Topic: 997c loses GPS position at slow trolling speeds, below .6mph

[mfbab]

This is an issue for people like myself who slow troll for crappie or also during times that I am trying to position the boat for casting to GPS marked underwater brushpiles while bass fishing. If the boat is not moving above approximately .5-.6 mph, I lose my gps position and the unit displays a circle on the screen. This circle has been affectionately named "The Donut." I am supplying a link to a thread on Crappie.com which was started with the intent of gathering a large group of HB owners together in an attempt to motivate HB to resolve this problem for us through a firmware update.

http://www.crappie.com/crappie/fishing-electronics-trolling-motors/162145-project-hb-donut.html

[Bob B]

+1  Humminbird please fix this

[RGecy]

I would suggest that you guys make sure to post your thoughts on the Wish List.  I believe this is lready in there, but you can certainly reiterate your desire to have it fixed.

I think it would be fairly simple in programming.  I would imagine with the older 16 CH gps units, the position would jump around so much that the boat icon would constantly change direction.  With the new 50ch gps, it should help to steady.  They could also use an algorithm at slow speeds to compensate for this and not allow it to change direction so drastically.

This fix is very doable!

Robert

[mfbab]

Thanks RGecy, from what I've been told I agree that it is a very doable fix.  For whatever reason, they just are not getting to it.  This has been brought up to HB CS reps by myself and many others for well over 2 years now and there have been many firmware uopdates (good ones too) during that time but still nothing addressing the donut. 

To be honest, I just feel a little silly having this expensive HB unit on my boat while simultaneously having to rely on marker buoys or other GPS units such as Garmin, Lowrance or Delorme to name a few, just to hold my boat on brushpiles or other underwater items I may have marked. 

BTW, the irony of the fact that I was able to mark many of these pieces of cover because of HB's revolutionary SI technology and the also revolutionary idea of integrating GPS into that same sreen is not lost on me

[mfbab]

I apologize if this is an obvious question, but how do I add this to the wish list?

Thx,
Rob

[sonar2000]

Just click on the line below....

http://forums.sideimagingsoft.com/index.php?board=48.0

Chuck

[Bob B]

I would suggest that you guys make sure to post your thoughts on the Wish List.  I believe this is lready in there, but you can certainly reiterate your desire to have it fixed.

I think it would be fairly simple in programming.  I would imagine with the older 16 CH gps units, the position would jump around so much that the boat icon would constantly change direction.  With the new 50ch gps, it should help to steady.  They could also use an algorithm at slow speeds to compensate for this and not allow it to change direction so drastically.

This fix is very doable!

Robert

Robert......Guess great minds think alike.

You know these units a LOT better than I do, but this is exactly what I was thinking.  I do programming for building controls, and this is exactly how I would compensate for an erratic reading....either by doing a time average, or setting a threshold.

This is probably part of the original code to compensate for the old GPS antennas accuracy.  They probably required a certain speed.....maybe 0.5mph before the boat is not considered to be stationary.  If that is the case, seems like all they would have to do is decrease the value of this paramater.....maybe to 0.1.

I thought about this a little more, and I think the best way to do it would be to combine both strategies to get the best of both worlds....instant response as well as direction at slow trolling speed. 

If the speed is above a "certain" threshold, give instantaneous direction information.  Below that threshold do a time average of the readings to allow speed and direction info at a VERY low speed.

[sonar2000]

Just click on the line below....

http://forums.sideimagingsoft.com/index.php?board=48.0

Chuck

[mfbab]

Thx guys, I added it to the wish list.  Others on the thread I linked in my original post are in agreement that new algorithms should solve this problem also. 

Rob

[FuzzyGrub]

If my memory is correct (big if), this became more of an issue after the S/W release where they increased the gps sampling speed.   I thought that release came out for 16 channel gps, ie before the 50 channel ones were released.   But, that was a number of releases ago. 

[mfbab]

This is the donut screen I'm referring to....

[rnvinc]

Here is an interesting interpretation of the HB "Donut" from some one way more knowledgeable than I....

It's over at the Unofficial HB website thread titled "Project HB Donut"...

http://www.xumba.scholleco.com/viewtopic.php?t=3246

posted by fsae99...

The donut issue is to combat erratic direction arrow in chart views. The request is really to write algorithms that have the GPS puck sample the data more often and determine if the heading has really changed before changing the direction arrow displayed on the screen. So you would do something like sample the gps signal 100 times a second and if the average position still had you going in the same direction the arrow would remain pointed in the same direction.

This would be to combat what I call the the GPS drift, the real issue. If you look a plot of position of any GPS (most all commercially sold) left in 1 position for any length of time you will see plot points all over the place. You would not want to increase the freq at which the data is sent to the head unit you would end up right back where you started, erratic direction arrow.

Alternately HB could add 3 axis tilt compensated Electronic compass with 3 axis accelerometers and 3 axis rate gyros to the GPS puck. Then use electronic compass for heading arrow and accelerometers + rate gyros to stabilize position updates. AIrmar has one the GH2183 GPS & Heading I think it is $2500 USD for just the GPS puck. I think that is too much and should be able to be done for an additional $400 USD to the cost of existing HB GPS puck. I think now even if you did buy the Airmar gps puck the HB unit would still show the donut under .5 mph due to the SW change.

While it is better now than it used to be, there is room for improvement on this issue without going to the extreme of changing GPS design.

HB has done a lot for us, was a time when removing the water column from the display was never going to happen (done now). The donut, waypoint management, and user created maps (vector) are the 3 things left to make it as close to perfect as you can gtfor $1K, $2K, and 3K machines.

Thanks for reading.

Jim


I think this actually sums up what we at "Project HB Donut" have been trying to bring out into the open  all along....

This is NOT a GPS issue...

This is simply a lack of HB writing the correct algorithm calculation programming into the HB chartplotter units....

What we are trying to achieve with "Project HB Donut" is.....

Lowrance and Garmin have already set the industry standard of showing the "boat shaped directional tracking icon" down to speeds of 0.1mph...

WE ,as "slow trollers need that boat shaped directional tracking icon

to show and plot down to 0.1mph...

We only want our HB units to do what the competition does....

Rickie Culp (an almost satisfied HB owner)...

[kosmo]

http://forums.sideimagingsoft.com/index.php?topic=2115.msg11656;topicseen#new

[Moose1am]

What you really are asking for is improved HDOP and accuracy.  If the GPS units were more accurate you could slow troll at 0.1 mph and still see that you moved more than 1 ft per second. 

.5 mph is only 1.47 ft per second.  The accuracy circle is about 10 ft in radius 95% of the time.  If you were anchored to a stump your gps would show you moving around in a big ten foot radius over time.  Your boat won't be moving but all but the most expensive surveying GPS units will show you moving around.
 
Ricky if you are only moving at 0.1 mph you are not going very far each second.  You only are moving 0.73 ft /sec. 
 
Good luck in your quest but I fear you are wasting your time.  If Humminbird could fix their unit to make it more accurate than it is now they would do it.  But the cost would be too great. 
 
My Garmin eTrex Vista has an electronic compass built into it and it still flip flops the map when I got too slow and have it in Heads UP Map display mode.  And in North UP mode the arrow changes directions at low speed due to the errors of the GPS signal.
 
Do me a favor.  Take your gps unit outside and sit it in one spot for a few hours.  Record the readings ever few minutes and write them down.  Plot those gps coordinates on a map and see if they are all the same of if they vary by about 10 ft from each other.  If you are lucky all the data points will be within a 10 ft radius.  95% of the data points should fall within a 10 ft circle. 5% of the data points on the graph may fall outside that 10 ft circle.

[kosmo]

What you really are asking for is improved HDOP and accuracy.  If the GPS units were more accurate you could slow troll at 0.1 mph and still see that you moved more than 1 ft per second. 

.5 mph is only 1.47 ft per second.  The accuracy circle is about 10 ft in radius 95% of the time.  If you were anchored to a stump your gps would show you moving around in a big ten foot radius over time.  Your boat won't be moving but all but the most expensive surveying GPS units will show you moving around.
 
I like the idea of the electronic compass or the accelerometers in the GPS puck antenna but that's expensive and adds to the cost.  Perhaps that would work until we get more accuracy in the GPS systems as a whole.        i dont think garmin or lowrance work any better