DURAVIT IN POINT WORLD

My last post took a look at some manufacturer BIM content and suggested that it could be improved considerably with a little extra effort: probably much less effort than it took to create the content in the first place.  One of my readers pointed out that there is some well-modelled content on the BIM Object website. It's certainly worth a look.  Probably over-modelled for some people's taste, and if you are going to aim for that level of accuracy you should consider using symbolic work in elevation views as well as plans.  But that's nit-picking.  Keep up the good work guys, we love you.

It was interesting to review the American Standard families, but I doubt that we will get to use thme here at GAJ.  Duravit is a different case.  They are quite popular in the Dubai market and when we use them there is probably an ID component in our scope, so it's worth having decent families.

 Over the years I have made several families based on downloads from their website.  They have very good 3d & 2d symbols, but no BIM content.  Sadly the 3d geometry often loses a face or 2 when loaded into Revit.  I'm not sure why.

Also there is often extra detail that we don't really want.  We don't section plumbing fixtures in Revit so we don't want to model hidden detail in 3d.  It creates an unnecessary load on the processor, especially on large projects.

Now that we can explode CAD geometry, there is a workflow that could be used to create smooth flowing forms in plumbing fixtures.  Model it in conceptual massing.  Export to .sat.  Bring into a standard plumbing template.  Explode
. 
So I decided to try this out on the Starck WC & Bidet that I had previously worked up using CAD imports from the Duravit web site.  It helps to have 3d CAD geometry as a guide when making these forms in conceptual massing. 

Set up a series of profiles.  Create form.  That's the basic bowl done.  Now for the rim.

Create form from one profile, gives an extrusion.  Make another extrusion crossing this horizontally. Change to void.  Voila.  But we are left with sharp edges and it's not so easy to round these off in Revit.  Where plane surfaces meet at right angles you can use a void sweep, but curved surfaces are tricky, especially when they curve in 2 directions.

Fortunately we are going to export to CAD anyway, and AutoCAD has solid modelling tools that eat this kind of task for breakfast.  Fillet edge, click on a set of edges, type in a radius, enter, enter ... done.

I went through a couple of iterations on this, going back into Revit and adding the recesses for the fixing bolts and cutting out a bowl for the bidet. 

 
This is the original Revit geometry, so I will have to round off all the edges again.  Export to SAT, open in Autocad, find the solid editing tools, Fillet Edge.  I used a soft curve to remove the edges on the underside also.  Probably you would never notice these, but it's easy to soften them out, so why not?   Softening the sharp corners does make a huge difference to rendered images.

Actually I discovered later on that those fillets bump up the file size fairly quickly, so don't get too carried away.

Bring the SAT back into Revit and explode.  No problem.  Apply materials and subcategories.  I modelled the seat directly in Revit using a void sweep to round off the edges.  Added a tap for the bidet (based on the CAD import from the original Duravit download)  Set up visibility controls to swap between WC & Bidet (one family, 2 types).

I already had the symbolic representation for plans & elevations from my previous efforts.  So that was that.  Half a day's work, divided by 1000 architects world wide ... has to be worth it.  Please, please Mr Duravit, give us some BIM content.

The image above shows my current collection of Duravit families.  I started by dropping all the cad downloads into one family.  That way I can copy paste them into individual families whenever I get time.  So far I've only converted a handful of these. I probably need 3 or 4 weekends to do the rest. Just imagine if all this was available as nicely crafted native Revit families.  It's an impressive range, and you have to admit their styling is really cool.

There's another post to come yet in this series.  After that I will try to bundle some of the best / most useful families up and make them available.  Maybe I can even persuade some of you to chip in and convert a couple.  Many hands make light work.

TOILET TRAINING

Why do I care what a toilet looks like ?  It's just a functional object ... right ?  An interesting question from various points of view.  Freud would have something to say I'm sure.  But it's clear that we do care a great deal about the appearance of our bathrooms.  Check out the websites of Duravit, Hansgrohe, Ideal Standard ... The whole "health & wellbeing" thing is very big when it comes to toilets.  People want to feel like they are going into an oriental spa to cleanse their spirit.

But we are just putting toilets on our drawings, aren't we ?  All we need is a symbol to say "put the toilet here" and a code to say "find more details under this spec reference".  Wrong again.  We have always cared what our drawings look like.  They should be crisp & clean with just a dash of style & verve.  It's been that way ever since we drew by hand, we cared about drawing style when we switched to CAD & we still care now that we are using BIM.

So this is (another) appeal for better sanitary ware families.  Where is the good looking BIM content to help us sell our bathroom designs to our clients ?  We could moan at Autodesk for not providing better generic families.  And I have shown before that this is not so hard. 

http://grevity.blogspot.ae/2012/05/plumbing-blockage.html

http://grevity.blogspot.ae/2012/04/my-weekend-on-toilet.html


But ultimately content is up to the end users.  Not just the designers,  I'm talking about the global construction industry.  Architects & Interior Designers can set the direction, Clients & Contractors can create the demand.  Manufacturers & Suppliers are beginning to grasp how they can participate in the BIM initiative.

Take a look at office furniture.  People like Haworth, Steelcase & Herman Miller have been working on their Revit families for a while now.  They are into the second & third generation versions of their content, listening to feedback, improving quality & thinking hard about how best to deliver this.  Just check out their websites.  I'm not saying they have achieved perfection.  Some will say that they want more lightweight content (2.5d perhaps) but you can't deny that they are trying hard.

Over a year ago I took a look at some plumbing families available on the internet. (link) 

Now we don't really use American Standard in our work.  Clients tend to go for the European brands.  But I looked at their stuff because, unlike Duravit or Roca, they do have BIM content for download.  I was hoping that by now they would have added more content & corrected some of the obvious shortcomings in their collection, but sadly no.  It seems that someone at American Standard took an interest in BIM 4 or 5 years ago, got a few families made and then lost interest. 

That's probably our fault.  Has anyone been giving them feedback or sending in requests ?

So here is my "open letter" to them.  It is offered in a spirit of positive criticism as a "neutral observer".  I believe it would be quite easy for American Standard to offer really good Revit content and I hope to motivate them to do so.  They are on the bus, but they need to start talking to the other passengers, maybe get a bit of a party going, turn it into a magic bus.

Point 1.  Please use symbolic representation in plan views.  This is an abstraction, like any other graphic symbol.  We need just enough detail to capture the spirit of the design and to look good at a variety of scales. 

Point 2 Please think carefully about the origin.  This is done by setting 3 reference planes to "defines origin".  If the fitting is wall mounted, the "centre front/back" should be the face of the wall.  The "centre left/right" is usually straightforward.  The horizontal origin plane could be finished floor level, or it could be the rim in the case of a sink.  It should not be the underside of a sink where the waste connects.  When the family is placed in a project, it

Point 3.  The 3d CAD inserts that occur in about half your current content can now be exploded to native Revit geometry.  This is a feature of Revit 2014.  I have tested this, and it so happens that it works really well with your existing families.  The advantage of this is that you can now assign a material parameter and subcategory directly to that geometry.  Also many end users object to embedded CAD objects within Revit, so you will make those people happy.

As an added bonus, exploding the geometry removes many unwanted edges & seams, giving a smoother appearance in shaded views.  Revit users will be now able to create very nice visuals for their clients.

Point 4.  Create collections.  You could load all your sinks into a single Revit project and make this available for download.  End users open the collection and select the product they want based on user-friendly visual information.  Copy-paste and you're done.  Let people know when the collection has been updated and they can download the whole thing again to stay current.  Easy.

Point 5.  Set up sheets in your collections.  Make it easy for the end user to see what your families look like:  in plan, front elevation, side elevation, 3d shaded view, 3d rendered view.  By setting up these views you can also do some basic Quality Control.  Are you happy with the way they look ?  Maybe you can also provide some examples of how your fittings can be tagged and scheduled.  Make people aware of the wealth of information in your BIM content.  Check that this is coming out the way you want it to.

So that's my open letter, ostensibly to American Standard, but potentially to any Sanitary Ware manufacturer/supplier in the whole wide world.  Some hits on how to "do better" on the BIM front. 

In the next post I'll take a look at Duravit who don't do BIM content, but do have some very tasty products up their sleeves.

2D BIM

First of all, a big thank you to Zach for some very kind words, for running his inimitable competition and for motivating me to get carried away again.

I intended to clean up my Desert Pumpkin families for upload during the course of last week, BUT work just got too hectic.  Towards the end of the week I got wrapped up in an exercise which I thought worth sharing.

It's a design and build project for a hotel with prefabricated bathroom pods.  The exercise involved enlarging the pods while maintaining the minimum area requirement for the guest rooms, which happens to be 24 sq m.  There are several different permutations of pod type and bedroom type.

We had to come up with a drawing that explains our proposal in order to achieve client sign-off before committing changes to the model.  A classic case for 2d BIM, which is my nickname for "drafting over model views" It would be possible to make a second copy of the model and implement changes in 3d.  But it seemed to me this would require more effort, and it would be difficult to show existing and proposed as overlays in the same view.

We are familiar with the concept of 2d BIM as a detailing technique, but I believe that it has many other analagous uses, and as such it is a method that deserves much more attention and respect.  It is particularly useful in a situation where you are exploring planning options, where it is directly analagous to laying sheets of tracing paper over hardcopies and sketching as you talk during a design workshop.

So I created a parametric detail item with 3 types, representing the 3 main pod types. There are a number of dimension strings that lock various fixed relationships in place, plus 3 variable parameters (also one visibility control).

Overlay this on a plan callout and dimension it up.  I'm using two colours of dimension for existing and proposed.  The resulting diagram is quite busy, but for team members who know these pods inside out by now, it's immensely informative.  Also you can go on generating variations until the cows come home, either by duplicating views with detailing, or by simply swapping out family types and printing off the sheet.

These same detail items are placed in a key plan, and used as guides for the edges of filled regions, one for each of the permutations of guest room/pod (we have six at present)  The edges of the filled region are stretched to align with the outer edges of the detail item and give a readout of the resultant area.  Unfortunately we can't tag the area of a filled region (seems like an unnecessary limitation) but it's no biggie.

So we have a key plan which summarises all the different situations and together with the various callouts for the pod types tells a very rich story.  Furthermore, the 2d world represented by this sheet is a very productive design space.  It's a great place in which to work while grappling with the problem at hand.  2d BIM is a very powerful tool for thinking out problems, analysing and abstracting, defining the critical issues, staying fluid and flexible.  We should use it more often.

So there you go: a nice, short, practical post to contrast against the epics of the past few weeks.  Call me schyzo if you like, but I enjoy using Revit to range all the way from drafting techniques to the conundrum that we call art.  That's why I call it my BIM pencil.