This is not a Caesar bashing thread!

But as the Stress Group Lead, i am under constant pressure to reduce the time for analysis from the Project managers. We have a new Engineer with us who has used all the major stress programs on the market, namely PSA5 Autopipe, Triflex etc.

On single line analysis, Caesar is pretty much ok, but start to throw in systems and we start getting bogged down in modelling etc.
My last system was a 1000 node steam condensate system which took 3 weeks to model by a junior. The ineractivity and nature of the Design require a single model hence the size.
The use of PCF files etc is a Pain and can be more trouble than its worth,
The PDS intergrapgh module is ok, but requires effort from both sides.

At the risk of burning in flames, is it time for Caesar to take a radical step forward in the way the mechanical models are created and manipulated?
(I could mention a conflict of intrest here with CADWORX)
Lots of things could be added to the program to improve it, such as
Drag and drop,
Improve (lots) Isogen interfaces
Improve PCF interfaces
Improve graphics capabillity.

From someone who used to be a Piping designer, Caesar can be a clumsy ignorant tool. And i can't help but think that the way caesar interact with CADWORX should be translated to other programs, and that the manipulation and creation of the models needs to be radically changed.

May i burn in hell, but i am trying to get 3 projects of the table at the moment and at every turn, caesar is hold us back instead of pushing us forward. I appreciate that because of caesar, semi skilled engineers are now employed throughout the world in stress analysis, but is now the time for a new direction??

A fair enough question really...

What comes to my mind is that other pipe flex programs probably have similiar limitations talking to 3D cad programs. We are all held captive due to a lack of coherent standardization.

CAESAR II's wonderful coherence with Cadworx is due in no small part because the same company COADE controls each piece of software. Autopipe probably works well with Autoplant but may not get along with ALL the various 3d packages.

One of the designers here wants software that will tell him interactively if his line routing is "bad" as he routes the line in the 3D model. While this is the goal I don't think software will ever be able to do this.

Anybody else has experience with other pflex programs collaborating with 3D packages????

(Having used punch cards and Mare Island, what we use now seems surreal at times to me.)

I use CadWorx and I use CAESAR. If you need drawings for manufacturing and erection doing the lines in CADWorx and then dropping the lines to CAESAR can save a lot of time. There are problems in getting the info form CADWorx to CAESAR. Even with problems I sometimes do the model in CADWorx first just to get the dimensions correct. Typically this is the case where lines are existing and I have to measure. In Cadworx I can model the building, vessels etc. I have also taken land surveyors pipe measurement data to CADWorx and from there to CAESAR. All electronic.

Couple tips:
- CADWorx model has to be drawn correctly. If you can get ISOGEN ISO then you should be able to get acceptable export
- When exporting export line at a time and combine in CAESAR. This way you get better node numbering
- Export before you include any "minor" items like drainpoints, instrumentation and similar. Such items just create useless nodes.
- In the beginning you need to test with the draughtsman that everything is OK
- You need to do some cleaning up. For instance supports are in correct position but most of the time wrong type.

Main time saving comes from the fact that only one modelling is required for two purposes. I have to admit some models are faster using CAESAR directly.

I guess that it would be too much to expect Autoplant to talk to Caesar II. However there is little excuse for Intergraph PDS, Smartplant, PDMS et al being so slow to produce a reliable and seamless interface. Intergraph did have a go years back, called PSTRESS. It collected line data, did a little plot and created a good CAESAR II input, but it seems to have been lost in the mists of time.

Punch cards and Mare island ... is that worse than my start point, punch cards and ADLPIPE ? I never, ever got a run to go first time.

Tim,

Let's go back a bit and take a "big picture" look at your problem statement.

"I am under constant pressure to reduce the time for analysis from the Project managers."

Aren't we all. But the nature of project managers is to never be satisfied - as soon as the technology becomes available for us to be able to model systems much faster than our PM's current expectations, the PM's will change their expectations. And you will still be under pressure, but with MORE projects going on at the same time and shorter deadlines.

ARE YOU SURE YOU WANT WHAT YOU ARE ASKING FOR?

As employees of various local, regional, national, or global engineering firms (for the msot part) we are all in competition with each other to get and do more work at lower rates. If the tools available limit the pace of your work, most likely they are limiting the pace of others similarly. (Unless, of course, your firm insists you do the analysis by hand.)

Likewise, CAESAR II is in competition with Autopipe, Triflex, etc. to create software that can boost the user's productivity. My guess is that they spend a god-awful amount of time and effort to try to figure out how best to import geometry data from CAD files.

As one who is nearly as old as Luf and Breen (usually I phrase this as "a lot younger than" when I correspond with them directly), I have also used punch cards and Mare Island for stress analysis. Personally, I will miss the day when I can just drag and drop a model from a CAD program into the analysis engine. No, not because I like the process of data input. But I believe that I gain considerable understanding into the probable workings of the piping system during the data input process, and also I have caught way more than a few silly design errors while doing data entry.

Also, note Jouko's caution - "- Export before you include any "minor" items like drainpoints, instrumentation and similar." I seriously doubt that this option will be available to you. So I expect that our fondly-desired drag-and-drop modeling will then require a lot of tedious examination of the model before analysis to remove extraneous nodes. And, since we are human, I'll bet that a lot of us take a while to adjust to this brave new world, with one or two gruesome errors along the way.

Adlpipe had an "improved" basis and though user hostile was less so than Mare Island....

In the old days of punch cards you were either fit for the work or you were excluded by your inabilities (became a hanger banger), you had to have an eye for detail.

The current state of affairs is often times a sad joke... people want to do meaningful work without doing ANY homework themselves, and often times post questions here and elsewhere that clearly indicate that they have done absolutely NO leg work on their part!

The problem as I see it is one of commercial interests. COADE because it writes both software packages (Cadworx and CAESAR II) can make damned well sure they talk to each other. But there is no commercial incentive for such cooperation between competeing software firms. As a matter of fact my guess is that the publisher of 3D Cad software might want to make sure that he either favors all or none I suppose.

Tim
Sometimes a change of tack works well with non-technical managers. Gently remind them that if something goes wrong with a system you have engineered, as it will be you and not the 'engineers' or 'designers' that will be held responsible, people may be hurt or killed and you could potentially go to jail for negligence. However also point out, with perhaps reference to say the recent Texas oil refinery incident or the Challenger disaster, that when it is later found that management put undue pressure on engineers, cut safty budgets, overruled competent advice or disregard any safety related topic in order to save money, that the law tends to come down much much heavier on them and the company, than any poor unfortunate engineer.
Putting things ins perspective and suggesting they could go to jail often helps justify budgets!

Schedules and budgets versus what the code says.... One has to juggle all these priorities.... a desire to reduce analysis time can be accommodated by not doing a computer analysis, in order to accomplish this the system must be layed out in a fashion such that its flexibility is readily determined by a simple visual examination.

So there’s your trade off if management decides to have the designers do the layout (not all designers are competent, just as not all pipe flex people are not competent) and the designers have a layout which is not readily judged to be adequately flexible, they have to decide to whether perform additional analysis and bear its costs and schedule impact or simply change the layout.

You cannot have both... right now as in the foreseeable future computerized analysis will take an increment of time. Thirty + years ago when I was doing Spielvogel calcs that took weeks I controlled the layout far more than now. Now I often get "You have to make it work" I usually come back with... "For enough money and time I can get almost any marginal, poorly thought out layout to work but bring out the cash."

Go Blues (Tie in The Fa Sheesh)

Back on topic and cutting to the chase.

I would like:

a) easier transition from model to Caesar (a la Cadworx) with whaterver cad package.
b) For COADE to start spending some development time on the interface, including all the little suggestions this forum.


Manual input is fine for high tech single lines, but i want to look at a system and don't want to spend weeks coding it. I'm fully capable of interpreting design without having to model it inch by inch.

Bentley,Triflex,PSA5 etc are already improving and reinventing there stress programs. I hope COADE don't get left behind

I looked at the Triflex web site.... seems to be a similiar interface to CAESAR II could you lend us some specifics please....

http://www.pipingsolutions.com/triflex/p...ode%20Pts.pps#6

I've only ever used Caesar, and for that i think it is a wonderfull program. The back up support is fantastic.
I have a new (but old) stress engineer started recently who is banging on about how good Autopipe and Triflex are and that these programs are in heavy development.

Little things like thining nodes out as per above have been sugested here numerous times, yet are ignored by COADE. WRT to the outputs, it could be argued that the program is actually going backwards.(ref finding max forces etc, or the speed of using the 3D output viewer)

Like i said, i am not caesar bashing, but as someone who has used the majority of 3D design packages on the market today, the interface could be a whole lot better.

A nice feature....