Scan data anomaly: Focus 330 out of calibration?

Faro S150, S350, X30, X130, X130HDR, X330, X330HDR, M70, Leica RTC360, HDS6*00, HDS7000, Z+F IMAGER 5016, 5010X, 5010C, 5006H, 5006EX & PROFILER 9012 etc
Post Reply
User avatar
sreed
V.I.P Member
V.I.P Member
Posts: 192
Joined: Mon Apr 28, 2014 4:31 pm
Full Name: Scott Reed
Company Details: Prologue Systems
Company Position Title: Principal
Country: USA
Skype Name: sabotva
Linkedin Profile: Yes
Location: Richmond, VA USA
Has thanked: 1 time

Scan data anomaly: Focus 330 out of calibration?

Post by sreed » Tue Aug 11, 2015 4:50 pm

Hey guys,

Any chance you have experienced what we have just observed from our scan data? In a recent project we noticed that the "overlap" between the beginning and end of the scan collection is mismatched. I don't recall ever seeing this overlap before, and we have measured it at 3/4" at about 75' distance, all in the verical. See attached example screen shot.

After discovering this on a recent project, we went back to investigate older scans and have seen the same phenomenon ever since we received the scanner from its first annual calibration. Coincidence? Is this overlap expected? If so, it should be perfectly aligned, correct? This could explain some registration problems we've experienced lately too. If targets are randomly scattered through the scan sweep, some may vary in z-location by up to 3/4" or more at far distances. X and Y does not appear to be affected.

I have a call in to Faro, but thought I'd push it out to the community for any advice.

Thanks,
-Scott
You do not have the required permissions to view the files attached to this post.

tbwester
V.I.P Member
V.I.P Member
Posts: 886
Joined: Fri Apr 20, 2012 1:49 pm
Full Name: Thad Wester
Company Details: WeWork
Company Position Title: Head of Reality Capture
Country: USA
Linkedin Profile: Yes
Location: SC/NYC

Re: Scan data anomaly: Focus 330 out of calibration?

Post by tbwester » Tue Aug 11, 2015 5:30 pm

I don't have a Faro, but we had a similar issue on another scanner. IMO the data is not usable, I noticed the error when our registration kept throwing large errors on a small loop closure.

Hopefully they can get it fixed for you, but it will probably have to go back in?


One tip. since it happened, and I had to re-scan that entire project, I've been a little crazy about checking the scanner to make sure the scan overlap has good closure. An easy way to check is to scan, and then color the scan by elevation at .005' increments, then you will clearly see if the overlap is off in the vertical.

chris_e
I have made 30-40 posts
I have made 30-40 posts
Posts: 34
Joined: Thu Aug 22, 2013 4:44 pm
Full Name: Chris El-Araj
Company Details: Underhill Geomatics Ltd
Company Position Title: Director
Country: Canada
Linkedin Profile: Yes
Has thanked: 1 time

Re: Scan data anomaly: Focus 330 out of calibration?

Post by chris_e » Tue Aug 11, 2015 5:41 pm

We do exactly as Thad outlined. We scan the floor in our warehouse and colourize at a 2mm interval. See attached image from our x330. The overlap line is clearly visible where the scanner starts and ends its 180 turn. This equates to 8mm over 15m. This unit is 5 months out of calibration, and has to go back. We, like Thad, had to return to a site and redo the work last year, as there was no way to salvage the data in the project timeline.

My question is there a reason why it goes out of calibration so fast, and is there a remedy other than sending it back to Faro for another calibration?

We check before and after every major project.

Chris
http://www.underhill.ca
You do not have the required permissions to view the files attached to this post.
C. R. El-Araj, BCLS,

www.underhill.ca

jedfrechette
V.I.P Member
V.I.P Member
Posts: 846
Joined: Mon Jan 04, 2010 7:51 pm
Full Name: Jed Frechette
Company Details: Lidar Guys
Company Position Title: CEO and Lidar Supervisor
Country: USA
Linkedin Profile: Yes
Location: Albuquerque, NM
Contact:

Re: Scan data anomaly: Focus 330 out of calibration?

Post by jedfrechette » Tue Aug 11, 2015 5:43 pm

I just got hit with this for the 3rd time in the last year. Once with one of our scanners that was out by a couple inches on the first scan after coming back from calibration by Faro and the other two times with contractors whose scanners were out. Supposedly in some cases Faro has been able to provide a calibration file that fixes the miscalibrated scans but I've never been that lucky.

It seems like this type of miscalibration should be pretty easy for the software to automatically detect, just compare the distance between the overlapping scan halves and warn the user if they are outside of a tolerance. The fact that it doesn't and users can go for weeks or even months collecting bad data because their visual inspection didn't find slight miscalibrations is pretty bad.

After the latest incident I've pretty much decided that we are going to have to develop our own automated validation tool to start checking for these types of problems. We're just losing to much money to these issues.
Jed

tbwester
V.I.P Member
V.I.P Member
Posts: 886
Joined: Fri Apr 20, 2012 1:49 pm
Full Name: Thad Wester
Company Details: WeWork
Company Position Title: Head of Reality Capture
Country: USA
Linkedin Profile: Yes
Location: SC/NYC

Re: Scan data anomaly: Focus 330 out of calibration?

Post by tbwester » Tue Aug 11, 2015 6:26 pm

Interesting comments. Definitely scary to not be able to trust the data. Re-scanning a site is a pain, but there are many projects where you only get one shot at it before they close an area, or for a variety of other reasons.

This wasn't with a faro, but I found the issue was completely resolved by turning off the compensator/inclinometer. Not ideal, but can be used in a pinch.

User avatar
MarkL
V.I.P Member
V.I.P Member
Posts: 129
Joined: Thu Sep 29, 2011 7:57 am
Full Name: Mark Lupton
Company Details: Wolverton
Company Position Title: LASER Scanning
Country: USA
Skype Name: [email protected]
Linkedin Profile: Yes

Re: Scan data anomaly: Focus 330 out of calibration?

Post by MarkL » Tue Aug 11, 2015 6:36 pm

I had this problem with a focus several years ago. It was fairly soon after getting the scanner, brand new, so I didn't think to check the overlap. I had a really hard time registering some data and spent a lot of time looking at it trying to figure it out. Eventually I found the overlap issue and have always been wary of using Faro scanners since. I think this is something Faro really need to get a handle on. Yes, they sell a cheaper product, but if you can't trust that your data will come out well, then I'd have a very hard time recommending it to anyone.

User avatar
sreed
V.I.P Member
V.I.P Member
Posts: 192
Joined: Mon Apr 28, 2014 4:31 pm
Full Name: Scott Reed
Company Details: Prologue Systems
Company Position Title: Principal
Country: USA
Skype Name: sabotva
Linkedin Profile: Yes
Location: Richmond, VA USA
Has thanked: 1 time

Re: Scan data anomaly: Focus 330 out of calibration?

Post by sreed » Tue Aug 11, 2015 6:44 pm

Great feedback! I've just requested the RMA from Faro for recalibration. Ughhh.

Maybe this is a question I shouldn't have to ask, but what is the easiest way to colorize by elevation without having to go through VloudCompare? (Scene?)

jedfrechette
V.I.P Member
V.I.P Member
Posts: 846
Joined: Mon Jan 04, 2010 7:51 pm
Full Name: Jed Frechette
Company Details: Lidar Guys
Company Position Title: CEO and Lidar Supervisor
Country: USA
Linkedin Profile: Yes
Location: Albuquerque, NM
Contact:

Re: Scan data anomaly: Focus 330 out of calibration?

Post by jedfrechette » Tue Aug 11, 2015 7:35 pm

Another issue we just had with our 330 is that intermittently it would start loosing all data in a ~10 degree arc where the scan halves should have been overlapping. The rest of the data seemed to be completely fine and would register perfectly, it was just missing that slice. Once it started happening it would affect every scan until the system was rebooted, after which scans would be fine again. The scanner didn't indicate that there was any problem while this this was occurring. Scene also didn't complain that any thing was wrong with the affected scans. As a result it wouldn't be detected until back in the office visually inspecting the data. On projects with hundreds of scans each scan doesn't get looked at until it has already been exported from Scene and loaded in other programs, as a result we ended up loosing not only the field time but also a significant amount of office time that wouldn't have been wasted had we known the data was suspect when we initially imported it in to Scene.

This is another situation where a simple automated sanity check comparing the overlap between the two halves (none in this case) would have picked up the problem. Every mechanical system is going to fail at some point. Overall, I've been happy with Faro's reliability and I don't have a problem with the fact that they go out of calibration. What I do have a problem with is that users aren't getting enough feedback so that they know when the system does start failing.

The worst thing about this particular problem is its intermittent nature. Even if you were doing test scans every morning before leaving the office chances are they wouldn't be affected so you could still go out and collect a bunch of broken data before realizing there was a problem.
Jed

User avatar
Phill
V.I.P Member
V.I.P Member
Posts: 593
Joined: Wed Feb 06, 2008 7:19 am
Full Name: Phillip Nixon
Company Details: SKM
Company Position Title: Surveyor
Country: Australia
Location: Sydney

Re: Scan data anomaly: Focus 330 out of calibration?

Post by Phill » Tue Aug 11, 2015 11:57 pm

Scott

We had this on a different brand of scanner, and there was no chance to revisit. The problem, if it is similar to ours, lies in the vertcial circle not being exactly perpendicular to the horizontal circle (also known as vertical collimation error). In modern total stations and some scanners such as the C10 & P20 this value can be calculated and applied by doing an in-field calibration (so I am led believe, havent needed to do one yet), but in the particular brand of scanner we had it was not possible, so I had to find a way to correct the data.

If you can calculate the value in seconds to correct the scans (by using the distance between the two scans at a reasonable distance from the scanner, and their distance from the scan head and an arctan and then dividing by 2) you could use the attached ptx correcter to possbily fix the error. Based on your numbers there (keep in mind I dont work in inches too much) I've got 180 seconds with a correction of 90 seconds. Wether it is -90 or +90 you will have to do a trial and error.

I am assuming Scene can export a ptx.

Cheers

Phill

edit: 21.12.2015 - new version of vertical circle corrector added (V1). The older version used a slightly "dodgy" method, the new version is more accurate but slower.
You do not have the required permissions to view the files attached to this post.
Last edited by Phill on Mon Dec 21, 2015 6:26 am, edited 1 time in total.

User avatar
Dedken
V.I.P Member
V.I.P Member
Posts: 363
Joined: Fri Mar 15, 2013 10:28 am
Full Name: Kenneth Bazley
Company Details: Sir Robert McAlpine
Company Position Title: Senior Geospatial Engineer for HDS
Country: UK
Linkedin Profile: Yes
Location: London

Re: Scan data anomaly: Focus 330 out of calibration?

Post by Dedken » Wed Aug 12, 2015 11:54 am

Been through all of this as well, but at least I now know what to look out for and how to check for it. For whatever reason, FARO is not dealing with this issue and I guess as long they continue to sell scanners it won't be a concern to them. If you send them your data they can send you a compensation file which might work if you're lucky. Last time it happened to us the compensation file appeared to correct the collimation error in the registration results but a physical check of the scans still revealed significant errors between some corresponding targets. The data was, in the end, unusable.

There are two ways to reduce the possibility of your Focus-type scanner ending up with this kind of problem which you can implement painlessly.

1) Phase-based scanners are fragile and prone to being knocked out of calibration in transit. Eliminate the possibility of accidental mistreatment by sending the scanner to site in a taxi with a responsible person who will (hopefully) not drop it or knock it around.

2) I would strongly suggest you check your scanner before you send it out on a job, and there is a simple way to do this.

1) Do one single scan on a level(ish) surface at a res that will give you decent range on a horizontal surface - ie 1/2 at 1x quality.
2) In Scene open the scan and highlight the overlap area (FARO engineers refer to this as the 'trigger point') with the lasso tool and right-click > 3D view.
3) In 3D View window select your point size as appropriate. Select rotation point and rotate the view towards the vertical. You will be able to see any divergence between overlapping circles and measure it. The tolerance value is +/-2mm at 20m from scanner so I was told. At its worst we had +/-16mm at this range!

If you have never checked your scanner in this way before, do yourself a favour and do it ASAP, you might save your company £$£$£$ and :oops:
All views are my own and are not representative of my employer, The Queen, God or anyone else for that matter.

"we need an instrument, to take a measurement" - I.MacKaye 1992

Post Reply

Return to “Phase Based Scanners”