Wann Tiefpass und Hochpass?

[[e1...]]

Hallo zusammen,

Ich vermesse zur Zeit einen Metallring mit einem Durchmesser von 340H7
Hier werte ich zusätzlich auch die Rundheit aus.
Nun meine Frage:
Wenn ich bei der Auswertung den Tiefpass auswähle, habe ich eine Rundheit von ca 0,1
Wenn ich den Hochpass auswähle, habe ich eine Rundheit von ca 0,01

Ich weiß aktuell nicht wann ich Tief oder Hochpass anwenden soll.
Im Zeiss Cookbook steht auch nichts drin.

Vielleicht könnte mir da jemand weiterhelfen

Danke im Voraus

[[To...]]

Low Pass for form.

[[Ky...]]

Hello Micro,

You'll have to forgive me for replying in English. I haven't taken any German in far longer than I care to admit.

What Tom said is absolutely correct, low pass is almost always what you want when measuring form and even location mostly. This covers most of what ZEISS machines measure, though there are a few exceptions. For instance, you may want a band pass if you are trying to do waviness. If you are doing some sort of roughness measurement, you'd want a high pass.

Mathematically what you are doing is blocking out either high frequency changes (in a low pass) or low frequency changes (in a high pass). Low frequency changes include things like deviations from being too oval or trilobed. Additionally, as a general rule, the amplitude of frequencies decrease as the frequency increases, so getting a better result with a high pass is exactly what I would expect.

Hopefully you find this useful.

[[e1...]]

Hello Kyle

It's no problem if you write to me in English. There is also a translator 😉
Then I'm doing everything right with the low-pass filter in a normal roundness evaluation.

I often work with the Zeiss Cookbook. But I haven't found anything about this.

Many thanks again

Greetings Mirco

Übersetzt mit DeepL (https://www.deepl.com/app/?utm_source=i ... ranslation)

[[e1...]]

Thanks Tom

[[Ky...]]

Please sign in to view this quote.

This does surprise me a bit. I know in the English version of the cookbook, the correct filter settings are mentioned in the probing strategies section in the area labelled as "C". I would think the German one would be similar, but I have not seen it. It doesn't go into the same detail as my last post, but that's a little outside the scope of the cookbook I think. I'm sure we can find a good book on Fourier Analysis if you want though, haha.

[[e1...]]

Thanks Kyle
I'll have another look through the cookbook. Maybe I have overlooked it.

Regards Mirco