rgleason / watchdog_pi

watchdog plugin for opencpn
GNU General Public License v3.0
1 stars 9 forks source link

Bug: Course Alarm #76

Open Corsair-63 opened 9 months ago

Corsair-63 commented 9 months ago

Windows 11 Pro ARM64 ver 22631.2506 OCPN 5.9.0-0 Watchdog 2.4.101.2

Corsair-63 commented 7 months ago

Windows 11 Pro ARM64 ver 22631.2861 OCPN 5.9.0-0+948.1ffe2ef Watchdog 2.4.103

still the same when setup the alarm, not read the heading device always COG, but on follow-up the alarm it follows the heading device reading.

rgleason commented 7 months ago

Corsair, I am sorry but I having a hard time understanding exactly what the problem is. Which value is incorrect and why? Devio is Deviation form the course = 20 degrees Rumbo Actual is Actual Course = 248
You are using the "Heading Sensor" (not GPS course)

Your instruments show Rumbo mag = Magnetic Course = 287 Rumbo verde = True Course = 287 real COG = Course over Ground = 249

So what is the problem?
Are you saying there is no current, and since STW Speed thru the water = 0 COG should be something else? Doesn't COG require movement of the boat to be accurate?

Corsair-63 commented 7 months ago

The course (COG) at that time was 248 (one second after was 249), you can see that the magnetic and true course are similar and they are 287.

the main problem is when you try to setup the reference course always is set to COG, and you know that when you're stopped is difficult for the GNSS to display a steady course, so it jumps on every NMEA data update.

and heading sensor is setup but not read when click on "Rumbo actual"

when the alarm is setup is curious, but the alarm follows the heading sensor reading and if not any exchange of course there is no alarm.

rgleason commented 6 months ago

@Corsair-63 Carmelo, we really need a VDR file with a ".txt" extension, showing this specific problem. It is necessary to debug and test. Can you upload a file as " ____.txt" using drag and drop to this Issue? Thank you.

Corsair-63 commented 6 months ago

Also, able to check at home with Pixel because it sends HCHDT & HCHDG sentences.

setup course alarm: click on heading device Click on actual course (Rumbo Actual) supposed to be set as heading device is reading, at this time Magnetic course 061, True course 059. COG undefined because I'm stopped, and the reading would be any after every click on "Actual course" button. click on the button displayed 258, after a second another click and 224 and .... when supposed to be set at 061 (magnetic) I manually entered 061 and setup the alarm and it follow the correct reading 061. but as reported previously always on every click to update the course and stated as "heading sensor" always goes to COG.

attached log file with the readings. Pixel_HDT_HDG.VDR.txt

I don´t know why is giving me too $XXHDT as virtual? I've done a double check about priorities, but I don't know where it comes the virtual input?

rgleason commented 6 months ago

Fixed Watchdog 2.4.105 Please Test and advise here. Thanks.

Corsair-63 commented 6 months ago

I could not test in deep because I continue with the problem of vessel taking water and I have the workshop on board, and I have to take the vessel from the water.

I did a quick test: downloaded latest WD activate it set-up the course alarm click on sync with heading sensor enabled and still goes to COG. manually setup to heading sensor reading and it follows it. Dashboard shows correctly the compass course.

rgleason commented 6 months ago

Thanks, still on the list to be fixed.

rgleason commented 6 months ago

Using your file, (I have not looked at it) run in VDR with Watchdog > Course open and click on "Current Course" it goes to 190 something using GPS, 200 something with Heading Sensor but Dashboard is showing 61 degrees.
Pixel_HDT_HDG.VDR.txt Screenshot (1930)

Screenshot (1931)

Isn't this a problem with the VDR file itself? I am not sure how this works, as I am simulating and don't have a heading sensor. I guess I've got to inspect the file you sent.

Search on HDG, looks like it is all 61 degrees. So why these results?

Search "HDG" (75 hits in 1 file of 1 searched) [Normal] C:\Users\fcgle\Data-Hesper\Nmea\Course\Pixel_HDT_HDG.VDR.txt (75 hits) Line 27: $HCHDG,061.0,,,000.2,E2C Line 59: $HCHDG,061.0,,,000.2,E2C Line 89: $HCHDG,061.0,,,000.2,E2C Line 123: $HCHDG,061.0,,,000.2,E2C Line 158: $HCHDG,061.0,,,000.2,E2C Line 185: $HCHDG,061.0,,,000.2,E2C Line 214: $HCHDG,061.0,,,000.2,E2C Line 250: $HCHDG,061.0,,,000.2,E2C Line 279: $HCHDG,061.0,,,000.2,E2C Line 312: $HCHDG,061.0,,,000.2,E2C Line 343: $HCHDG,061.0,,,000.2,E2C Line 371: $HCHDG,061.0,,,000.2,E2C Line 402: $HCHDG,061.0,,,000.2,E2C Line 431: $HCHDG,061.0,,,000.2,E2C

Corsair-63 commented 6 months ago

Hi Rick,

that's it if I set as "GPS course" and click on "current course", we should be agreed that it must read the COG. same if I set as "Heading device" doing the same it must read the Heading device value in that case 061.

rgleason commented 6 months ago

Using Watchdog v2.4.106 for test further testing.

Course-GPS

Course-MagHDG

Corsair-63 commented 6 months ago

If the vessel is stopped the COG exchange every second because any difference in decimal of minutes is the position makes a great exchange in COG, that's why on every click it exchanges. to properly check you should click on setup the alarm and at the very same moment check is dashboard the value shown as COG, and they should be the same, but after a second they won't match again.

rgleason commented 6 months ago

@Corsair-63 I am not understanding this

If the vessel is stopped the COG exchange every second because any difference in decimal of minutes is the position makes a great exchange in COG, that's why on every click it exchanges. to properly check you should click on setup the alarm and at the very same moment check is dashboard the value shown as COG, and they should be the same, but after a second they won't match again.

If the vessel is stopped (not moving), the COG exchange [changes?] every second, [because of small differences in position GPSD] decimal minutes, great exchange[???] in COG, that's why on every click [of Synch] it exchanges [changes?]

to properly check you should click on setup the alarm [alarm Edit?] and at the very same moment check is [if?] dashboard value [is] shown as COG, and [if] they [are] the same, but after a second they won't match again.

Is this right?

Yes, but nevertheless it is not working correctly.

Corsair-63 commented 6 months ago

take the e.g. GPS position features is that 99% of the time the position should be in 1 metre (I'm writing from my memory) so, just take an example, draw one point and make circle of one meter or radius, plot one position and inside this circle plot another one, the COG is calculated from last position to actual one, that means that you can have any course from 0 to 360 by the exchange of decimals in position and always been inside the circle defined.

rgleason commented 6 months ago

So what should we do about this?

Also I think the display of "Course" should be improved to be

"Course GPS" ) or "Course HDG") entered value of "Course" and "off course degrees" and the actual "off course degrees"

Example: Course GPS 106 15 </> Dashboard COG sign changes with alerts. Example: Course HDG 106 15 </> Dashboard HDGmag sign changes with alert.

Corsair-63 commented 6 months ago

So what should we do about this?

Also I think the display of "Course" should be improved to be

"Course GPS" ) or "Course HDG") entered value of "Course" and "off course degrees" and the actual "off course degrees"

Example: Course GPS 106 15 </> Dashboard COG sign changes with alerts. Example: Course HDG 106 15 </> Dashboard HDGmag sign changes with alert.

if you're stopped the first example is correct, if you're underway the COG should be close to HDG TRU/MAG. the second example if you setup as Heading device, stopped or underway, should not give alarm, because WD follows correctly the heading device course.

what it does no properly do is when setup, if you choose GPS it correctly gives you the COG when click on Sync but if you choose heading device it has to do the same with the actual HDG TRU/MAG that are coming from you heading device but it doesn´t and gives you the value of actual COG.

rgleason commented 6 months ago

I have to understand and think about what you are saying Corsair, but I'd like to amend the proposed display values as follows to more closely match what is being used for Apparent Wind.

Example: When enter course=105 range=15 Course GPS 90</>[value from dashboard]</>120
Example: When enter course=105 range=15 Course HDG 90 </>[Value from Dashboard]</>120 SIgn and color changes with each alert similar to Wind.

Corsair-63 commented 6 months ago

Erratic follow-up of off course alarm windows 11 ARM64 - 0.5.9.0 WD 2.4.107

first, as usual when set-up and selected "heading device" and click on actual course always goes to COG but never to the course read in the heading device. sshot-2

when set-up the alarm as "heading device" and manually entered the course, the plugin until this version followed the heading device course without any problem. now it jumps from alarm to no alarm, it seems that exchanges the reading of COG and heading device very frequent. sshot-1 sshot-3

I attach one NMEA recording where the alarm has triggered and restored several times. setting the alarm at 288 and 20 degrees offset. WD Course.txt

rgleason commented 5 months ago

Course is completely screwed up. Using Edit "Current Course" enters a spurious number that relates to nothing known. The course shown in the display is not the course being shown in Dashboard. This alarm needs some serious improvements.

rgleason commented 5 months ago

Werner reports:

Checking the course alarm. Almost concluded too soon it did not work, since I looked at the Mag HDG instead of the COG.

Alarm works well with the settings I tried. Cog 183, limit 10 degrees either side. Have to click 5 times though to make it disappear.