Wild Dog Adventure Riding

Riding: Plan, Report and Racing => Racing Section => Topic started by: Weedkiller - Adie on October 09, 2015, 10:24:25 am

Title: Locally Developed Navigation eq. Amageza Reportback
Post by: Weedkiller - Adie on October 09, 2015, 10:24:25 am

firstly, thanks to those who trusted us and used the units at Amageza.  Some were highly impressed and some were a bit peeved off.

For those who did not follow the original build thread it is here: http://www.wilddog.za.net/forum/index.php?topic=166589.0 (http://www.wilddog.za.net/forum/index.php?topic=166589.0)

Ok, Amageza is done and dusted. We evaluated some of the units that gave problems during the race and came to the following conclusion:

1: Buttons.  Although the buttons are IP67 rated they are CRAP.  About 80% of the issues was caused by 'button bounce' where the vibration activated the button and the unit will 'go crazy'.  Some buttons were permanently ON and this made the unit un-usable.

2: GPS connector.  We had 2 cases where the GPS and CPU momentary lost connection.  The screen 'froze' at the last reading until the communication was restored.

3: Power supply.  2 units responded erratically on the bike.  Once back in the workshop they worked 100% on 'stable' power.  One unit was on the BMW Dakar motor with a history of over charging (up to 19 Volts).

4. Software. A few units still ran on the old software that was a bit 'cumbersome' to use but fully functional.  The user interface of the system was way to complex for the novice user but once set up properly was a breeze to use.

Both units on Andrew's (Kamanja) 525 was set up with GPS only input.  Both functioned 100% up to his off (day 3 or 4??).  The 2mm stainless steel baseplate was bent and the GPS connection failed on one unit.  The spare unit I gave him worked 100% for the rest of the race.  The GPS and roadbook distances was almost always spot on.

The two units on my KTM 450 also worked 100%.  Both was set up with a GPS and one had a wheel pickup as well.  Both units gave exact same readings in GPS mode (distance, CAP and time).  The distances was virtually the same if I ran the one unit from the wheel  and the other from GPS.

Planned upgrades.
1. We will replace the switches with a 'tactile feedback' type. Vibration will not affect them, they go 'click' if pressed.  They are also much smaller than the current units. 

2. GPS connector.  We will re-design the whole unit to improve the GPS mount and connector.

3. Power supply.  We decided to use two regulators in the unit for ultimate power control and stability.  First a 9V to smooth the 12v supply and then a 5V to power the unit.  This arrangement ensure both regulators run nice and cool.  BUT the Power from the bikes are NOT as stable as I thought and hence the power issues on some bikes.  We will revert back to the 12v-5v arrangement. (The 5v unit will just run slightly hotter but the 12V will be able to cope with the erratic supply.

4. Software.  All internal math etc of the units worked flawless.  We will however make the user interface MUCH easier.  Actually, only the distance reset function was an issue.

We are investigating option to rent out complete tower with roadbook holder and GPS based Trip and CAP units for West Coast Baja.  The unit will be self contained and only need a power connection.  The handlebar switch will be included. (See next post on the roadbook holder.)


Title: Re: Locally Developed Navigation eq. Amageza Reportback
Post by: Weedkiller - Adie on October 09, 2015, 11:24:23 am
Roadbook Holder.

The roadbook holders worked pretty well except for the wiring that came loose on one unit and the motor dismantling itself due to vibration.  We will locktite all motor assembly's before we fit them in the unit.  We had a few units with a switch for the backlight.  this did not work as planned.  No problems where the backlight was constantly on.

On my unit (a pre-production test unit) the drive assembly was not locktite'd and came loose on day 2.  In the process of tightning the shaft I dropped the drive pulley. GONE, WEG, vanished in the sand. A quick solution was to put the 'O' ring over both shafts right next to the manual knobs.  I just had to manually wind the paper back and sometimes had to 'help' the forward movement if the friction was too much.

It looks as if the type of O-ring is not too UV stable.  The one holding the lid is perished quite a bit.  We will also protect the drive pulley from falling off with a circlip when you remove the belt for fast loading and un-loading of the paper roll.

What I liked was the quick removal of the drive belt to make loading and un-loading of the paper smooth and fast.

Title: Re: Locally Developed Navigation eq. Amageza Reportback
Post by: Rokie on October 12, 2015, 07:41:39 am
Well done!! Keep up the good work.
Very inspiring stuff (from a bit of a distance)