mirrored parts not acceptable 7

[duk748]

hello - our so called solidworks expert where i work has made it clear that we are not to mirror parts because they are too hard to detail & make changes to - i am a part time solidworks user & i am always getting in trouble w/ the so called solidworks police - if it is not acceptable to mirror parts then why does this command exist? - any experts here want to let me in on a secret to modifying & making drawings to mirrored parts - then maybe i can call myself an "expert" - really - any help or advice would be greatly appreciated - thank you

 

[ewh]

There shouldn't be many problems using mirrored parts if there are procedures in place and the discipline to follow them.
The benefits can be huge if used correctly, and the "police" would be much more effective if they helped develop such procedures, updating them as problems surface instead of creating short-sighted, expensive rules.
If you have thousands of LH parts on an aircraft, do you really think upper management would choose effectively modeling and detailing every part twice to document the RH parts, or establishing working procedures allowing the the effort to be nearly halved? I have created single details with opposite parts for years, and while it is true that they may be a bit more difficult to decipher on the floor, once the machinists and QA get used to it, it becomes second nature to them. And never forget, time = $, especially to management.

Technically, the glass is always full.

 

[ewh]

ShaggyPE's methodology is one that I am most familiar with, and it can (and does) work.

Some of the other sugestions here scare me; is pushing pulling lead on the drawing board next?

In some industries, everything has to be documented and controlled from production of raw material thru to the product reaching the customer. You can't let the floor change the product definition without full documentation. If they do, those resultant files still have to be documented and controlled in the same manner as the originals. By separating the parts to make them independent, you have essentially doubled the costs and efforts involved in any future modifications.
Planning and training are much more cost effective solutions.

Technically, the glass is always full.

 

[drawoh]

ewh,

There is no difference between mirroring and any other tabulated part or assembly. Tabulation saves modelling and drafting time, and it makes explicitly clear the similarities between a group of parts.

For example, I have drawings of a steel angle space frame in which each and every angle has the same hole pattern at the end. The fabricator can see that he can make one template or fixture to locate and punch the holes on every part. If each part had its own drawing, the template would not be a safe idea. In this case, manufacturing costs go up.

In-context models are what is dangerous in production. Done cleverly enough, your models will reconfigure your fabrication drawings depending on what assemblies you have open.

Your designers need to be able to decide when tabulation is good practise. At a lot of sites, management does not have this confidence.

At my site, tabulation is not allowed. We are extremely rule-driven. We want a process in which designers do not think. Tabulation looked complicated to our PDM[ ]group.

One of our designers worked around this. He created multiple configurations of his assembly model. He attached a separate assembly drawing to each configuration. No one noticed until I had to work on the top level assembly. What a mess!

If I am managing a tabulated production drawing generated in SolidWorks, I know there are multiple configurations to the model. The standard production design rule is to not change form, fit or function. If I intend to do such a change, I either must add a tabulation, or I must copy the drawing and model out and create a new drawing and part(s).

If you are manufacturing directly from the models, I don't see how you will recognize tabulation. In this scenario, the practise is dangerous. In mass production, design and drafting time is less important. In die-casting for example, similarities between parts probably cannot be exploited by production.

If you allow the village idiot to update your tabulated SolidWorks drawings, or any other SolidWorks drawings for that matter, you will get what you deserve.

JHG

 

[duk748]

hello again - in regards to our "police" being effective that has come & gone - when we decided to go to solid modeling the "policeman" & his pals sold the idea to management that our design time would be cut along w/ our errors in our product - all you have to do is push a button & it happens - our policeman was put in charge of setting up standards & writing a document w/ the procedures & rules to follow - that was of last november & still no document so we just hobble along like lemmings - except for me the trouble maker - i have tried to push to clean this mess up but i am left out in the cold & told i am trying to sabotage the program - i came from a company w/ tightly held standards & a manuel to go along w/ it along w/ 2 very good system administrators who kept reins on everyone in th cad dept. - i find it very difficult to work by flying by the seat of my pants - as before the problems are far greater then just solidworks & how to use it - thank you again maybe i am just getting tired! - have a great weekend

 

[CorBlimeyLimey]

Could you approach your previous company with a request to adopt their manual and standards?

 

[MikeHalloran]

I was the village idiot, hired with zero SW experience, for reasons that frankly elude me, to replace an engineering department that quit en masse. The surviving experienced SW users gave me what OJT they could spare in their copious free time, then were eliminated and replaced by me and a good mechanical designer who was still learning SW at about my level.

Thank goodness and Dassault for the tutorials and a super supportive local SW rep.

No thanks to the IT wizards who moved old (and multiply referenced) SW files among servers without warning, and Top Management who moved SW servers across town over protests. You haven't lived until you've loaded and edited a really big SW model over a T1 line.

Then the designer and I were eliminated and replaced by an EE who could sort of drive AutoCAD, and a fresh cheap MSME. At least they moved the SW servers back to the same net as the engineering workstations.

The results were pretty much what you'd expect...


Mike Halloran
Pembroke Pines, FL, USA

 

[ewh]

drawoh,
there is a difference; a mirrored part is exactly the same except opposite.
Tabulated drawings and configurations are related subjects. I was only attempting to address mirrored parts.

Technically, the glass is always full.

 

[drawoh]

ewh,

Mirroring or different lengths -- to tabulate or not to tabulate. It is the same problem.

If my sheet metal drawing shows right hand and indicates that left hand is opposite, the sheet metal guy makes one flat layout, punches two pieces and bends them opposite. The tabulated drawing shows an opportunity for the fabricator to save money and pass some of it back to us.

If I have a complex machined part shown as left hand and right hand, there might be an opportunity to flip the CAM program. There is a high probability of non-mirrored features creeping in. The tabulation is much less desirable, and way more likely to cause confusion.

If I have a drawing showing twenty angles of different lengths, all with the same hole pattern at the end, tabulation shows the fabricator an opportunity to work more efficiently. Again, it is a good policy. If the hole patterns at the end are different for each angle, tabulation creates an unreadable mess.

JHG

 

[ewh]

I respectfully disagree. I avoid tabulation whenever possible, but feel mirrored items are a special case, and as such are accepted by the major aerospace companies.

Technically, the glass is always full.

 

[ewh]

A little more clarification -
Tabulated parts do not always have changes to one which affect every other item tabulated (thus some of the documentation problems), as is the case with mirrored items. If the LH changes then by definition the RH changes. This can simplify design documentation significantly where it is taken advantage of.
It is not the same as tabulated parts.

Technically, the glass is always full.

 

[gwubs]

We use mirrored parts very successfully, because we INSIST that the mirror is ALWAYS the same except for the mirror aspect. We run code within our BOM software to confirm that the mass of mirror and non agree within .01%. If you make mirror and non mirror identical in ALL respects, except "hand" it is no problem.

I feel that there are no blanket, totally hard and fast ways to do things, but make sure that what you are doing actually works. Many fancy systems are set up without much thought about what can (and therefore will) go wrong. If you agree not to do things in SW that are error prone, and if you think everything through, you can decide how your group will do things. This must lead to setting policy, and enforcing it. If someone has a bright idea, then you look at it and decide whether is will actually work, and won't scuttle other things you already have in place.

At our company we are very creative in SW, we heavily use macros and code to speed up mundane tasks, our file structure is very refined, and we can open assemblies with 1000's of parts, years later without any errors. That said, our system is not perfect, and would not neccessarily be agood fit for others. It is stable, and reliable. We do not use PDM, we generate all our BOM's directly from the model in realtime, and we can do that because we have reasonable rules, and we enforce them.