DUSI - Windows Connection and Calibration Problems

Techniques, equipment and issues. Also visit the NSS Survey & Cartography Section.

Moderator: Moderators

DUSI - Windows Connection and Calibration Problems

Postby Extremeophile » Feb 22, 2010 10:56 pm

I recently purchased the DUSI v2 and have not been able to get it to communicate with the Data Viewer application in Windows XP. When I first plug the device into a USB port it prompts me to install drivers. When I manually choose the OpenMag.inf file it gives a message saying that the driver doesn't contain information about this piece of hardware (or something similar). I let Windows decide and it installs it as a "USB Serial Converter". It then tries to install a second device, which I again manually direct to the OpenMag.inf file and this shows up under port devices as "%DESCRIPTION% (COM3)" with a yellow exclamation point. The Data Viewer application tries to autodetect but says that "opening device failed". If I let Windows choose the driver for this second device it chooses "USB Serial Port (COM3)", and again the Data Viewer application does not work.I'm hoping others that own the DUSI have had some experience with this and can guide me.

I've also had some trouble getting the mag sensor to calibrate well and was hoping to adjust some of the settings using the Data Viewer software before trying again. I've probably re-discovered what many already know - that there is too much magnetic interference indoors to get a good calibration. There's also too much interference in my backyard and frontyard. There's apparently too much interference miles from civilization when placed on a schist boulder (maybe too many ferromagnesian minerals?). I think having it too close to the ground presents problems, though I'm not sure why. I'm getting FS/BS vertical angle agreement to 0.1 degrees (outstanding), but with some calibrations my FS/BS mag error can be >5 degrees (awful).

Any suggestions on the PC connection or hints / tips on calibration of the magnetometer would be much appreciated.

Derek
User avatar
Extremeophile
NSS Hall Of Fame Poster
 
Posts: 761
Joined: Dec 7, 2009 7:37 pm
Location: Littleton, CO
Name: Derek Bristol
NSS #: 34941
Primary Grotto Affiliation: Colorado Grotto
  

Re: DUSI - Windows Connection and Calibration Problems

Postby Phil Winkler » Feb 23, 2010 12:49 pm

Derek, just a guess, but many airports have an area especially designed for calibrating compasses. Well away from anything, usually. As for the Windows thing...sigh. It ain't fun, I know.
Phil Winkler
13627 FE
User avatar
Phil Winkler
Global Moderator
 
Posts: 2375
Joined: Sep 5, 2005 8:48 am
Location: Wilmington, DE and Dewey Beach
NSS #: 13627FE
  

Re: DUSI - Windows Connection and Calibration Problems

Postby Bob Thrun » Feb 23, 2010 2:53 pm

Does the DUSI have to be calibrated in the same manner as the Shetland Attack Pony? www.incavedigitalsurvey.com/lynnbrucker ... heSAP1.pdf
Bob Thrun
NSS Hall Of Fame Poster
 
Posts: 322
Joined: Jul 18, 2006 12:50 pm
  

Re: DUSI - Windows Connection and Calibration Problems

Postby Crockett » Feb 23, 2010 4:55 pm

You may have the tolerance set too high. Try the setting for simply "good" versus "excellent". The DUSI is like driving a high powered sports car. It is way too sensitive in default so you have to tune it down in the settings menus. I don't have one in my hand right now so I can't be specific.

Here is some advice from Sean D'Epagnier gleaned from an e-mail:

"It is not a good idea to only take points in a circle. The best
calibration would be to take points in circles (rotate through
azimuth) of various pitch (incline), but also rotating (rolling)
around the laser the whole time. I know this is confusing to
describe, but the idea is to get the points well spread in the unit
circle. I did something like 3 circles of 0, 45 and -45 incline, with
8 azimuths each, and 4 rolls per azimuth. Then also do straight up
and down."

and this:

"As far as the inaccurate measurements, first try to figure out why
pitch is off. I say this because it depends on less unknowns and when
resolved the azimuth errors may be gone too.

I suspect possibly the laser alignment might not be calibrated well.
This can be verified.. try taking some shots rotated around the laser,
and see how much the incline changes. If it changes more than .2
degrees, then the laser calibration is likely bad (considering the
accelerometer is set to excellent)."

The DUSI is indeed sensitive. Sean is really casual in his videos so he makes remarks like it is offhand and not important...but it is.

I don't have much to add about the desktop software. I loaded it but found no compelling use for it as you have.
Crockett
NSS Hall Of Fame Poster
 
Posts: 243
Joined: Feb 5, 2006 7:10 pm
Location: Pineville, Kentucky
Name: Mike Crockett
NSS #: 33730
Primary Grotto Affiliation: East Tennessee Grotto
  

Re: DUSI - Windows Connection and Calibration Problems

Postby driggs » Feb 23, 2010 9:32 pm

derekbristol wrote:When I manually choose the OpenMag.inf file it gives a message saying that the driver doesn't contain information about this piece of hardware (or something similar). I let Windows decide and it installs it as a "USB Serial Converter". It then tries to install a second device, which I again manually direct to the OpenMag.inf file and this shows up under port devices as "%DESCRIPTION% (COM3)" with a yellow exclamation point. The Data Viewer application tries to autodetect but says that "opening device failed". If I let Windows choose the driver for this second device it chooses "USB Serial Port (COM3)", and again the Data Viewer application does not work.


I don't have experience with configuring the DUSI to communicate with a computer, nor do I have much MS Windows experience, but...

The fact that your Device Manager lists "%DESCRIPTION% (COM3)" tells me that the .inf file and/or its associated driver files are corrupt or misconfigured. A concrete value should be filling in the placeholder string token "%DESCRIPTION%" in the .inf file; if that variable is missing, who knows what other critical configuration data is missing.

I would not go any further without completely uninstalling the driver(s), downloading fresh known-good drivers again, and re-installing them.
User avatar
driggs
NSS Hall Of Fame Poster
 
Posts: 495
Joined: Sep 12, 2005 9:40 pm
Location: State of Jefferson
Name: David A. Riggs
NSS #: 56189
Primary Grotto Affiliation: Monongahela
  

Re: DUSI - Windows Connection and Calibration Problems

Postby Extremeophile » Feb 23, 2010 11:09 pm

Does the DUSI have to be calibrated in the same manner as the Shetland Attack Pony?

The short answer is "no". The calibration routine is done without external references and only takes about two minutes to complete. My accelerometer calibrations always come back as "excellent". My magnetometer calibrations come back as "good" at best; however, these calibrations still have unacceptable FS/BS agreement. I was planning to set up a calibration course after getting a good calibration to validate the measurements I'd be getting against my Suunto and Sisteco.

You may have the tolerance set too high. Try the setting for simply "good" versus "excellent".

I'm not sure there's a way to set the tolerance with v2. After I complete the calibration I can check to see if it achieved excellent or good. I can set the magnetic range and I've tried setting this higher, which I think makes it less sensitive to magnetic interference at the cost of accuracy or precision. I didn't notice a difference when setting it one notch higher. There's one more notch to try. I also think there's a frequency setting that can buffer out magnetic noise (probably not the right technical description), but I can't seem to set this with the embedded software. I think this can be set with the external software, but that's not working. I believe the laser alignment is calibrated very well. I have a very low error value and get good vertical angle repeatability at all orientations around the laser axis. I've tried re-calibrating the laser several times and always get the same values (3.3 degrees on the accelerometer and 3.3 degrees on the magnetometer).

I would not go any further without completely uninstalling the driver(s), downloading fresh known-good drivers again, and re-installing them.

I'll try to contact Sean for a different driver file. He has some comments on his website about Windows being broken. Hard to believe anyone could find fault with M$.
User avatar
Extremeophile
NSS Hall Of Fame Poster
 
Posts: 761
Joined: Dec 7, 2009 7:37 pm
Location: Littleton, CO
Name: Derek Bristol
NSS #: 34941
Primary Grotto Affiliation: Colorado Grotto
  

Re: DUSI - Windows Connection and Calibration Problems

Postby Extremeophile » Feb 23, 2010 11:38 pm

I've been doing some work to make the DUSI more cave ready and thought I'd share my practices. I abondoned the rubber boot because it's heavy, bulky, not water or dust resistant and doesn't fully protect the LCD or keypad. I've found that the Pelican 1020 fits it perfectly.

Image
I've used Marc Ohms method for protecting my other instruments: www.caves.org/section/sacs/rr/suuntos.pdf
Below are a Suunto KB-20, Sisteco Surveymaster and Suunto Tandem all protected with a polycarbonate cover over the viewing window and several coats of Plastidip over the rest of the case.

Image
I used the same 0.1" polycarbonate to epoxy a protective window over the DUSI LCD and then coated the bottom, back and sides using a couple coats of Plastidip. This makes it a little more rugged when it's not in the Pelican case and gives it a little more splash and dust protection during a survey.

Image
I also epoxied a small Lego brick to the bottom of the case, where I visually aligned the LASER across the top and side - i.e. the point where the LASER axis intersects the bottom. This small nub gives me something to set on a survey station to know the LASER is in-line with the survey shot. This point also makes it much easier to calibrate the LASER alignment.

Image

Image

Derek
User avatar
Extremeophile
NSS Hall Of Fame Poster
 
Posts: 761
Joined: Dec 7, 2009 7:37 pm
Location: Littleton, CO
Name: Derek Bristol
NSS #: 34941
Primary Grotto Affiliation: Colorado Grotto
  

Re: DUSI - Windows Connection and Calibration Problems

Postby LWB » Feb 24, 2010 1:52 am

Nice job on the DUSI mods, Derek.

I use the Disto X calibration points for calibrating the DUSI. Mostly because that spaces them evenly - the DUSI calibration process can be simpler then the DistoX - but I was trying to be as consistent and repeatable as possible. The latest Pony software doesn't require "truth" data either - although Phil says it will be more accurate if calibrated with "truth" data. The DUSI calibration process is the easiest of the three.

I suspect there is a bug in the DUSI v2 software. I have never managed to get my v2 to perform as well as the v1 with the prototype software. I thought it could be related to poor laser alignment, so I modified my DUSI v2 to have an external laser in an aluminum rod on the side of the DUSI. This introduced another set of issues for mounting on my tripod for calibration. The first attempt to calibrate with it was not much better. I probably should do a handheld test (comparison to Theodolite data) since hand alignment is easier with the external laser. That should be the clincher for determining if it is really a software bug.

I haven't tried the v2 with the PC software. Sean is strictly a UNIX user. So any PC testing would have been done by someone else. If you have access to a UNIX machine, give that a try.

I have sent Sean an email. As far as I know he is still living on his boat (instead of in the Colorado Mountains), which makes it difficult to test devices sensitive to magnetic interference!

Sean's software is all open source, so if any of you are knowledgable about that stuff, take a look. He actually would like someone to take on the hardware design side - he really only likes doing the software. And a boat is not a good location for building hardware.

I like the DUSI approach - I wish it was performing as well as the beta software. It is probably an easy fix, but it has to be found to be fixed.

Currently of the(or at least my) three instruments - the Disto X is the only one I have taken on real cave surveys. And I find it needs to be recalibration every 3-4 weeks to produce good data. I think (for my accuracy requirements) the Pony needs a better algorithm (e.g. the Disto X / DUSI approach) and the DUSI needs (I think) a bug fixed. YMMV of course
LWB
Prolific Poster
 
Posts: 117
Joined: Oct 29, 2007 7:36 pm
  

Re: DUSI - Windows Connection and Calibration Problems

Postby Dan Henry » Mar 31, 2010 10:48 pm

Does anybody have a working DUSI with v2 on it or know of anybody who does? There's 4 of them sitting collecting dust at a project I've worked on because nobody can get them to calibrate well enough to trust them. Sure would be nice to know what's wrong...
Dan Henry
Asheville, NC
NSS#47794
User avatar
Dan Henry
Frequent Poster
 
Posts: 54
Joined: Feb 13, 2008 4:48 pm
Name: Dan Henry
NSS #: 47794
Primary Grotto Affiliation: Flittermouse Grotto
  

Re: DUSI - Windows Connection and Calibration Problems

Postby Roppelcaver » Apr 2, 2010 8:59 am

The .inf file contains the driver for the device. When plugging in a device PnP searches the driver store for the device id of the plugged in device. If none are found, it prompts you for the driver that might be on a CD or downloaded to your desktop. If after pointing to the driver, if the device ID does not match, you will get a message saying this is the wrong driver (I forget exactly). Or, if the inf file is otherwise corrupted or incomplete, you will get a similar message. What you have sounds like a null file, or something similar. Ask the manufacturer for an updated driver.
Roppel Caver guy
Roppelcaver
Prolific Poster
 
Posts: 158
Joined: Sep 4, 2005 11:23 pm
  

Re: DUSI - Windows Connection and Calibration Problems

Postby Dan Henry » Apr 3, 2010 9:30 am

Sorry, I think you responded to the OP's question about the driver problem, but i was being a bit more broad. It seems these units just don't check out when taking backsights, even after working drivers and careful calibration. Clinos apparrently work fine and check well on backsights, but compass readings are consistently 5+ degrees off while the users are doing the same things that they would do that get consistent readings from Suuntos. If you have to strip naked and be 500' from anything metalic, it's just not practical. Has anybody found a way to get these V2 units to work well in the field and get repeatable checks with backsights? What settings are you using?
Dan Henry
Asheville, NC
NSS#47794
User avatar
Dan Henry
Frequent Poster
 
Posts: 54
Joined: Feb 13, 2008 4:48 pm
Name: Dan Henry
NSS #: 47794
Primary Grotto Affiliation: Flittermouse Grotto
  

Re: DUSI - Windows Connection and Calibration Problems

Postby Sau » Jul 14, 2010 9:34 pm

If you need download date in windows do:
in hyper terminal -standart win program make connection with settings
speed 38400 8 NOPARITY ONESTOPBIT

In console print: get survey/measurements

driver find at internet, ft232 chip. Version of file CDM 2.04.16.exe

Here program on python: http://ifolder.ru/18542321
Sau
Infrequent Poster
 
Posts: 6
Joined: Jul 29, 2009 5:54 am
Location: Тюмень
  

Re: DUSI - Windows Connection and Calibration Problems

Postby DBryce » Jul 16, 2010 4:52 pm

Derek:

I feel your pain. I have a DUSI V1 and have never got it calibrated, nor communicating with my Windows machine.

I would like to try to calibrate it again, but can no longer view the videos on Sean's website (http://digitalsurveyinstruments.com/downloads) that show how.

I, too, manually installed the driver, but didn't have the problems you did. It seemed to work fine. DataViewer runs "autofind", finds the device, and I can even run "Populate" and it seems to work, but I cannot seem to get any Survey data off of the DUSI (junk data I took sitting at my desk, but about six survey stations).

Sean is a very smart guy, maybe too smart. He can't talk to us normal I.Q. kind of guys (and I'm a Mechanical Engineer...I have to deal with electronics and programing stuff often, PLCs, Data loggers, cameras, etc.).

Could one of you types out there who has gotten this thing working please write a manual for the rest of us?

Otherwise, I got a DUSI V1 here for sale....real cheap.

Dan
Daniel Bryce
Front Range Grotto / Colorado Grotto / NSS 53632
User avatar
DBryce
Infrequent Poster
 
Posts: 13
Joined: May 18, 2007 10:21 pm
  

Re: DUSI - Windows Connection and Calibration Problems

Postby DBryce » Jul 16, 2010 5:02 pm

Oh, forgot to mention, I can HEAR the videos, just can't see them. Must be a codec thing, but I can't figure it out.

And my Windows Device Manager says the same thing yours does, under "PORTS", it is listed as %DESCRIPTION% (COM 6).


Image


Image
Daniel Bryce
Front Range Grotto / Colorado Grotto / NSS 53632
User avatar
DBryce
Infrequent Poster
 
Posts: 13
Joined: May 18, 2007 10:21 pm
  

Re: DUSI - Windows Connection and Calibration Problems

Postby Extremeophile » Jul 19, 2010 11:02 am

Funny. I downloaded copies of those calibration videos and I can see the video but the audio is garbled. Maybe between us we can put together the instructions. I'll bring my copies to the next grotto meeting.

I've been playing with the sensitivity settings and recalibrating and think I may have found settings that give a good calibration. I'm defining a good calibration as one that gives the same azimuth and clino readings as the device is rotated around the LASER axis (e.g. screen facing up, right, down, left), and good front sight / back sight agreement. And by same I mean azimuth and clino within 1 degree. Using the default settings on the instrument I've been getting very good calibration for the clino, but very bad numbers for the azimuth - sometimes >5 degrees of error. Once I get these settings sorted out and feel confident they are reproducible I'll post a guide.

Derek
User avatar
Extremeophile
NSS Hall Of Fame Poster
 
Posts: 761
Joined: Dec 7, 2009 7:37 pm
Location: Littleton, CO
Name: Derek Bristol
NSS #: 34941
Primary Grotto Affiliation: Colorado Grotto
  

Next

Return to Survey and Cartography Forum

Who is online

Users browsing this forum: Majestic-12 [Bot]