hadn't quite thought enough 'outside the box' (sorry) enough to think of using a background as an element. Scott, I was having to clip the xRef because the architect had included blocks to the side with the instructions "DO NOT DELETE" (although I've never actually tried deleting them to see what would happen), so I clipped the xRef so they wouldn't show up anymore. But now that I think about it, making modelspace viewports should take care of this for me as well.

4) I think I'm going to have to get creative with the Layer States here. The last time I tried 'Match Sheet Layers to Viewports' it gave me undesired results, since it would set both the ductwork sheets and piping sheets to the same layer settings. The 'Save Layer States' option for the mspace view really seems like it would do the trick, but unfortunately, it causes my system to crash every time I try to use it. Separate constructs for the ductwork and piping is something I had considered, but the one of the main issues there is connecting the equipment to both ductwork AND piping. I know that you can still use AMEP snaps when you overlay a construct, but even then the 'disconnect marker' still shows up, and it's hard to tell whether or not all your connections are made. Furthermore, it would be tough to create sheets for the piping if all of the equipment was in the ductwork construct (or vice-versa), and adding the equipment to both drawings seems time-consuming and difficult to keep coordinated.

5) Great idea about using elements for notes. I still need to get a better feel as to how our company uses notes, but that seems like a good option.

7) For now, we have our project set up to use relative paths, so that he can download the entire project folder and work from his machine. Having him keep all the files on his computer is no good, sine there will be others here in the office that need access to the files as well. This is beginning to seem more like an issue where we need an additional piece of network management software, so perhaps I'll find a networking board to post the question to.


Thanks again for the replies, they've definitely been a huge help in understanding AMEP.
-Brad

[bradg]

Good news and bad news regarding #4:

Good news first: Autodesk was nice enough to grace me with an email containing SP1 for MEP, telling me that it should fix an error I was experiencing (no specifics as to which one). It appears as though, after installing the service pack, my AMEP no longer crashes when attempting to add Model Space Views with saved layer states to a Sheet. Huzzah!

So, here's what I was hoping to do with them: I have a feeling our company is going to want to draw ductwork and piping into one drawing, therefore, I'll have to 'separate' them inside the View/viewports. What I tried was to add the mechanical construct to a View, create two Model Space Views (one for ductwork, one for piping), turn off the piping layer(s), right-click the Mspace View for ductwork, click Save Layer State, turn on piping, turn off ductwork, save the Layer State for the piping view, save the View drawing. So that way have two Model Space Views, one for ductwork (that has all the piping layers frozen) and one for piping (that has all the ductwork layers frozen).

Now for the bad news: I first tried placing the Mspace Views onto a sheet with the 'Match Sheet View Layers to View' property set to 'Yes', but it seems that this property is like a 'master override' for any layer states. That is, no matter which Mspace View I added, the viewport showed whatever configuration the View drawing was last saved in. So then I tried adding the Mspace Views with the 'Match Sheet View Layers to View' property set to 'No.' With the propert off, I can drag the Mspace Views to the sheet, and while moving the view around while selecting an insertion point, everything looks fine: the ductwork view previews the ductwork, and the piping view previews the piping. But as soon as I click an insertion point, everything inside the viewport disappears. It seems that what's going on is once I place the viewport, all of the layers from the original construct become frozen in the viewport. Why?!?

Having to go back through inside the viewport and freeze/thaw the layers I want completely defeats the purpose of (what I think) the Save Layer State does in the first place.


What I'll probably end up doing is creating two separate View drawings for ductwork and piping, and leave 'Match Sheet View Layers to View' set to 'Yes'.


(P.S. changing the VISRETAIN didn't seem to have an affect on the results either)

[david_peterson]

If you need to get some info on a "how to" and Concepts of PN, you can go to Paul Aubin's website and download chapter 5 of his book "Setting up the Building model". You may find this info very useful.

[stelthorst]

Hi Brad,

Here are my responses to your two previous posts.


I put my schedules in a view drawing for two reasons. First, to keep my construct "clean" (I am hung up on that aren't I?) and secondly I work on large scale projects and break my drawing up by panel or area and adding the schedule to the view allows me to add multiple constructs to the schedule.
Glad you found a fix. I've always used ctb plot styles but will have to look into stb more to see what advantages there are to my work flow.
I do a lot of file sharing with other contractors, engineers and architects so I would be hesitant to add the background as an element but if it works for you more power to you.
I have had similar problems with multiple model space views in one view drawing. My final suggestion would be to turn the layers on and off that you want and save the layer state via the layer dialog box (upper middle of the dialog) and then importing the layer state into your sheet.
Glad you found a solution
I'm not familiar with any software to do this but if you find one please post it so we can all see what it is.Hope this helps,

[cridder_]

bradg

any fix for layer issue in viewport?

http://forums.augi.com/showthread.php?p=1054632#poststop

[bradg]

c.ridder,

Can you be more specific as to which one? It's been a while since I wrote the questions, so forgive me if I'm not so quick on the uptake. I'd be happy to share my experiences though.

-Brad

[cridder_]

sorry about that. it would be the layer issue with placed views.

'' But as soon as I click an insertion point, everything inside the viewport disappears. It seems that what's going on is once I place the viewport, all of the layers from the original construct become frozen in the viewport. Why?!? ''

it worked great in adt/abs 2007:razz:, not so much in 2010 arch/mep:cry:. I have a ticket with dealer and autodesk and they are working hard to solve. i can recreate with 2010 32 and 64 bit using out of the box profile and sample project.



c.ridder,

Can you be more specific as to which one? It's been a while since I wrote the questions, so forgive me if I'm not so quick on the uptake. I'd be happy to share my experiences though.

-Brad

[bradg]

Ok, gotcha.

Unfortunately it's been over a year since I set up that project, and our company has since shied away from using all of the Views and Sheets inside the PN (too much hassle without a clear payoff benefit). However, what I *think* I ended up doing was setting up Layer States inside of the Constructs (one for ductwork, one for piping, etc.) and then once the Views are created, going back into the Layer States Manager and applying the States that are referenced from the Construct. (I didn't set this up/figure this out until late in the project, so it was kind of a pain going back through all of the Views and such.)

If I recall, I think I gave up on the "Match Sheet Layers to Viewports" option, just because I couldn't get it to function correctly and reliably.

It also took some time to figure out the proper way to use VISRETAIN in all of the Constructs, Views, Sheets, xrefs, etc., so be sure to think that through to save some headaches down the road.

Hopefully that helps. If not and you'd like some more info, I may be able to take some time later to look through the old project and see how I set it up.

-Brad

[cridder_]

Thanks for the info. No need for extra effort on your part, we know how to do the layer states. Unfortunate that it worked fine in 2007 but now it is broke.

I have dealer and autodesk on it.

[dzatto]

I may be way off base here, but why not create a separate construct for piping, one for duct work, etc. etc. Then just create your views using whatever construct you need? Wouldn't that work?

[cridder_]

Not to much off base and we do have seperate constructs for each discipline, existing demo, new, etc so no problem there. The issue we are having is that when a view is placed on sheet all the layers are frozen in the viewport. worked in 2007, 2010 not so much. dealer and autodesk have recreated issue and are working on it.