Strange Therion/survex disparity in altitudes
HI Therion friends, I'm working on a dataset with a friend, We have survex and Therion working independently, but putting the aven models [.3d] from the survex dataset and the therion dataset side-by-side we're coming out with differences. We nailed the x/y difference down somewhat with the magnetic corrections in the .th file due to dates. But we're now left with a difference in the height/altitude between the two models. The cave is a single entrance cave which is higher at the end (with a streamway 260-340m below), All of the differences are weird but the weirdest difference in altitude is in the middle, where there's a 60m difference in the survex/therion models, however further on at stream level, there is only a 20m difference [665 - 685] ##### The models are kept at: Survex: http://cave-registry.org.uk/svn/Andara/Tresviso/ Therion: http://cave-registry.org.uk/svn/Andara/Tresviso/Nacimiento/therion/Nacimient... run therion using Nacimiento2020_master.thconfig ##### I sent the following message to my friend to describe the issue: # "Hmm, I think the next thing to try and crack is the elevations. as I think there’s something going awry there, which might point us further towards the remaining 50m in n-s separation. On the .3d file generated from therion I get the following Entrance= 470m Dr2000stream_pina.23038= 665.30m Terror_firma.6= 933.34m On survex model I get: Entrance= 480m Dr2000_streamway2.23038= 685.26 Terror_firma2.6= 1025.53 I really can’t answer the 10m diff at entrance, the 20m difference at stream level or 92m difference at the top. Extrapolating down, Jurassic highway.0 is still 90m separation. However, satan_dr2000.1202 is 41m separation (higher in survex). Increases again to 66m sep at garden Eden.16 vs consort_climax.16E" If anyone can give me some help I would be thankful, as i'm confused! Regards, Alastair Gott.
Hi Alastair,
Fun that you choose to use a different file/folder structure for the two versions. Makes them a little awkward to compare ;) Fixes inline in one, and in a global file in the other...
On the .3d file generated from therion I get the following Entrance= 470m
Because that's how it is fixed. http://cave-registry.org.uk/svn/Andara/Tresviso/Nacimiento/therion/Nacimient... fix 1 0363921 4789684 470
On survex model I get: Entrance= 480m
Because that's how it is fixed. http://cave-registry.org.uk/svn/Andara/Tresviso/FixedPoints.svx *fix T13 0363921 4789684 480; Cueva del Nacimiento (Cueva del Agua) That accounts for the 10 m error, right? I notice you also fix the entrances in the Therion version with "cs UTM30N" but you don't do that in the Survex version.
92m difference at the top.
There is something hugely different in one part of the cave. Aside from the crazy rotational error. dr2000_aven is really tall in the .svx version, but goes up and down in the Therion version. Look for station 2214 in domino_satan in the Therion version to find the same station. DOWN has been mis-converted into +90 rather than -90, when converting from Survex format to Therion format. Survex: 2213 2212 B 3.49 - DOWN Therion: 2213 2212 3.49 311 90 I'm guessing that error may turn up elsewhere. consort_climax looks similarly wrong. tinkerbell_bigpitch (Survex name) is also completely different (wrong?) in the Therion version. I think maybe you have a cluster of equates in the domino_colins area (Therion name) that are not supposed to be joined. It's really hard to look for differences with all the extra duplicated data scattered everywhere though. Making my head hurt. Working on an isolated and simplified set of data would really help. I need to get some sleep. I will pick this up when I have some time tomorrow, if you don't manage to spot the rest of them. Cheers, Tarquin
DOWN has been mis-converted into +90 rather than -90, when converting from Survex format to Therion format. Survex: 2213 2212 B 3.49 - DOWN Therion: 2213 2212 3.49 311 90 I'm guessing that error may turn up elsewhere. consort_climax looks similarly wrong. As a point to note -90 and Down at treated differently for clino correction, they can be treated ed the same
From the survex manual "*infer plumbs on" tells cavern to interpret gradients of +/- 90 degrees as UP/DOWN (so it will not apply the clino correction to them). This is useful when the data has not been converted to have UP and DOWN Andrew
HI Tarquin and Andrew, Many thanks for that, We were just lost, I think we'll be able to pick it up from here with your helpful guidance. Should just be a case of looking through the .th files and looking for all the 90's I hope you are both well. Al. On Fri, 26 Feb 2021 at 07:33, Andrew Atkinson <andrew@wotcc.org.uk> wrote:
DOWN has been mis-converted into +90 rather than -90, when converting from Survex format to Therion format. Survex: 2213 2212 B 3.49 - DOWN Therion: 2213 2212 3.49 311 90 I'm guessing that error may turn up elsewhere. consort_climax looks similarly wrong. As a point to note -90 and Down at treated differently for clino correction, they can be treated ed the same
From the survex manual
"*infer plumbs on" tells cavern to interpret gradients of +/- 90 degrees as UP/DOWN (so it will not apply the clino correction to them). This is useful when the data has not been converted to have UP and DOWN
Andrew _______________________________________________ Therion mailing list Therion@speleo.sk https://mailman.speleo.sk/listinfo/therion
Hi therion people, I’m working on the dataset again with my friend. We’ve found that the survex model uses a compass calibration in some places and are trying to work out if this needs transposing across to the therion model. But in the process I have found out that when we give therion a compass reading it’s always taking off 0.9 to 1 degree from the reading given in the .th data. I’ve just run changes to entrance-boulder in both therion and survex. Below results just for first leg 1-2 No date+compass calib= 133.6 deg survex// 132.6 deg therion Date + no compass calib=131.0 deg survex// 130.1 deg therion No date+no compass calib=131 deg survex// 130deg therion Date + compass calib=133.6 deg survex// 132.7 deg therion So clearly therion is correcting the data 1[] 2[] 13.4[] 131[] 10.0 By always taking one degree off the compass irrespective of what we tell it. (For information: When calibration used survex= *calibrate compass -2.65 // therion= calibrate compass 357.35) Does anyone have any ideas where therion is pulling this 1 degree correction from? Regards, Alastair Gott.
On 26 Feb 2021, at 07:43, A Gott <alastairgott@gmail.com> wrote:
HI Tarquin and Andrew,
Many thanks for that, We were just lost, I think we'll be able to pick it up from here with your helpful guidance. Should just be a case of looking through the .th files and looking for all the 90's
I hope you are both well.
Al.
On Fri, 26 Feb 2021 at 07:33, Andrew Atkinson <andrew@wotcc.org.uk> wrote:
DOWN has been mis-converted into +90 rather than -90, when converting from Survex format to Therion format. Survex: 2213 2212 B 3.49 - DOWN Therion: 2213 2212 3.49 311 90 I'm guessing that error may turn up elsewhere. consort_climax looks similarly wrong. As a point to note -90 and Down at treated differently for clino correction, they can be treated ed the same
From the survex manual
"*infer plumbs on" tells cavern to interpret gradients of +/- 90 degrees as UP/DOWN (so it will not apply the clino correction to them). This is useful when the data has not been converted to have UP and DOWN
Andrew _______________________________________________ Therion mailing list Therion@speleo.sk https://mailman.speleo.sk/listinfo/therion
First when you say calibration, you do mean calibration not declination, (or calibrate declination, which is an older form of declination) which have slightly different functions? There is more about these in the documentation, which can be useful when dealing with old data sets. I started a quick test hoping this would only take 5 mins, but got no where, but found something interesting, I have run out of time, not read the manual etc, so just putting this up as it could be useful I produced two files identical except in the annoying format differences. See below I was expecting to see some difference between the 1980 and 220 as the declination has moved about 3° I was interested to see what happened to the no date one. However, the answer was nothing happened everything came out the same, with not adjustment, both in survex and therion. I noticed I didn't get a list of declinations, so ran one of my older files, and it was there geomag declinations (deg): 2010.1.1 -2.6622 2011.1.1 -2.4922 So at some point I knew what I was doing, but now I don't, but there are no errors and no way to tell its not working! Now I think that the problem with Therion is it is doing something weird with the date. Happily accepts date 9999999 or date 999.11 Not happy with date 99.99999 invalid date month Goodness know how it is interpreting those Survex give a sensible error message about the dates that can be use. Andrew Therion 5.5.7 survey all equate 1@2020 1@nodate 1@1980 survey 2020 centerline cs OSGB:ST date 2020 data normal from to tape compass clino 1 2 15 90 0 2 3 10 180 0 endcentreline endsurvey survey nodate centerline cs OSGB:ST data normal from to tape compass clino 1 2 15 90 0 2 3 10 180 0 endcentreline endsurvey survey 1980 centerline cs OSGB:ST date 1980 data normal from to tape compass clino 1 2 15 90 0 2 3 10 180 0 endcentreline endsurvey endsurvey Survex *Begin all *Equate 2020.1 nodate.1 1980.1 *Begin 2020 *cs OSGB:ST *date 2020 *Data normal from to tape compass clino 1 2 15 90 0 2 3 10 180 0 *End 2020 *Begin nodate *Data normal from to tape compass clino *cs OSGB:ST 1 2 15 90 0 2 3 10 180 0 *End nodate *Begin 1980 *cs OSGB:ST *Date 1980 *Data normal from to tape compass clino 1 2 15 90 0 2 3 10 180 0 *End 1980 *End all On 15/03/2021 14:45, Alastair Gott wrote:
Hi therion people,
I’m working on the dataset again with my friend.
We’ve found that the survex model uses a compass calibration in some places and are trying to work out if this needs transposing across to the therion model. But in the process I have found out that when we give therion a compass reading it’s always taking off 0.9 to 1 degree from the reading given in the .th data.
I’ve just run changes to entrance-boulder in both therion and survex. Below results just for first leg 1-2
No date+compass calib= 133.6 deg survex// 132.6 deg therion Date + no compass calib=131.0 deg survex// 130.1 deg therion No date+no compass calib=131 deg survex// 130deg therion Date + compass calib=133.6 deg survex// 132.7 deg therion
So clearly therion is correcting the data 1[] 2[] 13.4[] 131[] 10.0 By always taking one degree off the compass irrespective of what we tell it.
(For information: When calibration used survex= *calibrate compass -2.65 // therion= calibrate compass 357.35)
Does anyone have any ideas where therion is pulling this 1 degree correction from?
Regards, Alastair Gott.
On 26 Feb 2021, at 07:43, A Gott <alastairgott@gmail.com <mailto:alastairgott@gmail.com>> wrote:
HI Tarquin and Andrew,
Many thanks for that, We were just lost, I think we'll be able to pick it up from here with your helpful guidance. Should just be a case of looking through the .th files and looking for all the 90's
I hope you are both well.
Al.
On Fri, 26 Feb 2021 at 07:33, Andrew Atkinson <andrew@wotcc.org.uk <mailto:andrew@wotcc.org.uk>> wrote:
> DOWN has been mis-converted into +90 rather than -90, when converting > from Survex format to Therion format. > Survex: > 2213 2212 B 3.49 - DOWN > Therion: > 2213 2212 3.49 311 90 > I'm guessing that error may turn up elsewhere. consort_climax looks > similarly wrong. As a point to note -90 and Down at treated differently for clino correction, they can be treated ed the same
From the survex manual
"*infer plumbs on" tells cavern to interpret gradients of +/- 90 degrees as UP/DOWN (so it will not apply the clino correction to them). This is useful when the data has not been converted to have UP and DOWN
Andrew _______________________________________________ Therion mailing list Therion@speleo.sk <mailto:Therion@speleo.sk> https://mailman.speleo.sk/listinfo/therion <https://mailman.speleo.sk/listinfo/therion>
_______________________________________________ Therion mailing list Therion@speleo.sk https://mailman.speleo.sk/listinfo/therion
Hi Andrew, I'm afraid I there is no answer as to whether I mean calibration, declination or calibrate declination. As the survex model has the *calibrate compass -2.65 already in it, and I don't know whether this is because they knew the compass was out, or whether they corrected for the declination in the survex model when the data was put in. When I refer to the calibration in Therion, I've just applied the same function calibrate compass 357.3 I was hoping that running the comparison between therion and survex on the first leg of the survey, I could work out by trial and error what I should do to make the data the same. I'm happy to apply a fix to all the .th files, to ensure that the models align, but this doesn't seem like the therion way given that the underlying data is the same. PS. I know there will be parts of the datasets where there are differences, but till I can make the first leg the same, I think it will be difficult for me to spot even the biggest difference. Therion.th survey entrance_boulder centerline cs UTM30N station 1 "Cueva del Nacimiento" entrance fix 1 0363921 4789684 470 date 2011.08.14 #calibrate compass 357.35 units length meters units compass clino degrees data normal from to length compass clino 1 2 13.40 131 10.0 2 3 10.50 194 1.0 endcentreline endsurvey config source therion.th export model -o therion.3d Survex.svx *begin entrance_boulder *entrance 1 *fix 1 0363921 4789684 470; Cueva del Nacimiento (Cueva del Agua) *date 2011.08.14 ;instruments: SWCC compass and clino, colins laser pointer ;*calibrate compass -2.65 *calibrate clino -0.5 ;**ENTRANCE TO BOULDER HALL** ;**************************** *data normal from to direction tape compass clino 1 2 13.40 131 10.0 2 3 10.50 194 1.0 *end entrance_boulder Regards, Alastair Gott. AlastairGott@hotmail.com, M: 07931779380. ________________________________ From: Therion <therion-bounces@speleo.sk> on behalf of Andrew Atkinson <andrew@wotcc.org.uk> Sent: 15 March 2021 16:53 To: therion@speleo.sk <therion@speleo.sk> Subject: Re: [Therion] compass correction Re: Strange Therion/survex disparity First when you say calibration, you do mean calibration not declination, (or calibrate declination, which is an older form of declination) which have slightly different functions? There is more about these in the documentation, which can be useful when dealing with old data sets. I started a quick test hoping this would only take 5 mins, but got no where, but found something interesting, I have run out of time, not read the manual etc, so just putting this up as it could be useful I produced two files identical except in the annoying format differences. See below I was expecting to see some difference between the 1980 and 220 as the declination has moved about 3° I was interested to see what happened to the no date one. However, the answer was nothing happened everything came out the same, with not adjustment, both in survex and therion. I noticed I didn't get a list of declinations, so ran one of my older files, and it was there geomag declinations (deg): 2010.1.1 -2.6622 2011.1.1 -2.4922 So at some point I knew what I was doing, but now I don't, but there are no errors and no way to tell its not working! Now I think that the problem with Therion is it is doing something weird with the date. Happily accepts date 9999999 or date 999.11 Not happy with date 99.99999 invalid date month Goodness know how it is interpreting those Survex give a sensible error message about the dates that can be use. Andrew Therion 5.5.7 survey all equate 1@2020 1@nodate 1@1980 survey 2020 centerline cs OSGB:ST date 2020 data normal from to tape compass clino 1 2 15 90 0 2 3 10 180 0 endcentreline endsurvey survey nodate centerline cs OSGB:ST data normal from to tape compass clino 1 2 15 90 0 2 3 10 180 0 endcentreline endsurvey survey 1980 centerline cs OSGB:ST date 1980 data normal from to tape compass clino 1 2 15 90 0 2 3 10 180 0 endcentreline endsurvey endsurvey Survex *Begin all *Equate 2020.1 nodate.1 1980.1 *Begin 2020 *cs OSGB:ST *date 2020 *Data normal from to tape compass clino 1 2 15 90 0 2 3 10 180 0 *End 2020 *Begin nodate *Data normal from to tape compass clino *cs OSGB:ST 1 2 15 90 0 2 3 10 180 0 *End nodate *Begin 1980 *cs OSGB:ST *Date 1980 *Data normal from to tape compass clino 1 2 15 90 0 2 3 10 180 0 *End 1980 *End all On 15/03/2021 14:45, Alastair Gott wrote:
Hi therion people,
I’m working on the dataset again with my friend.
We’ve found that the survex model uses a compass calibration in some places and are trying to work out if this needs transposing across to the therion model. But in the process I have found out that when we give therion a compass reading it’s always taking off 0.9 to 1 degree from the reading given in the .th data.
I’ve just run changes to entrance-boulder in both therion and survex. Below results just for first leg 1-2
No date+compass calib= 133.6 deg survex// 132.6 deg therion Date + no compass calib=131.0 deg survex// 130.1 deg therion No date+no compass calib=131 deg survex// 130deg therion Date + compass calib=133.6 deg survex// 132.7 deg therion
So clearly therion is correcting the data 1[] 2[] 13.4[] 131[] 10.0 By always taking one degree off the compass irrespective of what we tell it.
(For information: When calibration used survex= *calibrate compass -2.65 // therion= calibrate compass 357.35)
Does anyone have any ideas where therion is pulling this 1 degree correction from?
Regards, Alastair Gott.
On 26 Feb 2021, at 07:43, A Gott <alastairgott@gmail.com <mailto:alastairgott@gmail.com>> wrote:
HI Tarquin and Andrew,
Many thanks for that, We were just lost, I think we'll be able to pick it up from here with your helpful guidance. Should just be a case of looking through the .th files and looking for all the 90's
I hope you are both well.
Al.
On Fri, 26 Feb 2021 at 07:33, Andrew Atkinson <andrew@wotcc.org.uk <mailto:andrew@wotcc.org.uk>> wrote:
> DOWN has been mis-converted into +90 rather than -90, when converting > from Survex format to Therion format. > Survex: > 2213 2212 B 3.49 - DOWN > Therion: > 2213 2212 3.49 311 90 > I'm guessing that error may turn up elsewhere. consort_climax looks > similarly wrong. As a point to note -90 and Down at treated differently for clino correction, they can be treated ed the same
From the survex manual
"*infer plumbs on" tells cavern to interpret gradients of +/- 90 degrees as UP/DOWN (so it will not apply the clino correction to them). This is useful when the data has not been converted to have UP and DOWN
Andrew _______________________________________________ Therion mailing list Therion@speleo.sk <mailto:Therion@speleo.sk> https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmailman.speleo.sk%2Flistinfo%2Ftherion&data=04%7C01%7C%7Cb0afac4de572448cf95508d8e7d2f653%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637514240552340058%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=0kBJbg6XfaDY5GZiSEKNy%2Bp7uJ%2BNB4VGUZndn8sbuYg%3D&reserved=0 <https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmailman.speleo.sk%2Flistinfo%2Ftherion&data=04%7C01%7C%7Cb0afac4de572448cf95508d8e7d2f653%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637514240552340058%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=0kBJbg6XfaDY5GZiSEKNy%2Bp7uJ%2BNB4VGUZndn8sbuYg%3D&reserved=0>
_______________________________________________ Therion mailing list Therion@speleo.sk https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmailman.speleo.sk%2Flistinfo%2Ftherion&data=04%7C01%7C%7Cb0afac4de572448cf95508d8e7d2f653%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637514240552340058%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=0kBJbg6XfaDY5GZiSEKNy%2Bp7uJ%2BNB4VGUZndn8sbuYg%3D&reserved=0
_______________________________________________ Therion mailing list Therion@speleo.sk https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmailman.speleo.sk%2Flistinfo%2Ftherion&data=04%7C01%7C%7Cb0afac4de572448cf95508d8e7d2f653%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637514240552340058%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=0kBJbg6XfaDY5GZiSEKNy%2Bp7uJ%2BNB4VGUZndn8sbuYg%3D&reserved=0
On Mon, Mar 15, 2021 at 05:50:18PM +0000, alastair gott wrote:
I'm afraid I there is no answer as to whether I mean calibration, declination or calibrate declination.
As the survex model has the *calibrate compass -2.65 already in it, and I don't know whether this is because they knew the compass was out, or whether they corrected for the declination in the survex model when the data was put in.
It would help a lot to determine what -2.65 includes. If it wasn't recorded and you can't usefully ask the people who recorded the data, you could try adjusting the data assuming this includes the declination and again assuming it doesn't and see which version gives better loop errors. Aven can colour by error which provides a way to visualise this.
When I refer to the calibration in Therion, I've just applied the same function calibrate compass 357.3
I was hoping that running the comparison between therion and survex on the first leg of the survey, I could work out by trial and error what I should do to make the data the same. I'm happy to apply a fix to all the .th files, to ensure that the models align, but this doesn't seem like the therion way given that the underlying data is the same.
A key difference here is that Therion automatically corrects for declination and for grid convergence (the difference between true north and grid north) if you specify a coordinate system and put a date on a survey. It does this by calculating the corrections at the location which is the average of all the specified fixed points (this doesn't seem to be documented, but that's my understanding from reading the source code.) Survex takes a different approach and you have to turn on the equivalent feature and specify the location to calculate the corrections at, e.g.: *declination auto 0410600 5282000 1234 I didn't follow therion's automatic approach because it has the downside that adding a new fixed point slightly changes the location of *ALL* survey stations (apart from those which are also fixed points).
PS. I know there will be parts of the datasets where there are differences, but till I can make the first leg the same, I think it will be difficult for me to spot even the biggest difference.
Rather than trying to maintain two parallel version of the data set, my suggestion would be to save yourself a lot of work and just tell Therion to use a .3d file produced from the Survex data set for the station coordinates (see the "import" command). Cheers, Olly
participants (6)
-
A Gott -
Alastair Gott
-
alastair gott -
Andrew Atkinson -
Olly Betts -
Tarquin Wilton-Jones