ben@geowhiz.UUCP (Ben Abernathy) (10/06/84)
[]
We are considering purchasing a Wintek circuit board layout system
and would like some information from other sites. The ad appears in the
October 4 issue of Electronic Design (page 216). This system costs $895
and is designed to run on an IBM PC.
Some of the questions we are wondering about are:
1) A real obvious one, can you really take the output of the Epson
printer or the Houston plotter straight to the photographer and
expect a decent PC quality negative ?
2) What is the minimun line width and packing density that a system
such as this could produce ?
3) Would the use of such a system *really* be easier and more productive
than taping PC artwork by hand ?
4) Is there any other such systems that we should be considering ?
We are basically a research facility and typically produce a board
quanity in the 1-10 range, ie our production runs are smaller than industry
prototype runs. Can anyone out there give us a hand ? Since there is not
a lot of traffic in this newsgroup, please post replies directly to
net.analog.
Ben Abernathy
UW - Madison
Geophysics and Polar Research Centerpadpowell@wateng.UUCP (PAD Powell) (10/08/84)
Message-ID: <135@geowhiz.UUCP> > We are considering purchasing a Wintek circuit board layout system > and would like some information from other sites. The ad appears in the > October 4 issue of Electronic Design (page 216). This system costs $895 > and is designed to run on an IBM PC. > Some of the questions we are wondering about are: > 1) A real obvious one, can you really take the output of the Epson > printer or the Houston plotter straight to the photographer and > expect a decent PC quality negative ? This is very dependent on the following items: 1. Make BLOODY DAMN SURE that your printer has a 1:1 aspect ratio across the entire page, and over a the size of the plot/print 2. Make <insert favourite epithet> sure that your plotter can plot straight lines at the plotter speed used. 3. Paper will do horrible things- use as good a paper as you can. If you can, use drafting mylar with a very light matte. Now the following problems. First, the resolution of the artwork. The usual standard for PCB artwork is pretty high. Cruddy artwork almost always results in poor boards. If you want 25 mil (.025 of an inch) wide lines with 12 mil spaces, you better be sure that you have that spacing on the artwork. The quality of the edges of the lines is very important: the sharper the contrast and straighter the edge, the better the resulting negative will be. By the way, the Institute for Printed Circuits has a VERY informative set of booklets available to its members, and I bet that they will send you some if you ask. If you are a Research or University, they will allow you to join as an associate, at a ridiculously low fee, and they even provide you all the technical stuff you want at a ridiculously low price (free, some of it). In addition, they have a list of people who specialize in making artwork, some of whom will come and TELL you how to do it, FREE (well, we had to give them lunch...). > 2) What is the minimun line width and packing density that a system > such as this could produce ? If you print at 4:1. you can get .025 lines with .025 spaces, but I would be very carefull, and look at all "narrow" places. > 3) Would the use of such a system *really* be easier and more productive > than taping PC artwork by hand ? If you are making small boards, say 4 inches by 6 inches, with moderate spacing, and smallnumber of components, they win hands down. Larger boards are a real pain. The big problem is displaying information and detail. I have tried a system that had a very nice multiple window, and allowed you to display several different parts of the board in the windows, allowing you to route and place things. This was a blessing, and I cannot understand how I lived without it before. If you are trying to "pack" things, or doing really tight and accurate work, I strongly suggest a system that will produce photoplotter driver output. This is a pain, but the artwork quality will pay for the extra headaches. > 4) Is there any other such systems that we should be considering ? I hate to name names, but Racal-Redac has a nice system, the PC-Cad is a nice toy, SCICARDS by Scientific Calculations is nice, Calma, Applicon, CompterVision, are the 3 Big Guns, and have nice systems. > We are basically a research facility and typically produce a board > quanity in the 1-10 range, ie our production runs are smaller than industry > prototype runs. Can anyone out there give us a hand ? Since there is not > a lot of traffic in this newsgroup, please post replies directly to > net.analog. > Ben Abernathy > UW - Madison > Geophysics and Polar Research Center Patrick ("A Boardlayout a Day keeps Poverty Away") Powell
cem@intelca.UUCP (Chuck McManis) (10/15/84)
I saw the Wintek Board layout program at NCC and was impressed, I
do not think it is better than say VersaTec's system but it is
probably sufficient for ones to tens. The print quality of the
PC layers (I saw it only working with two although the salesperson
indicated it could do up to 8) was ok on the Houston Plotter but tended
to smudge, the epson output was better, but would reccomend a new
ribbon for any plots on their way to a mask maker. I suspect it
would probably meet the needs of a small degign house that wasn't
designing boards bigger than Multibus I, or S-100 size. I would hate
to try to register a taped together mask layout that spanned three
or four page widths on the epson.
--Chuck
--
-- Chuck - - - D I S C L A I M E R - - -
{ihnp4,fortune}!dual\ All opinions expressed herein are my
{proper,idi}-> !intelca!cem own and not those of my employer, my
{ucbvax,hao}!hplabs/ friends, or my avocado plant. :-}
ARPAnet : "hplabs!intelca!cem"@Berkeley