lotechman said:
The point I am making is that often weldments are built to ridiculous tolerances. This costs time and money. The designer sticks his chest out so proudly and boasts that the completed monster is plus or minus one millimeter but doesn't take into consideration the cost.
Sometimes yes and sometimes no. I will admit that sometimes cost is forgotten about. Just as often though it doesn't mean much in the overall sceem of things. Somtimes on the manufacturing floor yesterday is to late to get something done. I've seen parts made out of stainless because it was laying around when plastic would have doen the trick. In the end the cost is nothing compared to watching a prodcution line sit idle with people twiddling their thumbs.
A typical example is a local bridge that was upgraded with new deck sections. The main girders were large wide flange beams. The engineers required that the bottom flanges be located accurately to plus or minus one mm while they were thirty feet apart. This is illogical!
When the difficulty was pointed out the motivation for such a ridiculous tolerance was revealed. The man-carrying trolley that traveled under the bridge deck was hung from the lower flanges of the two girders. The genius who designed the trolley decided that his guide wheels were to be rigidly located thirty feet apart. There was no design allowance for movement of the bottom flanges. They had to redesign the wheel guides.
Atleast the problem was addressed in some manner. But I think you are over stating the case a bit, one millimeter over thirty feet isn't that extreme.
Many times people design with close tolerances simply because they can. I have seen pin bores that are 1/4 inch oversize in diameter to the pin yet the pin and the bore are toleranced plus or minus five thousandths of an inch! The designer obviously is in another world apart from reality.
There are a number of problems here. Number one of which is lazyness with CAD systems. Other times the engineer might siply want the hold drilled with a specfic drill bit and forgets to label as such. Soemtimes yes they are also in another world.
It is a waste of money to machine legs on a table if it is going to sit on a concrete floor.
This is where I absolutely disagree with you. The fact that the machine sits on concrete means absolutely nothing, that is why you have screws under the legs. The frames (there is always more than one in this sort of work) are to be arrainged and assembled as a unit The coupling may be one frame bolted to another or in some other way coupled. You would not end up with a usable production line if tolerance wheren't held on the frames. Say you have a production line that is 40 feet long made up of fifteen of so frames (machines) bolted together, how would that line function if each frame was held to a random tolerance. Imagine more so doing this sort of thing for several lines quickly.
The concrete floor is not a reference surface. In some instance a lazer is the refernce to which the machiens are setup to.
The floor is not going to be flat enough and you will have to grout or bolt in any case. Of course there is a designer out there who will specify that his table sit only on a concrete floor with a flatness of plus or minus one thousandth of an inch per foot span. Yes! He will stick his chest out and tell everyone how accurate the floor is and how accurate the table is.
I've actually worked in plants where they layed new foundations for rmachines because the old floor wasn't flat enough or strong enough or both. I remember one case where an engineer got all worked up because he thought he need to have is new concrete pad planned down.
In the special machines / production line business you get around this problem by building a line that doesn't depend on the floor.
I apologize for the rant but I see this all the time at work. The European system of engineer training is superior in this respect. They have to spend time on the floor to get their degree.
Well I take it as a rant but that doesn't bother me. What does bother me though is that you seem to be set on this position when there is good reason not to be. There are considerable advantages to this approach.
One is the components can be purchased or supplied via other elements in or out of the plant. Further if we are talking about making several frames the machining can happen at one time. A big one is that frames that go together square tend to weld up much nicer and require fewer adjustments at final demensioning.
That is production related, but then you need to consider the need for precision. This you seem to discount the most and is really the part of this discusion I'd like to get you to reconsider. If you let your tolerances flop all over the place you end up with huge issues at tying the rest of the machine together that relies on the frame. The frame becomes the reference for every thing else you do. Without a good refernece everthing else (conveyors, roobots, spindles, fluid power devices) becomes an effort to fit.
Now I'm not saying that everything needs to be held to 1 tenthousands of an inch on the frame. What I'm sayiing is that the use of a mill to demension and square parts gives you consitancy that you need for this sort of work. It is not something you can do on a band saw all day long. The concrete floor has nothing to do with the machining of the parts.
Thanks
Dave
