Trek Fuel EXe 500wh Battery Project

[rzr]

Kinda surprised as the charger is 4A and the cells handle 1.7A continuous charging

? charger charges full battery pack, not single cell.
They are serial and parallel connected.
Theoretically there is enough power to charge one cell with 2A (which is around .6C)

 

[Pakere]

I'll continue 'discharge' the battery tomorrow

 

[Psythepie]

I am hoping to be able to help with the discharged value tomorrow but the weather here is making riding unpleasant at the moment & my ride may be cancelled. As soon as I can help I will.

 

[MamboN5]

Thanks !!

 

[shure2]

The truth is that I have not tried it but at 10% the bike automatically goes into ECO or some kind of limited mode. So if it were as you say, which is very possible, a solution would have to be found.
If with the original battery the dead battery voltage is close to 3v per cell, the bike is almost certainly running according to calculations and not voltage readings.

I rode down to 3% and the power was down, but I presumed that was due to lower capacity available, or voltage sag. You are probably correct about an Eco mode though. Hopefully you can work it out.

 

[MamboN5]

I rode down to 3% and the power was down, but I presumed that was due to lower capacity available, or voltage sag. You are probably correct about an Eco mode though. Hopefully you can work it out.

have you measured the voltage? I need to compare the voltage data with the spent battery.

 

[Mteam]

I rode down to 3% and the power was down, but I presumed that was due to lower capacity available, or voltage sag. You are probably correct about an Eco mode though. Hopefully you can work it out.

the TQ manual states that at 10% battery and lower, the motor limits power to 110w max.

have you measured the voltage? I need to compare the voltage data with the spent battery.

If nobody else replies, I'll do this next time I ride the bike, the problem is that the weather is terrible at the moment, so no opportunity to ride.

 

[MamboN5]

the TQ manual states that at 10% battery and lower, the motor limits power to 110w max.


If nobody else replies, I'll do this next time I ride the bike, the problem is that the weather is terrible at the moment, so no opportunity to ride.

Thank you very much mate.
Yes, I know how the bike works at 10% battery, the question is whether the arrival at that point is done with a battery voltage measurement or with an algorithm that calculates how much battery you have spent.
That is why it is important to compare the voltage of my discharged battery with the original one, something that I could have done before disassembling it, but that possibility did not even cross my mind.

 

[Mteam]

Thank you very much mate.
Yes, I know how the bike works at 10% battery, the question is whether the arrival at that point is done with a battery voltage measurement or with an algorithm that calculates how much battery you have spent.
That is why it is important to compare the voltage of my discharged battery with the original one, something that I could have done before disassembling it, but that possibility did not even cross my mind.

yep, understand, that part of my message was just replying to shure2 about the motor/battery behaviour at 10% and under.

regarding how the battery percentage is derived - just as a point of discussion - surely the only reliable method for the BMS to determine state of charge would be by measuring voltage - what advantage is there for TQ in using an algorithm using current draw (or similar) to calculate state of charge? I'm just interested in why a manufacturer would choose either approach.

 

[MamboN5]

Thank you very much mate.
Yes, I know how the bike works at 10% battery, the question is whether the arrival at that point is done with a battery voltage measurement or with an algorithm that calculates how much battery you have spent.
That is why it is important to compare the voltage of my discharged battery with the original one, something that I could have done before disassembling it, but that possibility did not even cross my mind.

I can't think of many reasons to do it this way, but from the moment I connected the BMS to the new battery and the bike gave 100% when the cells were at 50%, it's clear that it doesn't work with voltage readings alone.
There is a memory that stores the state of the battery when disconnecting the bike, why do that if you can read the voltage and give the real state?
Maybe the balancer makes the battery level drop between uses and they want to hide it?
Don´t know.
Battery voltage varies as a function of engine load, perhaps you want to avoid constant voltage readings that should also be at times when there is no load and that's why you calculate and don't read?
Don't know.
The TQ system has several things that are out of the norm, we will have to investigate.
In the battery that I made for my giant reign 2022 that worked perfectly with a BMS Daly, the bike gave the remaining battery by means of 5 leds in the frame, with my battery it also gave it but unlike the original when pedaling it always lowered a led that recovered when the load was released.
Sometimes things that seem simple are not.

 

[Mteam]

I can't think of many reasons to do it this way, but from the moment I connected the BMS to the new battery and the bike gave 100% when the cells were at 50%, it's clear that it doesn't work with voltage readings alone.
There is a memory that stores the state of the battery when disconnecting the bike, why do that if you can read the voltage and give the real state?
Maybe the balancer makes the battery level drop between uses and they want to hide it?
Don´t know.
Battery voltage varies as a function of engine load, perhaps you want to avoid constant voltage readings that should also be at times when there is no load and that's why you calculate and don't read?
Don't know.
The TQ system has several things that are out of the norm, we will have to investigate.
In the battery that I made for my giant reign 2022 that worked perfectly with a BMS Daly, the bike gave the remaining battery by means of 5 leds in the frame, with my battery it also gave it but unlike the original when pedaling it always lowered a led that recovered when the load was released.
Sometimes things that seem simple are not.

the TQ manual does say that on very first use of the bike the motor will cut out at a displayed 15% of charge and will not go down to 0%, but after this first cycle, it will correctly show 100% down to 0%. I thought this was odd, but presumably its some kind of learning process for the BMS or display , so maybe there is a way of resetting the system to make it repeat this process?

 

[Pakere]

Here are my results. Hope it helps.

 

[MamboN5]

Here are my results. Hope it helps.

View attachment 104424 View attachment 104425

It helps and a lot, my reading at 1% battery was 49.72 v.
Your reading at 10% of 43.85v which equals 3.13v per cell is logical.

Thank you!!

 

[Pakere]

It helps and a lot, my reading at 1% battery was 49.72 v.
Your reading at 10% of 43.85v which equals 3.13v per cell is logical.

Thank you!!

One more thing: it was my first ride, so the battery level changed by 10% at a time on the display and never went below 10%. But the battery more or less was empty, the displayed blinked, there was no(or very very little) assist by the motor .
So maybe it's not the most accurate measurement, but I can ride from now on.

 

[MamboN5]

One more thing: it was my first ride, so the battery level changed by 10% at a time on the display and never went below 10%. But the battery more or less was empty, the displayed blinked, there was no(or very very little) assist by the motor .
So maybe it's not the most accurate measurement, but I can ride from now on.

Be that as it may, it is the only reading that I have with a value close to 0% battery.

 

[Psythepie]

Unfortunately I won't be riding tonight as it's currently heavy rain again & the trails are rivers so will have to wait until the weekend for a low reading, sorry. I did just remeasure my battery & it has dropped from 58.1Vdc when it had recently finished charging to 57.9Vdc by just being sat for 24hrs in a cold'ish room.

 

[Mteam]

I know you already have a lot of data for the 100% SoC, but I figure more data cant hurt and I was in the garage anyway....

mine has been sat for a few days at 100% SoC and is showing 57.4v

 

[MamboN5]

the TQ manual does say that on very first use of the bike the motor will cut out at a displayed 15% of charge and will not go down to 0%, but after this first cycle, it will correctly show 100% down to 0%. I thought this was odd, but presumably its some kind of learning process for the BMS or display , so maybe there is a way of resetting the system to make it repeat this process?

I have not found this information in the manual... do you have a link or screenshot?

 

[Mteam]

I have not found this information in the manual... do you have a link or screenshot?

heres a screenshot, its in the trek FAQ - not the manual as I stated up there : here (FAQ - Fuel EXe | Trek Bikes (GB))

Have you tried doing a factory reset on the display? thereis some combination of buttons to hold down to reset it, maybe this will repeat the learning process? I dontknow because I have never reset it.

 

[MamboN5]

heres a screenshot, its in the trek FAQ here (FAQ - Fuel EXe | Trek Bikes (GB))

View attachment 104524Have you tried doing a factory reset on the display? thereis some combination of buttons to hold down to reset it, maybe this will repeat the learning process? I dontknow because I have never reset it.

Yes, I did the reset yesterday, we'll try this afternoon to see if it worked.

 

[MamboN5]

It is confirmed that the battery percentage that appears on the screen is a calculation and not a voltage reading, today when I reached 1% I continued until the battery ran out and I have done 17 km and 600 meters of unevenness with that 1%.
Very good news because autonomy improves according to my initial calculations, around +40%, now we have to see how to solve this little big handicap.

 

[Mteam]

It is confirmed that the battery percentage that appears on the screen is a calculation and not a voltage reading, today when I reached 1% I continued until the battery ran out and I have done 17 km and 600 meters of unevenness with that 1%.
Very good news because autonomy improves according to my initial calculations, around +40%, now we have to see how to solve this little big handicap.

Now that you have run the battery down to a real 0% state of charge, perhaps the BMS will learn the full correct range when you charge it up from this real 0% level? Maybe?

 

[MamboN5]

Now that you have run the battery down to a real 0% state of charge, perhaps the BMS will learn the full correct range when you charge it up from this real 0% level? Maybe?

Let's hope so, tomorrow we will leave doubts

 

[Mteam]

Let's hope so, tomorrow we will leave doubts

Did you make any progress in getting the bms recalibrated?

 

[Psythepie]

Finally got out for a ride (& a couple of laps of the village to knock off those extra percentages). Temperature is 8 degrees Celsius/46.5 farenheit. Hope this is still useful to you @MamboN5?

 

[MamboN5]

The reset has not worked and completely draining the battery seems to have had an influence on the data shown the next day but little, I still do not get data close to reality. I'm still investigating how it works but it seems that the theoretical capacity of the original battery plays an important role in the calculations.
On Friday I took the bike to the store and we connected it to see if the BMS could be reset and there is no option that allows us to do anything like that. A complete test was done and everything came out ok, battery status 100%.

View attachment 104628
Finally got out for a ride (& a couple of laps of the village to knock off those extra percentages). Temperature is 8 degrees Celsius/46.5 farenheit. Hope this is still useful to you @MamboN5?

yes!! Perfect!! Thanks!!

 

[Mteam]

The reset has not worked and completely draining the battery seems to have had an influence on the data shown the next day but little, I still do not get data close to reality. I'm still investigating how it works but it seems that the theoretical capacity of the original battery plays an important role in the calculations.

yes!! Perfect!! Thanks!!

I think somehow you need to reset the bms so it goes through that first learning process again that it's described in the trek faq,but I have no idea how you would force it to reset,I guess you have already removed any voltage from the bms when you replaced the cells?

Maybe there is a jumper on the bms circuit board to erase any persistent memory store?

 

[Zimmerframe]

I think all the cells need to be either fully drained or fully charged also. (Unless you've already done this, in which case ignore my ramblings ) Therefore you may need to break some of your solder connections temporarily to make sure everything starts off the same.

As said, you had some in one state and some in another - but with a wide variation, this maybe outside of the scope of any balancing.

Imagine you have two jugs of water .. with a float in each connected to the same float arm. One was already half full .. you add half a litre to each, the float in one says they're full - but they're not both full.

 

[OldnSlow]

I think all the cells need to be either fully drained or fully charged also. (Unless you've already done this, in which case ignore my ramblings ) Therefore you may need to break some of your solder connections temporarily to make sure everything starts off the same.

As said, you had some in one state and some in another - but with a wide variation, this maybe outside of the scope of any balancing.

Imagine you have two jugs of water .. with a float in each connected to the same float arm. One was already half full .. you add half a litre to each, the float in one says they're full - but they're not both full.

Perhaps you could run that analogy using beer Zimmer. The logic seems sound but I'm having difficulty visualising.

 

[Zimmerframe]

Perhaps you could run that analogy using beer Zimmer. The logic seems sound but I'm having difficulty visualising.

Ok, well as that seems rather apt ..

I'm just drinking some beer .. and one tasted completely different ..

I then noticed it was completely different .. I thought they had a brewing problem, but it seems they have a packing problem.

Ok, lets try ..

Like @MamboN5 you're sitting on the beach surrounded by beautiful women and handsome guys - obviously you're easily the most handsome so you're not threatened or even worried in the slightest by anyone else there. This is a typical evening for anyone Spanish before heading inside to eat Paelle and work on your battery experiment.

At this point I'm rather unreasonably assuming @MamboN5 is a man and not a lady. However if MB5 is a lady, 95% of men of the forum will be booking flights to Spain to meet the wise battery lady with the Sexy Trek bike.

So lets say you have a 12 pack of beer .. We won't say San Miguel, because that's crap- but if it's hot and you're sweating like hell, it probably tastes ok .. ;-)

Now picture that 4 of those beers are the San Miguel Extra Strong (ok, I contradicted myself) .

You're sitting with 2 friends and dish out the beers, one of your friends ends up way more happy than everyone else and sooner ... This is beer inbalance.

You need to even things out so everyone gets beer charged at the same rate, or someone will fall over before the others and that's when the party stops.

Therefore, you need to go back to the off licence/beer shop/man with a donkey on the beach and get the same beer, or open all the beers and pour them in a tub and mix them all up ..

Sorry, that's all I can do on short notice, my food needs taking out of the oven.