Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[togo]

> I've been meaning to finish my program to analyze logs and come up with a more accurate SoC to voltage look-up table that takes into account temperature and cell degradation, but I got swamped at work and I always feel like I'm reinventing the wheel when this has been done so many times.

I don't know if anyone has done it rigorously and released the results.

[princec]

The one time I had a ride on an Energica, it read 60% SOC in the showroom, only to plummet to 0% and conk out exactly 1 mile from the shop. Seems that voltage measurement isn't entirely reliable technology yet. Maybe it depends on having a lot more cells in parallel to get a better sample.

Cas

[talon]

I figured a data-driven model might be helpful. There's so much I wanted to do with it, including what Teslanomics did for releasing their analysis of how certain models of Tesla batteries were holding up over time. It would be so cool to interface with Hassler's page and scrape anonymous data (with users' consent) and see how Zeros across the board are doing. Start showing trends. (surely Zero has something like this on their support mail, aside from a decoder)

I did a small example long ago in Java for [my] ease in porting to Android. I thought it would be cool to be able to visualize logs right on the device that pulls them from the bike. I also wanted to look into doing it from the ground up in Python. It's been many years since I put on my programming hat, and everything is so easy now I feel like there is always an easier way than what I am doing.

[princec]

Aha... I'm a Java programmer as it happens (https://www.puppygames.net) ... been coding for 40 years now.

Cas

[MVetter]

The one time I had a ride on an Energica, it read 60% SOC in the showroom, only to plummet to 0% and conk out exactly 1 mile from the shop. Seems that voltage measurement isn't entirely reliable technology yet.

Really? What voltage was the bike reading?

[princec]

Didn't check that at the time as it has a "remaining range" reading which read "60 miles".

Cas

[MVetter]

So how are you blaming the voltage for being inaccurate? The range estimates are based off the last 10 miles. If someone drove around at 10 miles an hour, for whatever goddamn reason, for an entire hour it would show a ridiculous range which will quickly wear off as soon as the bike sees highway speeds. I'm guessing you also didn't look at the SoC.

[TheRan]

He said it also had a 60% SoC though, no way that should be dropping to 0% in just a mile of riding.

[MVetter]

At first he said 60% SoC and blamed the voltage for being unreliable. Then he said 60 miles and said he had no idea what the voltage was. I am merely clarifying.

[togo]

I figured a data-driven model might be helpful. There's so much I wanted to do with it, including what Teslanomics did for releasing their analysis of how certain models of Tesla batteries were holding up over time. It would be so cool to interface with Hassler's page and scrape anonymous data (with users' consent) and see how Zeros across the board are doing. Start showing trends. (surely Zero has something like this on their support mail, aside from a decoder)

I did a small example long ago in Java for [my] ease in porting to Android. I thought it would be cool to be able to visualize logs right on the device that pulls them from the bike. I also wanted to look into doing it from the ground up in Python. It's been many years since I put on my programming hat, and everything is so easy now I feel like there is always an easier way than what I am doing.

Absolutely

[princec]

Sorry - I actually meant it had 60% SOC displayed - which the dealer said equated to about 60 miles. Don't know the voltage (and frankly don't care).

Cas

[MostlyBonkers]

Sorry - I actually meant it had 60% SOC displayed - which the dealer said equated to about 60 miles. Don't know the voltage (and frankly don't care).

Cas

Did the dealer mention how long it had been sat in the shop for? I'm guessing that it hadn't been ridden for a long time, along with you discovering a bug in the software.  Do Energicas drain their batteries when parked? My Zero is excellent at holding its charge for months and only dropping a   couple of percent.  However, I always fully charge it before riding after a stagnant period.

[princec]

We pieced together what had happened back at the showroom. The trip meter on the bike read, very suspicously, 108 miles. It turned out it had been taken out for a long demo ride, then left with no charge in for some time and not plugged in (may have been for weeks). Somehow when switched back on, the system reckoned it had 60% SOC based on the voltage reading. It looks like the battery voltage must "collect" for a while while it's idle, resulting in a small surge of power for a little while, fooling the calculation. Were I an Energica engineer at this point I'd be thinking, it should probably maybe also take the last trip duration since charging into account to limit its estimates - in this case such a formula would have yielded a 0% SOC as there's no way it could have done 108 miles without draining pretty much the entire battery.

If the bike weren't unfortunately so bloody heavy (which is what you discover when you have to get it out of the main road in a hurry when it conks out...) that would have cost them the sale... but yeah it is too heavy, I could barely move it.

Cas

[MostlyBonkers]

If the bike weren't unfortunately so bloody heavy (which is what you discover when you have to get it out of the main road in a hurry when it conks out...) that would have cost them the sale... but yeah it is too heavy, I could barely move it.

Cas

Interesting, Cas.

I’m a little confused by your last paragraph. Did the weight of the bike cost them the sale?

[princec]

Chances are I wouldn't have thought too much about the weight if it hadn't conked out, but as it is, the bike doing the exact one thing that everyone is paranoid about electric vehicles doing on its test ride was enough to make me decide "no". The SR/F was far easier to haul about, but if that had broken down in the same manner it would also be getting a "no". As it is I'm just far too poor to afford the SR/F anyway Still saving pennies, searching down back of sofa, etc

Cas