TSDZ2 - VLCD5 Display Odometer Issues

[nbrus27]

I have a new tsdz2 750w Conversion Kit fitted to my bicycle and I configured the VLCD5 display setting wheel diameter to 700c (I also tried 28, 27, and 26) and I have the units set to miles, but the odometer is reading very high regardless of wheel diameter setting ... I did a trip of 18 miles and the odometer reads 27 miles for the trip. The speed readout seems fine from what I can estimate from experience, but the distance measurements are way off.

Is this a common issue and will the open source firmware fix it?

Is anyone else having this problem? A google search turned up zero results ... maybe I have a faulty display, but it seems to be working fine in every other way except for the odometer.

I also noticed the battery capacity gauge doesn't start to drop until battery is below 60% remaining (measured using another more accurate battery meter), but this is a minor issue compared to the excessively high distance readout.

 

[nbrus27]

I swapped a VLCD-5 display from another kit and the distance on odometer and trip are still behaving strangely ... the distance randomly increments and it even jumped from 0 to 0.8 miles after cycling a few meters ... very odd! The speed readout seem pretty accurate though and shows no issues.

As the same problem exists on two VLCD-5 displays it seems that the distance must be coming from the motor controller unit rather than the display. That's just a guess as I don't really know and don't have any info on how this system works, but if anyone here knows, then please reply and it will same me from guessing and I will know what replacement parts I need.

 

[Elinx]

.... please reply and it will same me from guessing and I will know what replacement parts I need.

It could be a bouncing reedcontact of the speedsensor.
This means you get a lot of pulses in a short time, with every rotation of the wheel and happens if the distance of the spokemagnet is too close.
It must be about 10mm.
Most easy to try is rotating the spokemagnet away.
With later OSF versions this problem is overruled by the software, because the software counts just one of these pulses.

 

[nbrus27]

Thanks for your suggestion Elinx ... it seems odd that the speed reading seems good, but the distance is way off ... I will try moving the speed sensor and see what happens. Much appreciated.

 

[nbrus27]

I just put my bike on a stand and adjusted the speed sensor so its further away from the magnet ... I will do a proper test ride tomorrow as its late here in UK ... but so far your suggestion Elinx seems to work as the distance reading doesn't seem erratic any more when spinning the rear wheel on a bike stand. I'll post a proper update tomorrow, but so far your insight seems spot on.

 

[Elinx]

... it seems odd that the speed reading seems good, but the distance is way off ... ...

It depends how these values are calculated.
The difference is the time value for speed, which isn't needed for distance.
It is a known problem that the tsdz2 controller is extremely sensitive for the right magnet distance.
As said, the latest OSF does some valuation of the speedsensor signals to prevent wrong calcutation of the distance.


"Added a filter to the speed sensor to avoid problems of adjusting the distance from the magnet."

 

[nbrus27]

... it seems odd that the speed reading seems good, but the distance is way off ... ...

It depends how these values are calculated.
The difference is the time value for speed, which isn't needed for distance.
It is a known problem that the tsdz2 controller is extremely sensitive for the right magnet distance.
As said, the latest OSF does some valuation of the speedsensor signals to prevent wrong calcutation of the distance.


"Added a filter to the speed sensor to avoid problems of adjusting the distance from the magnet."

Good insight !!! ... from what you said it would seem that the best way to calculate distance is from the speed and time value as the speed part is working, then everything would be properly matched with no strange discrepancies ... a simple bit of math. I learned a lot today.

 

[nbrus27]

In case anyone is wondering ... I moved my speed sensor further along the chain stay so it was just overlapping the magnet on the wheel spoke. I went on a cycle ride today and recorded mileage was accurate, so this issue is resolved ... big thanks to Elinx for helping me out.

 

[Elinx]

... I went on a cycle ride today and recorded mileage was accurate, so this issue is resolved ... ...

:thumb:
Sometimes there is an easy solution, if you know which one :wink: