Cycle Analyst V3 preview and first beta release

[hjns]

I only care about total voltage while driving. I have set lvc at a pack voltage that is similar to an average cell level of 3.4v. This works for me.

 

[amberwolf]

I think the only real safe battery gauge to run lipo is cell voltage anyway, and we all know that never below 3.6. Would just be easier in my opinion to see these numbers as there would be no confusion ever in your conversion.

If you were actually monitoring the cells themselves, and the voltage displayed was that of the lowest cell, then this would be true.

But unless you are doing that, and not making an artificial average of cell voltages, it could be dangerous to do this kind of display, and at the least is not any more useful than the existing display of pack voltage.

What I did on mine was calculate the hard LVC, the desired LVC (where I'd rather not run the pack below), and charging HVC for my pack, and note it on the pack itself, and I just remember what my desired LVC is to compare to the display. The hard LVC for pack voltage is set in the CA, and then cell-level emergency-cutoff LVC at the pack itself will be taken care of by Methods' boards (not currently installed but will be after I find all the parts in the piles of stuff after the (unrelated to ebikes) house fire).

 

[izeman]

Pack value has EXACTLY the same meaning as cell voltage. No one expects to know if the pack is still in balance or of there is a dull cell by reading the pack voltage. Neither does anyone by reading the average cell voltage.
It's very simple: for some people it's just easier to remember which pack level is seen as 'good' and some like to know it on cell level base.
Eg. I have a 12s lipo pack, and see 48v it means nothing to me. But 4.0v means a lot for me. That doesn't mean that there isn't one cell below 2.5v in the pack. But therefore I haver my BMS to take care of

 

[hjns]

Having experience with bad cells, pack voltage to me is NOT the same as individual cell voltage. I don't worry about discharge during driving. There is no danger in overdischarging the cells. The danger comes when CHARGING overdischarged cells, or when BULK OVERCHARGING normal cells due to bad cells keeping total voltage down.

Therefore, instead of monitoring cell voltage during a ride, it makes much more sense to do individual cell voltage checking prior to charging. If there is any clue of bad cells, the only trustworthy way to find out is checking individual cell voltage prior to charging and directly after charging and making a nice discharge curve using a good RC balancing charger/discharger. No voltage monitoring while driving will give you that info, and during driving you should pay attention to the road anyway.....

Oh, and I am the BMS....

 

[izeman]

I totally agree with you, but what you say has nothing to do with our request/wish That's what I was trying to explain: it's not about LVC or HVC, and not about watching (individual) cell levels. It's about how different people prefer different views of the same thing.

 

[EdwardNY]

The problem I have with total voltage is that, yes I can remember that with 18s, 3.7 is 66.6 and 4.15 is 74.7.

But for me it is so much easier to know my state of charge with a 4.0 volts per cell then 72 volts total.

When you start pushing towards the end, like 3.65, 3.6. It is much easier to see 3.6 then 64.8. Since I do not push my pack so low I usually am not sure where my per cell voltage is.

I guess the problem is I like to think in terms of individual cell voltage. Especially since I use a RC hobby charger to parallel charge my packs and also have been using 12s, 18s, and even 24s. Since I just bought my first cycle analyst I am honestly surprised that it does not have the option to display average cell voltage.

 

[amberwolf]

I see your points.

I only think it could be dangerous as a display of voltage because there *will* be people that will select that option and somehow believe they are actually looking at the real per-cell voltage, and not an artificial average. How they then respond to that display, especially for those running cheap RC LiPo with no actual per-cell monitoring, will determine if they get to later star in that year's "dramatic thermal event collection" thread.

for those (most people) that do understand what that display actually means, it could indeed be a useful way of displaying the data.

I just know from personal experience with other people and other things that there are going to be people that refuse to understand no matter how many times it's explained to them, or in what way, and will end up with a problem because of it.

 

[Merlin]

Edward, i can follow you...but i think your on status: beginning using cav3
i come from RC section too....after some more or less "Bike-Cycles" you know the voltages very well.

i start from 75v (4,17v)and stop discharging at 64v (3,55v)

you will also know how your range differs using heavy throttles

the most important part of the fantastic CA is that you can push the throttle max and the CA will reduce really smooth the Power that you
never drop under LVC (like 64v)...
so you can ride your bike till it stops with 0 power and have great 3,55v on each cell.

(Iam sure on CAv4 we will have an external "Modul" to watch Cell Voltages up to 30s)

 

[hjns]

OK, Edward, I think I get it.

I think the problem you describe is really only an issue when you change packs often. For one pack, voltage range stays the same. For my 20S lipo config, looking at unloaded voltages, it means anything above 80v is fully charged (4.00v cell level), anything above 74v (3.70v) means I don't have to worry, and everything below 70v (3.50v) is a danger zone for me.

As you rightly say, computations get more complicated once you need to convert weird configurations like 13S and 17S. Then again, I would assume that after having calculated the 3 most important values more than 5 times, you would know them by heart as well. I don't need a CA for that.

If you really want cell level voltage on the go, why not add a couple of cell logs to the balancing wires with a beeper or flash? I once had 4 celllogs attached to the battery while driving. I never once looked at it.... so I removed it again.

I need my CA for the "really important" things, like power throttle and thermal loop feedback. Those two things are really not related to my batteries, but are essential in me being able to handle my powerful bike.

 

[EdwardNY]

OK, Edward, I think I get it.

I think the problem you describe is really only an issue when you change packs often. For one pack, voltage range stays the same. For my 20S lipo config, looking at unloaded voltages, it means anything above 80v is fully charged (4.00v cell level), anything above 74v (3.70v) means I don't have to worry, and everything below 70v (3.50v) is a danger zone for me.

As you rightly say, computations get more complicated once you need to convert weird configurations like 13S and 17S. Then again, I would assume that after having calculated the 3 most important values more than 5 times, you would know them by heart as well. I don't need a CA for that.

If you really want cell level voltage on the go, why not add a couple of cell logs to the balancing wires with a beeper or flash? I once had 4 celllogs attached to the battery while driving. I never once looked at it.... so I removed it again.

I need my CA for the "really important" things, like power throttle and thermal loop feedback. Those two things are really not related to my batteries, but are essential in me being able to handle my powerful bike.

hjns, I guess if I used the same voltage long enough and got used to reading total pack voltage it would be non issue. But somehow I think I'm always set on converting everything to individual cell voltages. I guess its like when driving with KPH when I grew up with MPH, I like to always convert it to MPH. Same with using Celsius temp. I always like to convert to Fahrenheit no matter how long I am in Europe and reading Celsius.

I just think it should be an added option to convert it to how you like. I am not tyring to change anything just would be nice to go into setup preferences and change it to display average cell voltage in the main window. Of course put a warning in the instructions saying this is only an average cell voltage and individual cell voltages could differ.

 

[litespeed]

So I decided to get a V3 for my (to me!) monster power DH Team/24s lipo powered H4065. After riding the piss out of it the other day when it was hot and I heated the motor up for the first time. Question is should I get the programing cable? Not sure if I really need it or not. I have them for all my R/C things, lyen controller and all my chargers but don't use most of them.

Thoughts?

Tom

 

[litespeed]

Couple other questions,

Did anyone figure out 100% how to make the Clyte H4065 10k transmister (Duh?) work with the V3?

& Should I buy from Grin Cyclery if I live in the USA?

 

[John in CR]

HJNS,

I think there's a risk during discharge too, not just during charging. If a cell runs dry and you continue to run current through it from the other batteries, then it's internal resistance will shoot up and create enough heat to start a fire. Even though it doesn't have electric energy to rapidly accelerate the fire, the adjacent cells do. The risk is probably lower than with a pack during charging, since the battery contains less energy at the time, but I believe there's still significant risk.

 

[NeilP]

Not always, as I have found out recently.
Sometimes cells can die and have near enough zero internal resistance.

My 20s4p pack became a 19a 4 p pack a week or so back, and I did not notice
Had done a balance charge only a few days earlier and all well.
Got a bit slack over past year, not checking balance before each bulk charge.
So the result; 19s being charged multiple times (10 plus ) to 83.4 volts. Some cells going to 4.4 or more. Lucky no fire and only some puffing, 15to 20 mm across 10 cells .

But no high internal resistance on bad cells, and no noticable heating. This was with peaks of 120 amps and 40 to 50 continious, every day for a week

 

[hjns]

HJNS,

I think there's a risk during discharge too, not just during charging. If a cell runs dry and you continue to run current through it from the other batteries, then it's internal resistance will shoot up and create enough heat to start a fire. Even though it doesn't have electric energy to rapidly accelerate the fire, the adjacent cells do. The risk is probably lower than with a pack during charging, since the battery contains less energy at the time, but I believe there's still significant risk.

OK, I did not know that, so I learn something new every day. Though, a vid would be nice

So, how would we design an experiment like that, and would we even consider putting a BMS-like system into the CAv99? Or would we just recommend Methods' LiPo saver in combination with the CAv3?

 

[Architectonic]

By the way, is the v3+Thun considered a legit way for lycra wearing cyclists to measure their power output?

Assuming they can deal with the hole for the wiring, square taper.

 

[shorza]

I am not sure if this has been answered or not, but with the latest release i am getting slight cut-outs upon acceleration.

Everything is ok when accelerating slowly and when cruising.

Speed limit is set to 500 kph and ramp up speed is at minimum.

 

[teklektik]

I am not sure if this has been answered or not, but with the latest release i am getting slight cut-outs upon acceleration.

Everything is ok when accelerating slowly and when cruising.

Speed limit is set to 500 kph and ramp up speed is at minimum.

A quick look at your bike suggests that you may be experiencing problems with the DSGain setting due to rapid acceleration. This can happen with more powerful bikes. Please review section 5.2 of the Unofficial User Guide and see if that helps.

 

[Luge_a_foin]

Hello,
First of all congratulations for all your work and posts! Nevertheless I've got an issue with the CA V3 and may be you can help me.
My conf:
CA V3 (last firmware available) + speedometer
Lyen Sensorless 12 Fets with direct plugin to CA.
Here is my issue:
I've decided to reroute the throttle to the CA. I've read carefully the teklektik's UM. In particular the annex so I've derived the CA's throttle output to the controler's input (as I suppose there is possibly a diode in the controller). I use the Pass-Trough conf. I've checked voltage mapping : it's OK with at least an empty load. 4.3V output (mesured with a voltmeter) when full throttle which fits with the original one.
But my issue is very very strange for my understanding. I've got a speed limit at around 30km/h. I've removed all limits in CA & checked the Limits flags in the associated display : No caps and however the displayed voltage output are 4.2 which should generate a full speed.
However my speed sensor is on the front wheel and I've made the test with only the back wheel lifted. The wheel cannot reach full speed. I've reflashed the controller. But what is very weird is when I reuse the throttle without going through the CA, it reaches again full speed.
I don't know If I've been clear but I must miss something but what? Especially if the displayed voltage in the CA fits with outputs.
So any help is welcome...

p.s. there is a small bug I think in the initial conditions of the pass trough configuration : if I use the throttle as soon as the CA is turned on the full throttle voltage increases progressively during several seconds and the full throttle is only available after maybe 1 mn. After that it works correctly. May be it is because I've deactivated the limits with some 0 values as indicated in the UM. Nevertheless it's a kind of bug.

 

[teklektik]

You seem to have gone through the Guide, but here's a few questions to help understand the situation:

...I use the Pass-Trough conf. I've checked voltage mapping : it's OK with at least an empty load. 4.3V output (mesured with a voltmeter) when full throttle which fits with the original one.

Please help clarify: Are you saying that the CA ThrOut is 4.3v as measured with a multimeter when not connected to controller?
What is displayed on the Diagnostic Screen as the OUT voltage at Wide Open Throttle when not connected to the controller?

... however the displayed voltage output are 4.2 which should generate a full speed.

Again to clarify the situation: Are you saying that when connected to the controller the diagnostic screen shows 4.2v?
Is this the setting you made from the procedure in step 5 on p. 17 of the Guide ("Adjust ThrO->MinOut and ThrO->MaxOut...")?

So - this is a little odd. Were you able to get the motor to achieve full speed when making the p.17 ThrO->MaxOut adjustment? If so, did it work during Setup but not after ThrO->MaxOut was configured?

Does the bike actually achieve full speed under load (not on the stand)?

...if I use the throttle as soon as the CA is turned on the full throttle voltage increases progressively during several seconds and the full throttle is only available after maybe 1 mn. After that it works correctly. May be it is because I've deactivated the limits with some 0 values as indicated in the UM.

This is very strange. This is pretty unusual behavior and I suspect there is a configuration issue in play, not the firmware. Please ensure that AuxPot, Temp, PAS, and Trq are all deactivated and that ramping is set to the defaults. You say you have set some limits to 0 as indicated in the Guide. You actually should have all the settings at the default to get the throttle working - there are so many possible interactions that it is important to follow the setup instructions carefully to avoid introducing other unwanted behavior.

Please verify that you have not set IntSGain=0 - if you do this, the throttle will respond very slowly and the bike may not achieve full speed. The lowest you want to go with that is IntSGain=1 (p.21).

This second issue is likely tied somehow to the first...

 

[shorza]

I am not sure if this has been answered or not, but with the latest release i am getting slight cut-outs upon acceleration.

Everything is ok when accelerating slowly and when cruising.

Speed limit is set to 500 kph and ramp up speed is at minimum.

A quick look at your bike suggests that you may be experiencing problems with the DSGain setting due to rapid acceleration. This can happen with more powerful bikes. Please review section 5.2 of the Unofficial User Guide and see if that helps.

You were spot on, thanks for the help.

 

[Luge_a_foin]

You seem to have gone through the Guide, but here's a few questions to help understand the situation:

...I use the Pass-Trough conf. I've checked voltage mapping : it's OK with at least an empty load. 4.3V output (mesured with a voltmeter) when full throttle which fits with the original one.

Please help clarify: Are you saying that the CA ThrOut is 4.3v as measured with a multimeter when not connected to controller?
What is displayed on the Diagnostic Screen as the OUT voltage at Wide Open Throttle when not connected to the controller?

... however the displayed voltage output are 4.2 which should generate a full speed.

Again to clarify the situation: Are you saying that when connected to the controller the diagnostic screen shows 4.2v?
Is this the setting you made from the procedure in step 5 on p. 17 of the Guide ("Adjust ThrO->MinOut and ThrO->MaxOut...")?

So - this is a little odd. Were you able to get the motor to achieve full speed when making the p.17 ThrO->MaxOut adjustment? If so, did it work during Setup but not after ThrO->MaxOut was configured?

Does the bike actually achieve full speed under load (not on the stand)?

...if I use the throttle as soon as the CA is turned on the full throttle voltage increases progressively during several seconds and the full throttle is only available after maybe 1 mn. After that it works correctly. May be it is because I've deactivated the limits with some 0 values as indicated in the UM.

This is very strange. This is pretty unusual behavior and I suspect there is a configuration issue in play, not the firmware. Please ensure that AuxPot, Temp, PAS, and Trq are all deactivated and that ramping is set to the defaults. You say you have set some limits to 0 as indicated in the Guide. You actually should have all the settings at the default to get the throttle working - there are so many possible interactions that it is important to follow the setup instructions carefully to avoid introducing other unwanted behavior.

Please verify that you have not set IntSGain=0 - if you do this, the throttle will respond very slowly and the bike may not achieve full speed. The lowest you want to go with that is IntSGain=1 (p.21).

This second issue is likely tied somehow to the first...

Thanks for the quick answer. I will do some new tests tomorrow. But as a summary:
-The motor reach full speed with direct throttle,
-The CA throttle output is correct in open loop (consistent between expected values, displayed values on CA and outputs measured values with a voltmeter),
-There is a limit in motor speed when use of CA throttle (even with a speed = 0 for the CA : speedometer still)
According to the behavior I've observed it seems to me that the limit comes from the controller but I cannot figure out how it would make the difference between 4V coming from the CA & 4V coming directly from the throttle. To be continued...

 

[panurge]

Hello guys
Since I'm mulling about ideas and solutions for my 2nd build and a CAV3 is on the list (I'm happy using the beta LRC version for my RC bike, actually) I have 2 question for the CA gurus (Justin, Teklektik etc.).

The first is about how to use (if possible) a force sensing resistor like this instead of my actual Magura pot i

as a throttle with the CAV3.

The second is a bit complex for me to explain, but that is:
I plan to use a BB as jackshaft where motor imput and pedal imput combine together and go with a common final stage to the wheel, so that means 2 BB used on the bike, the lower for cranks only and the upper as jackshaft with 2 imputs and 1 common output.
The question: Is in any way, at least in theory, convenient to have a sensored thun BB used as jackshaft to measure the motor torque so that the CA could handle these datas and accordingly regulate the pulses? this jackshaft-BB would see 300rpm max.

Another simple way to explain what I mean is: forget the 2 BB and figure out an hard tail e-bike that has not pedals, but where the BB were used as jackshaft for the motor trasmission, there is any convenience to use a thun coupled with the CA measuring Motor torque at that stage and let the CA handle it?

Hope some of you can understand

 

[John in CR]

Has anyone else tried dual controllers with the CA3 throttle by wire? It didn't work well for me with EB2xx Infineon type controllers. It worked but the controllers seemed to fight over the throttle signal and not share equally. I ran it for a while and I seem to have messed something up in the CA3's throttle circuitry in the process. I was able to get home, but the throttle became like an on/off switch, and with 22kw on tap with a 100mph 137lb ebike, a throttle that's an on/off switch is Liveforphysics kind of excitement, not mine.

John

 

[Luge_a_foin]

You seem to have gone through the Guide, but here's a few questions to help understand the situation:

...I use the Pass-Trough conf. I've checked voltage mapping : it's OK with at least an empty load. 4.3V output (mesured with a voltmeter) when full throttle which fits with the original one.

Please help clarify: Are you saying that the CA ThrOut is 4.3v as measured with a multimeter when not connected to controller?
What is displayed on the Diagnostic Screen as the OUT voltage at Wide Open Throttle when not connected to the controller?

... however the displayed voltage output are 4.2 which should generate a full speed.

Again to clarify the situation: Are you saying that when connected to the controller the diagnostic screen shows 4.2v?
Is this the setting you made from the procedure in step 5 on p. 17 of the Guide ("Adjust ThrO->MinOut and ThrO->MaxOut...")?

So - this is a little odd. Were you able to get the motor to achieve full speed when making the p.17 ThrO->MaxOut adjustment? If so, did it work during Setup but not after ThrO->MaxOut was configured?

Does the bike actually achieve full speed under load (not on the stand)?

...if I use the throttle as soon as the CA is turned on the full throttle voltage increases progressively during several seconds and the full throttle is only available after maybe 1 mn. After that it works correctly. May be it is because I've deactivated the limits with some 0 values as indicated in the UM.

This is very strange. This is pretty unusual behavior and I suspect there is a configuration issue in play, not the firmware. Please ensure that AuxPot, Temp, PAS, and Trq are all deactivated and that ramping is set to the defaults. You say you have set some limits to 0 as indicated in the Guide. You actually should have all the settings at the default to get the throttle working - there are so many possible interactions that it is important to follow the setup instructions carefully to avoid introducing other unwanted behavior.

Please verify that you have not set IntSGain=0 - if you do this, the throttle will respond very slowly and the bike may not achieve full speed. The lowest you want to go with that is IntSGain=1 (p.21).

This second issue is likely tied somehow to the first...

I've done some investigations and I have a clue:
The CA output is OK in Open loop but not in close loop (bike still, speedometer still, back wheel lifted): the CA output on the debug display (4.2V) does not fit with the measured one (saturation at around 2.2V). As a reminder in this conf Measured Speed = 0, Consumption is very low (around 50W). Could this come from a throttle current draw of the controller that the CA is not able to supply? Because the only difference is that the Gnd & 5V of the throttle controller input are not connected in this conf. Should not the debug display output always fit with the measurement? weird, no?