All Activity

Hi Roy , try moving the first starting position of the calibration plate up by about 1 cm so that there’s still space to move at position 12, then it should work. A similar problem occurred for us when calibrating the GOM Scan 1, of course make sure to check if you have the correct adapter installed (different one for the Atos Q and a different one for the GOM Scan 1).

😎 The same problem on the other side of the Moon. 🤣

You could move the plate ... Which sensor and MV are we talking about?

Hi, I attached a clearer drawing. To get my 2D line, I created an other point(Ubis) close to pointU, and by point recall of U and Ubis i created my 2D line. I'm looking to measure the distance between U and U1 according to the drawing. The dimensions of the part are approx. 2/3 cm on each axis and the thickness 2mm. I hope that's clearer now

Do you have at least drawing? I don't understand how you got this 2d line. Did you use recall or recall measured points? I think that using such sketch is not much helpfull, because everyone from us are looking onto problems differently. Just to summarize: tolerances, problems with definition for CNC scan, varying results in what range, drawing vs used characteristic. There can be many right choices with very different approaches. Also sometimes scale of real part vs magnified drawing can be misleading.

Hi Martin, Thank you for your advice. I set my point with a plane point but i don't see any improvement in my results they are not stables. I noted that there's 3 different option to determine the normal: - Probing direction - Plane measurement always - Plane measurement in manual I didn't find in the manual any explanation on this matter, but they affect a lot my results and the stability. When i choose ''plane measurement always'' because my faces are little, the probe miss the surface and then the normal vector is affected and of course my results. Also, when i create my line ( i attached a figure), the line is not connected to the 2 points. Is it a graphical issue or a problem in my construction? Thank you for your help. IMG_1670.HEIC IMG_1668.HEIC

From ISO 1101:2017 Extracted integral feature: the real, tangible, non-perfect feature as probed by the CMM. Essentially the probed point collection belonging to that feature (a hole, a plane, etc.) Extracted derived feature: the feature computed from one ore more integral features. Examples are real (non-perfect, i.e. curved) median lines (centerlines) and median surfaces (like the median plane of a slot). Optionally, it is possible to specify that the associated (fitted) feature, for instance a Gaussian fitted plane or the perfect centerline of a Gaussian fitted cylinder shall lie inside the tolerance zone. If the Gaussian fitted element (feature) must lie inside the the tolerance zone, rather than the extracted element, this is indicated by adding (G) in the second compartment of the tolerance indicator. See picture below. In many cases, getting the Gaussian associated plane inside a tolerance zone is way easier than the extracted surface itself. So, answering your question, no, the Gaussian (least‑squares) fitting is not the default method for the the feature to be measured. By default, the extracted (real, probed, or calculated from probed points) feature must be measured.

We're using old Visual IPI, I believe it's called differently now as they got bought out. We're paying 5k a year for outdated software so it's got to go. I look for something opposite - as simple as possible - easy to use by someone who doesn't understand drawings very well. I've had DISCUSS, guthrie and 1factory on trial. I'm going to take a look at HighQA and Capvidia. Solidworks inspection looks nice, but I don't think they'll be at IMTS and they don't offer trials.

I'd love to find one that didn't cost an arm and a leg, and had a perpetual license. I got fed up with all of the big names, and spent time developing our own. It was very rough around the edges at first, but after a lot of work we have something that works well for us. Depending on your needs, I might suggest going that route for long term use.

Can you explain how this works? Its peeked my interest.

I had a very similar design "in my head" that incorporated 6 grooves (just extend the current 3 grooves to the opposite side of center) so plate could be rotated 180° for convenience. My design also had the 4 sides ground so block could be turned onto its' sides, also for convenient access to the 3 screws on the back. With ground sides, the block could be held a sin vise or plate, or the adjustable v-block that Richard posted.

Not selling our RDS Contura would have saved a lot of time and money, too. lol

@Jeff Our lab holds the temp at ±0.5° 24/7, so for something like this job, it's not such a big deal. Yes, I know handling causes some expansion but that is minimal. When we're trying to hold microns, the CFX extensions come out. Does Zeiss make CFX adapter plates?

We have a fix assist. It was worth the money. The only issue that we ran into is that the angles were slightly off - especially noticeable at really long distances, so we created new references mark to account for this.

@Roy Espinoza Is this a GOM Scan 1 or an ATOS Q? I ran into issues with the last GOM Scan 1 that I was using where the arm ran out of room to move closer. There was nothing I could do, so I ended up calibrating with the tripod. I believe there is a different mounting bracket for the GOM Scan 1 to the ATOSQ, so make sure you’re using the correct one.

@Keith Forche Thanks. This worked very well for setting the rotation angles however I had to resort to a Starrett Protractor to eyeball the angles of the knuckle joints. Was quite surprised with the results. After qualifying stylus systems and measuring cylinders with shafts, all the angles were within ±0.5°. 🙂

@Martin Jánský The confusion was that I was looking for an answer to my question about my base alignments and I didn't believe you addressed it. I mentioned I had measured the 2 other pins to see if they were 120° apart. Since they were not, it caused me to question the validity of using 1 pin for the planar as I didn't know if it was in the right location. Obviously, the sensor uses all 3 pins to create a repeatable connection. Are the 3 mating connection points on the sensor square with the machine? Seems like I am making a lot of assumptions when I create a base alignment from the adapter plate. I was trying to set and verify the angles using the cmm. I don't want to set an angle, qualify the stylus, measure a cylinder from the shaft on a sphere, to find out what the angle is. Then, repeating it until I get the desired results. That seems like a long set of steps.

This is a template I use for the VAST XTR sensor; it can easily be used for any other sensor. I have another for RDS CAA sensor as well, nothing much different. It uses basic Calypso settings and many other custom settings for file output. Use as you see fit. The "5x" is number of decimal places. Eaton_Inch 5x_Template_v2022_04-17-2024.zip

What are you using now? We have Discus, we may need something more robust that can cover a lot more with planning, check lists, etc.

I'll be there. Got to find a new ballooning software and browse for new stuff.

Nowhere, it's a really bad joke. Took me a while to find a good Greek name.

And I just found where the reference is, buried in some PCM that calculates a bolt circle from all related cylinders determined by bore count. See the highlighted code in the image. 🫣

I am running into this error, and I can't find the reference being referenced. Depending on the part configuration it can have from 3 cylinders (with 3 scans per cylinder) to 13 cylinders (with 3 scans per cylinder). "Circle1_Groove OD" has no visible or searchable reference to any of the "BASE_CIR_xxx" features. Same with "Circle1_Groove ID" (no error) and ""Circle1_Groove OD" was created as a copy of ""Circle1_Groove ID". Parameters shown in the reference image are numeric values from an Excel file for the Para side, there are no references to any of the "BASE_CIR_xxx" features or related characteristics. It's all PCM and Para file controlled. I do not believe the PCM controls the number of cylinders to scan for my tertiary datum, which is "BASE_CIR_DATUM -C- 2" used in the Base Alignment. All features shown in the "Base Cir features" image are not used for this mini plan. The PCM only controls where the program reads code based on the number of cylinders. The last part of code is well after the first section of code. Any assistance is appreciated. if (num_bores==9) and (miniselection=="Rough Bore Characteristics") if (tongue_inspect==1) then setRunID("9 Broach Rough Bore Characteristics") //New 12.12.23 inspPath = getActualInspectionDir()+"\" fileName = "setRunID.txt" deleteFile(inspPath+fileName) addToFile(inspPath+fileName, getRunID()) //End New else setRunID("9 Rough Bore Characteristics") //New 12.12.23 inspPath = getActualInspectionDir()+"\" fileName = "setRunID.txt" deleteFile(inspPath+fileName) addToFile(inspPath+fileName, getRunID()) //End New endif endif Except the I have this code that should bypass the above (for any number of cylinders). //New 08.27.26 Back Face Groove diameters if PN=="00-363107" miniselection=="00-363107 Grooves" setRunID("00-363107 Grooves") Machine="Groove Characteristics" //New 12.12.23 inspPath = getActualInspectionDir()+"\" fileName = "setRunID.txt" deleteFile(inspPath+fileName) addToFile(inspPath+fileName, getRunID()) spd_grv="0.29650" endif

Invest in one of these, I can set any angle quickly and easily. ZX, ZY, XY or any compound angle you can imagine. You may also need a small set of gage blocks.

In Calypso 2025 you can still apply it using Loose but there’s a reason they got rid of it.