Jump to content

PCM training


---
 Share

Recommended Posts

I'm going to Maple Grove next week for PCM training. Will I see any of you there?
Lots of pressure on me to absorb this as best I can. We have many, many programs
that require PCM (part families)
Link to comment
Share on other sites

Ask them why the (shallow) equality boolean return is behaving diffrently then inequality. I would love to here their explanation. And don't settle for a "don't know". You paid to get answers!

Don't forget to report back, I bet this will be priceless ๐Ÿ˜ƒ

Explanation of what Im saying:
(Highlight and compute one at a time. )
"one" == "one"
"one" <> "one"
Oh, this is a good one also. Force them to explain this:
test=1101000110100010010010011001010001001000100011001
message(test)
Link to comment
Share on other sites

I will be glad to ask that question. If you hadn't explained what
that actually meant in layman's terms, I wouldn't have known.

How exactly would this be used in a program?
Link to comment
Share on other sites

Good luck.
I took my PCM training probably 4 years ago and we basically spent a day writing a little bit of code to log the elapsed time that a program runs and saves (append) it to a text file. That was about a year before "measurement Duration" became a thing, so now i have all my programs fit with several lines of code in the presettings & post settings that could be replaced with a single command if i ever wanted to take the time to edit everything.
Then we spent 2 days on "parafiles" which can be very handy for part family programming because you can enter "Variables" into a text file that the program will read when you start a program, and adjust your nominals or tolerances or pretty much anything at all.
so you'll get a text box that says "what part are you running" and you click on "Part1 Part2 Part3" and depending on what part you click the program will read the parafile for that part and adjust your nominals and go.
I would say that PCM is probably the class that i got the most out of. I thought Basic & Advanced were very lacking in that i don't think anyone who's completely new to the Zeiss world can take those classes and be considered a programmer of anything more advanced than a washer. Curve and FF similarly left me with more questions than answers.

So you wont know what questions to ask until about a month after you're done with the class and your boss says "I need you to make 1 program that will measure every part we make", but then you'll just come to the forum and thats where youll get the most help.

B_lQeKFyKYYGhEiMc3-FmczG4HxqnDMBtiWYNVcaExg.jpg942539ddfc6becd73ed2e2801422ea0c.png

Link to comment
Share on other sites

I highly doubt the ordinary Calypso trainer has that much insight into PCM's mode of operation.
The best you can expect is them trying to ask the software department, but as far as I know this exclusive ivory tower is located in Germany... ๐Ÿ˜Ž
Link to comment
Share on other sites

I have to admit that I'm a little intimidated by this. There will be high expectations put on me when I get back. We have taken over a huge transfer job from one of our other plants. This involves many part families, so yes, we need one base program to be used for multiple sizes. Hopefully, my previous career didn't give me too much brain damage, from all the shot's to my head.
Link to comment
Share on other sites

"Hope the answers help clarify the questions"
// ---------- ---------- ---------- ---------- ---------- ---------- ---------- ----------
Value_1 = "one"
Value_2 = "one"

if Value_1 == Value_2
message("Condition 1", ": ","equal")
endif

Value_1 = "one"
Value_2 = "two"

if Value_1 <> Value_2
message("Condition 2", ": ","unequal")
endif

// ---------- ---------- ---------- ---------- ---------- ---------- ---------- ----------

"As String"
test = "1101000110100010010010011001010001001000100011001"
message(test)

"As decimal"
test = 1101000110100010010010011001010001001000100011001
message(test)

"PCM handles the parameter as decimal"
"Maximim decimal, value 1.1*10^16 internal"
test = 11010001101000100 //10010011001010001001000100011001
11010001101000099,4304 // output at nominal also wrong
"inside input"
test = 11010001101000
11010001101000,000 // OK
message(test)



"decimal overflow"
test = 110100011010001001 //0010011001010001001000100011001
message(test)

"I think CALYPSO is still 32 and WINDOWS ifself are 64"
"Windows biggest"
test = 11010001101000100100100110010100
"Windows crashes"
test = 110100011010001001001001100101000

"Lichtjahr: 1 Lj = 9โ€ฏ460โ€ฏ730โ€ฏ472โ€ฏ580โ€ฏ800 m โ‰ˆ 9,5ยท10^12 km = 9,5 Billionen km"
"all systems are limited to -macro and micro-"
Link to comment
Share on other sites

Dear Klaus, you sort of missed the point with the bloolean. The question is why there is no return value to the user in the GUI.
Not how equality/inequality works.

The second is indeed precision issues, and 1.1*10^16 is correct.

These questions are not complicated, It's just fun to ask stuff that the tutors don't get asked very often I think.
Make 'em sweat a bit ๐Ÿ™‚

Ask them why quad-precision is handled in the same way as double. Eg:
123q5
123d5
Link to comment
Share on other sites

 Share

×
×
  • Create New...