Advice for CAD application developer working in hydraulics?

[Phoil]

Hi, guys,

I'm not an engineer, I'm a computer programmer by trade. Over the last few months, I've begun taking over some of the development on a CAD application geared toward hydraulics and sprinkler systems meeting NFPA standards.

My background and education is primarily in mathematics, so understanding the hydraulic calculation algorithms are no problem. However, not being an engineer, some of the terminology and practical aspects of this area of work have stumped me at times. I'm learning quickly, but there's still quite a bit I feel like I don't know about the perspective of an engineer using the software I'm working on.

Anyway, I wanted to first say "Thank you" for this forum. I've found a few answers to questions I had in the previous threads here.

But I also wanted to ask whether anyone here has any tips about learning the jargon and the usual progression of how these systems are designed. Anything would be appreciated, especially if it helps me more quickly grasp the core issue of problems and feature requests that come my way.

Thanks again, guys!
- Phoil

 

[SprinklerDesigner2]

I've yet to see what I would consider a "good one" yet.

I need an upgrade and took a look at the industry leader. It was nice, would do the job but the problem I had with it is everything had to be input just perfectly or it wouldn't work.

The elevations had to be input taking time. Seemed to me it took more time then to just input it to HASS and do it.

With the exception of a gridded birdcage, I am partial to those when it comes to two and three story 13R systems, there aren't a whole lot of systems I couldn't calc in 30 minutes to an hour at most. A straight tree or grid can be done in less then 15 minutes.

I am finishing up a rather large project that I had allotted 400 man hours to. 50 man days. In that 50 day period I might have spent two, three at most, doing calcs.

Guess I could never buy into the idea it was a time saver.

 

[Phoil]

Ohhhh-kayyyy. :-|

I wasn't asking for general critiques of software, I was asking for tips on how a non-engineer like myself could learn the jargon and techniques of this field.

Anyone else?

 

[TravisMack]

SD2:

I would say that those problems may be user error. No offense intended. I can calc a system in about 3 seconds with AutoSprink. It is also very easy and intuitive to draw with. HydraCAD is another fairly easy program to use. You will have an initial learning curve, but then you will more than double your production. I would say that with AutoSprink, I am 4x as fast as someone drawing on just AutoCAD and using HASS from start to finish of a project.

Phoil:

As far as terminology, you may want to try to get a copy of NFPA 13. That is the standard to which fire sprinkler systems are designed to in the US (for the most part). There is a section of definitions in the beginning that will give you some of the terminology.

As SD2 said, there is no real good "primer" on fire sprinkler system design. We have mostly all been taught from the trenches and it is all passed down from one guy to the next. Also, sprinkler layout is still an "art" to a certain extent. You will likely never get 2 identical answers from anyone on system layout, except for definitions of the terms.

If you have some specific questions, I am sure that many on here would be more than happy to assist you!



Travis Mack
MFP Design, LLC

 

[stookeyfpe]

Phoil:

You may want to consider this course from Oklahoma State University. It was a required course in my Junior year when I was obtaining my Bachelors of Science degree:

http://is.okstate.edu/pricelist2.aspx?id=7708

It will educate you in the basics of fire protection hydraulic calculations. It will not guide you in pipe layout, components, pipe erection or support or dealing with the protection of high challenge commodities or storage occupancies. OSU has other courses but you must be enrolled in the School's Fire Protection and Safety Engineering Technology program to take those courses.

If you want to learn the mechanics of automatic sprinkler design, you'll need to retain SD2, Travis and myself at a nominal fee. Realistically, it sounds as if your attempting to automate a design process that has already been automated by software available from different vendors that the industry is very comfortable in selecting and using. It's not just calculations - it's about economically designing a reliable fire protection water distribution connected to properly selected fire sprinklers.

Competition is good and you can hopefully produce the next great thing. Your employeer needs to be collaborating with folks such as SD2, Travis and myself to get a better understanding of the engineering involved. As Travis said, you need a copy of NFPA 13 (I would suggest the NFPA 13 handbook) and in conjunction with the mentioned course, sit down with paper, pencil, eraser, calculator and a straight edge and actually understand the mechanics of preparing a design. If you chose to do that, you should have SD2 looking over your shoulder because his experience is a rare and prized commodity.

Please don't take my comments as being arrogant. Rather, the design and construction of these systems is how many of us make a living and you posting such a broad question on a web-based message board is very hard to answer in short, concise answers.

 

[Phoil]

Maybe I should have been clearer. I'm not working on a new application, I'm one of the newer developers for AutoSPRINK.

To put it as simply as I can, I'm finding that the biggest obstacle to bug-fixing and especially implementing new features that come my way isn't the code, it's understanding the terminology and how the application is used. I've already learned a lot in the months I've been working here (and have had good success in tracking down bugs and adding new features), but I'd like to have a good resource for negotiating the "coder-to-user gap".

I appreciate the suggestions, guys; I'll get a copy of NFPA 13, and possibly look into some other materials. I did some of my graduate mathematics work at Oke State a few years ago, so thank you for that link, stookeyfpe.

It also appears there is some vast experience here, so I may drop by from time to time with my questions.

P.S. Feel free to ask any questions about my perspective as a coder (just don't send me bug reports and such, those should go through the proper channels).

Thanks, guys!

 

[SprinklerDesigner2]

Travis,

I use third party software that does much of the automation for me. It was good when it first came out but the firm folded up, there hasn't been updates in years and I hate throwing my $3,000 (early 1990 dollars) investment away.

I tried hydracad and didn't like it. Tried SprinkCad years ago. When two of their people showed up at my office to demonstrate only to have their software crash it didn't leave a good impression.

 

[Cidona]

Phoil. Good to hear you're working with Autosprink. That one was going to be hard to bypass. Are you going to to be incorporating Pararell computing into the program??? I would like the plotting, saving and stocklisting process to be executed quicker.
As far as getting to know termonogy in the industry, the NFSA has some very good classes. Their 2 week in Colorado will definitly give you lots to think about. If that is not practical they have shorter 'intro to sprinklers/plan review classes' as well as some on line seminars. Though personally I find the inperson classes a far superior experience to the online. Their website is

http://www.nfsa.org.

Apart from that start looking up everytime you go into a buliding (sicknesss within the industry). Start paying attention to how pipe is positioned in relation to the structure and other building features (for hanging the pipe, sprinkler location for clearance, situation where grid and tree systems are used, materials used ...ask lots of specific questions). Invite some fitters over for a BBQ.

 

[Phoil]

Cidona,

I don't know of any specific plans for implementing any parallel processing. I definitely agree that it could help for users with multi-core machines, but I don't know if it's considered a priority yet.

By the way, thank you for suggesting the NFSA classes here in Colorado; I'll certainly check that out.

 

[TravisMack]

Phoil:

I know that myself and Cidona are heavy AS users. I know that I would be happy to help in any way possible to get a grip on terminology and practical applications as well. AS is probably the best fire sprinkler design program out there, with HydraCAD being a close 2nd.

Again, feel free to ask any questions here.

SD2:

I only lease the AS program and hate paying that lease payment each year. However, I only do design and generate enough profit to pay for the program by the end of January each year. It is really that good! So, throwing away the $3k program may be the best thing you can do, unless you are already keeping up with your design load and hanging in there.



Travis Mack
MFP Design, LLC

 

[SprinklerDesigner2]

Travis,

I need to look at the AutoSprink because I am not keeping up.

Hard to compare designer output it's something gross dollar figures comes closest to.

I estimate most large projects, certainly the larger ones, sell and design. It it is anything to do with design, including conducting flow tests and field surveys, I'm the only one that can do it. Sometimes I get so behind the phone ringing scares me. I've got three jobs to do and they are all due next Tuesday sort of thing. What one do I do first?

Dollar figures I run through between $1.5 and and $2.0 million a year. So far this year I've done a 175,000 sq. ft. school, two hospital additions totaling 60,000 sq. ft., two four story motels with standpipes, 60,000 sq. ft. metal storage building and that is just the larger ones off the top of my head.

Something like hospitals take time especially when you get in intensive care units and operating room areas.

If I could be convinced my productivity would increase by 25% I'd be an idiot not to upgrade but I am not so sure. I'd hate to spend $30,000 only to endup where I am already.

5 years ago the rule of thumb was $1 million per year per designer. If a company did $30 million a year they'd have 30 designers some would be experienced NICET IV's while some would be new ones with 6 months experience. With increased material costs maybe that figure would be closer to $1.5 million a year today.

There's also the one sheet a week method. If you pack stuff in close generally speaking you can get one sheet a week done not counting plot plans. It's about what I do but I do pack a lot of stuff onto a single sheet. If I can fit two floors (1/8 scale) onto a single sheet I'll do it.

 

[Cidona]

Phoil.
Thank you for the info on the status of the parallel processing. Hope it will become a higher priority in the near future. I'm sure they have lots of other fun stuff to keep you busy.
I just was on the NFSA website and I noticed that they are no longer doing the 2 week class in Colorado. They have them in RI, TX and Chicago.
I hate recommending software to anyone. I can't say exactly what percentage I increased productivity since switching to AS. There are some improvments that I would very much like to see for it to realise it's full potential. Training and learning time are substantial (mostly due to the vast flexibilty and features within the program). It's expensive and you need an expensive computer. That being said, it smokes anything I've seen before. It brings with it a new era in sprinkler design. (As corny as that might sound).

 

[TravisMack]

SD2:

We used to base designers to the number of fitters. 1 designer should be able to keep 6 fitters / yr busy as a minimum. We used to figure that 1 fitter was about $250k / year, so 6 fitters would be 1.5M. That was in a union only market. I went out on my own several years ago. Prior to hiring someone, I was able to keep about 40-50 fitters going by myself. Granted, that was a lot of 70+ hour weeks, but that is the production level that AutoSprink was able to afford me to do.

So, give it a look again if you haven't in awhile. They have a lease plan where it is about $4k the first year and about $3k each additional year. I am not an AutoSprink salesman (I promise), but it is that good. You have my phone number from one of my emails (I think), feel free to give me a call about it some time if you like.


T

Travis Mack
MFP Design, LLC

 

[SprinklerDesigner2]

Travis,

6 fitters sounds about right.

70 fitters sounds nearly impossible even with 70 hour weeks.

 

[TravisMack]

It was easily 40-50. I have several small companies that I am the only designer for. If you add up all of their fitters, it comes out to about 35. Then, I figure the other larger companies that I design for accounts for another 10 or so. Alot of it is warehouse work and retail buildings. It is crazy, but possible. Give me a call sometime if you like.

Take care and have a great weekend!



Travis Mack
MFP Design, LLC

 

[SprinklerDesigner2]

Travis,

Enjoyed our conversation yesterday and thanks for the pdf files.

If AutoSprink does what it claims I can easily see a doubling or even tripling in productivity which makes it a no brainer as far as investment.

I can see the ability to keep 20 fitters busy which I think is a lot when you consider I estimate work, bid work, sell work, conduct my own flow tests and oftentimes conduct surveys of existing jobs for retrofits. 50% of my year is spent with tasks other then layout and if I could keep 20 fitters busy without having to hire a designer (non available) I would be ecstatic.

I spent all yesterday afternoon researching the product. It's really come a long way since I last looked. Real time calcs look real impressive and since I go for the exotic, bird cages along with loops fed by grids etc, I can see where I could save 10% of what time I do spend ln layout.

 

[TravisMack]

It was my pleasure. If you decide to get the product, I hope you enjoy it and find it as productive as I do.



Travis Mack
MFP Design, LLC

 

[Oremus]

Travis,

No offense but I've worked for a large union company who used Autosprink exclusively. I came aboard with nothing but Autocad, Hass, and my extensive library of custom Autocad LISPS that I have developed over the years. Hands down I could design, calculate, and list the entire job in less time than the designers using autosprink. The problem was with all the time it took for programming in 3d, entering elevations, outlets etc...

I've been considering releasing my cad routines as a package to show the speed you can maintain without breaking the bank. I've tried Autosprink & hydracad and I would prefer autosprink to hydracad - hydracad, althogh autocad based, is just way too clumsy - I think the programmers were learning coding as they developed this software and it shows. While autosprink is nice you still need autocad for general cad cleanup and I know some autosprink designers who will layout the sytem in cad then convert blocks to sprinklers and that saves them time as well.

To each their own. But I must warn you: Autocad w/ my lisp library are unconquerable! En Garde!

 

[TravisMack]

Oremus:

That is awesome. I have been an Autosprink user for about 4 yrs now and have seen production go through the roof. When you utilize things like the voice recognition and such, it goes pretty quick.

However, as you say, to each his own

But, I am seeing a greater movement toward 3D coordination which AS is great at. However, if you can draw 3D quickly in AutoCAD, more power to you I know that I can't.

Again, the point is to find the tools that work best for each person and utilize them to the greatest extent possible. I don't have a vested interest in Autosprink whatsoever, but I would put it's speed up against any other program or group of routines.

If there is ever a fire sprinkler system speed design competition, we may meet to see what is the best

Take care and have a great day!

Travis Mack
MFP Design, LLC

 

[Cidona]

That's a pretty bold statement Oremus. I'd be curiuos to see what you have. Please let us know here if you decide to do something with your package.
Also just to mention, I can't see how someone would be doing anything other than adding time by drawing in acad and doing convertions in autosprink. I do certain clean up operations on my drawings in ACAD (with some very quick tools for 'more typical' type buildings)(burst, purge, remove duplicate lines, etc...) But then I'm drawing in Autosprink and feel layout is just as quick as in ACAD with the added benifit that the elements are inteligent componets (take out/calcs/stocklisting) rather than just lines etc.
This comparison is not against you and your tools but rather against the users who are doing (by choice), layout in autocad only to convert to again to components in Autosprink. Maybe you can point out a benifit I'm missing.