printing problems

[112275]

It would seem that when i try to print full scale drawings with a hp 750 line printer that the width of the part is correct but the lenght is off by quite a bit. about 10mm on a line that should be 735mm long. Anyong have a guess as to why.

thanks in advance.

 

[SBaugh]

Check the printer settings and the page setup setting on your computer.

Regards,

Scott Baugh, CSWP

http://www.3dvisiontech.com

http://www.scottjbaugh.com

faq731-376

 

[112275]

Thanks for the help.

Changing the settings should do it? The reason I ask is because our shop is looking at buying a plotter. So for a test to see if it would suit our purpose i had a printing shop print a part for me. but i need the printed part to be quite accurate, is this possible with a large format printer?.

 

[uncleal52]

I assume your printing 100% and not "scale to fit".

 

[112275]

yes 100%

 

[SBaugh]

Many years ago I had to make overlays for pinions and gears for our Quality control dept. The overlays where used to measure the gears for accuracy. We would put the pinion or gear in one of those large scopes and blow up the teeth onto a screen. Then we would take the overlay and lay it on the screen and measure the gears that away. If the gears fell inside the overlay it was good part, if it was outside of the overlay it was bad.

The Overlay was made by me according to calculations that I was given. I made configurations of the part and showed them in a drawing, giving the outside and inside allowable varations. Including the nominal area.

The parts were very accurate but the problem came to printing them out on, and now old plotter (HP 350C). I made a 1 x 1 box in the upper right corner. I messed and messed with the printer settings and the page settings till I got the right accuracy. Once that box measured 1 x 1 then I knew I had it. It wasn't something that came easy, but that's really the only way to find it. Is by continuing to play with the settings. Maybe 100% isn't good enough try going up or down. But use something like a box, so it makes it easier to measure. Then step up once you get that far.

As in the movie "What about Bob?" ------> Baby steps

Regards,

Scott Baugh, CSWP

http://www.3dvisiontech.com

http://www.scottjbaugh.com

faq731-376

 

[112275]

thanks sbaugh. I wonder how close i can get on a 200" printout?. I am trying to transfer the part for the pc to paper, then from paper to aluminum. this should speed up our prototype hulls. at least till we get a cnc router.

 

[JNR]

Should be a set-up option to calibrate the plotter (ie. a plotter option). It is probably self contained within the plotter. As you say, the width is fine because that is contolled by a fixed encoder device (though often you can fine tune this too). The length is controlled by rotory encoders. The paper is driven through by rollers. Believe it or not even paper thickness and surface finish (friction!) will efect your plot scale along the long axis. Also the rollers wear and their diameter decreases (it don't take much!). BTW: You can buy replacement rollers. I am not familiar with the 750 model specifically, but it must have this calibration capability. Try looking in the plotter manual.

We have need to plot very accurate optical targets and other things from time to time, so we always go through a plotter calibration check first.

(Note that although the paper is "flat" going through the plotter, the upper surface speed/distance is dependent on the total effective radius to the paper surface - ie. the roller radius plus paper thickness.) The rollers are not very large, so even a few thousandths of an inch difference in thickness can make a significant change in the total plot length.

The usual process is it prints a cal plot that has two lines of a supposed length - one longwise, one crosswise. You measure them and enter back in the actual length. The plotter than adjusts and stores the new settings. It then plots a new cal plot and you ususally have to go around two or three times depending on how accurate you want it.


John Richards Sr. Mech. Engr.
Rockwell Collins Flight Dynamics

There are only 10 types of people in the world - those who understand binary and those who don't.