CellLog 8 hacking

[EBJ]

ok, I successfully followed directions and got my cell-log to cut-out my throttle at set lvc.
I now have a question:

Is there any way to "gradually" cut-off the throttle so I dont go flying over the handle-bars when the thing kicks off ?
First time it turned the throttle off I almost ate it. I now ride nice and slow when I think it will happen.

lol.

Thanks.

[torqueon]

Think its better to use a loud alarm. rather than shut down the motor. The batteries are the least concern when crossing a busy street ! Set L V C a bit above spec. If the alarm goes off, go easy on throttle and pedal home.

[hillzofvalp]

Even still, u will still be able to move. Usually the l vc kicks off when voltage drop recedess.. So if you're light on the throttle u can make it through the intersection.

[amberwolf]

ok, I successfully followed directions and got my cell-log to cut-out my throttle at set lvc.
I now have a question:

Is there any way to "gradually" cut-off the throttle so I dont go flying over the handle-bars when the thing kicks off ?

Is your CL set to pull down the throttle signal line? Or does it engage an ebrake line?

If the former, then sure, you could setup a ramping circuit with either an op-amp or transistor, and a resistor and capacitor.

If the latter, then usually ebrakes are on or off.


Since it sounds like you ahve regen braking enable,d then depending on your controller's capabilities, you might be able to change the regen mode, if you don't normally use regen anyway.


However, a better solution is to have it turn off the controller's "keyswitch" or "ignition" line, which would just need either a FET, transistor, or relay to cut the power to it on command from the CL.

[Alan B]

Disabling the controller while in motion has on some occasions caused controller failures. Not to me, but it has been mentioned here before. Seems like a bad idea to pull the power while there is lots of energy in the motor, the FETs may do something unfortunate while the power decays slowly away.

Reminds me of Honda Outboards. The older ones shut down the motor when the oil got low. The newer ones sound a buzzer and let the skipper decide whether to burn out the motor or get swept into the rocks.

[amberwolf]

Ouch. I may have missed the controller failure threads caused by disabling it. I know that disconnecting power completely from them while in use can certainly result in bad things.

[hillzofvalp]

having a micro controller change the current limit a few minutes before LVC will give the user a warning that he needs to start thinking about less-risky riding, pedaling more, and utilizing battery power at the crucial moments. Takes $30-40 of parts. I suppose you could do it with the cell log.. or have you already? If you got it to cut out a throttle then you should be able to get it to cut out one or more of the shunts in your controller...

[fechter]

One way to do it would be to have the CellLog cut the throttle in half rather than fully off.

Pulling down the throttle signal through a diode/RC filter could provide a somewhat gradual limiting effect. With a large enough capacitor, it could be very gradual.

Usually if the LVC hits and the throttle drops, the battery voltage will immediately bounce back and clear the LVC. This will give you sort of a 'studder' that's hard to miss. I can see how a very high powered setup could be damaged by this. It might be better if once LVC hits, it latches into the reduced power mode instead of returning to full power.

[whatever]

I"ve ordered a couple of cell logs, my idea is to mount them on the handlebars with long balancing wires going to the pack at the rear of the bike.
I realise these longs wires will change the voltage read by the cell logs, can anyone see any big issues with this?
I guess it would be possible to lower ( increase?) the lvc/hvc alarms by same amount of voltage loss due to the long balancing wires in order to keep them accurate.
Has anyone already tried this?

[fechter]

The CellLog only draws a few milliamps, so the voltage drop in the wires won't be too much. If you knew the wire size and length, it would be possible to calculate the drop. My guess is it would only be a few millivolts, so not enough to worry about. Using heavier wire will reduce the drop.

[whatever]

that is excellent news, at last my cobalts will be relatively safe ( constant monitoring of all cells)

[hillzofvalp]

How would I use a multiplexer to change the cell groups the cell log checks? I know how it could work, but I'm unsure there is a multiplexer that would do it.. That way I wouldn't need 3 cell logs and 3 serial inputs on my arduino to monitor my pack.

[Samson]

I thought I understood how a Cell Log 8M connected to a battery pack (40 Volts) from reading the rather vague PDF manual but while connecting one using pins 1 and 9 I heard a spark and the cell log would not power up. The polarity was + on pin 9. The manual for the CD appears to show the full pack voltage connected between pins 1 and 9 so I do not know what caused this. The unit will power up with the USB power connected but when I opened it up it looks like a component near the pins has blown open. I have another unit but I am reluctant to try it until I understand the cause of the problem. Perhaps I misunderstand how it connects. The Cell log was new. Any ideas?

[Alan B]

Cell Log can only handle 8 cells.

[hillzofvalp]

maybe mwkeefer can come out of his mancave and share his findings on serial use

[fechter]

I thought I understood how a Cell Log 8M connected to a battery pack (40 Volts) from reading the rather vague PDF manual but while connecting one using pins 1 and 9 I heard a spark and the cell log would not power up. The polarity was + on pin 9. The manual for the CD appears to show the full pack voltage connected between pins 1 and 9 so I do not know what caused this. The unit will power up with the USB power connected but when I opened it up it looks like a component near the pins has blown open. I have another unit but I am reluctant to try it until I understand the cause of the problem. Perhaps I misunderstand how it connects. The Cell log was new. Any ideas?

Manual says maximum voltage is 43v, so seems like it should have been OK.
If you connect pins 1 and 9 only, it will alarm on the unconnected channels. If you connect to pins 1 and 2, it should read the pack voltage.
Not sure why you got a spark and it died. Reversed polarity would do it.

[dumbass]

I thought I understood how a Cell Log 8M connected to a battery pack (40 Volts) from reading the rather vague PDF manual but while connecting one using pins 1 and 9 I heard a spark and the cell log would not power up. The polarity was + on pin 9. The manual for the CD appears to show the full pack voltage connected between pins 1 and 9 so I do not know what caused this. The unit will power up with the USB power connected but when I opened it up it looks like a component near the pins has blown open. I have another unit but I am reluctant to try it until I understand the cause of the problem. Perhaps I misunderstand how it connects. The Cell log was new. Any ideas?

As stated by Alan B the max amount of cell you can connect are 8. IT sounds like you did things correct but just to be sure I go thouth how I always connected mine. I connect to 8 cells....the CellLog8 has 9 connectors numbered "0" through "7" and a "+" mark. I wire all my cells together then attach the wire from the "0" to the negitive of my first cell (this would be considered the negitive you would use to connect to your controller...or the neg you would use to messure the total pack voltage). Then connectt the #1 to the next cell's negitive and so on. the last connection should be the "+" wire from the CellLog8 and it is connected to the positive terminal of the last cell. This is where you would connect if you were checking the total voltage of the pack.

Note: If by chance the CellLog8 was active when you started connecting it will power-up when you connect the 3rd. wire (#0, #1 and #2). It will read the voltage of the first 2 or 3 cells. Most people that have a problem connecting the CellLog8 find that they connected it 180 degrees backward. I've worked with people on the phone on a few occasion helping them connect their CellLogs just to find they started with the wrong cell. Just remember the "0" and "+" go on the terminals that are used to messure the total pack voltage.

Bob

[Samson]

I thought I understood how a Cell Log 8M connected to a battery pack (40 Volts) from reading the rather vague PDF manual but while connecting one using pins 1 and 9 I heard a spark and the cell log would not power up. The polarity was + on pin 9. The manual for the CD appears to show the full pack voltage connected between pins 1 and 9 so I do not know what caused this. The unit will power up with the USB power connected but when I opened it up it looks like a component near the pins has blown open. I have another unit but I am reluctant to try it until I understand the cause of the problem. Perhaps I misunderstand how it connects. The Cell log was new. Any ideas?

As stated by Alan B the max amount of cell you can connect are 8. IT sounds like you did things correct but just to be sure I go thouth how I always connected mine. I connect to 8 cells....the CellLog8 has 9 connectors numbered "0" through "7" and a "+" mark. I wire all my cells together then attach the wire from the "0" to the negitive of my first cell (this would be considered the negitive you would use to connect to your controller...or the neg you would use to messure the total pack voltage). Then connectt the #1 to the next cell's negitive and so on. the last connection should be the "+" wire from the CellLog8 and it is connected to the positive terminal of the last cell. This is where you would connect if you were checking the total voltage of the pack.

Note: If by chance the CellLog8 was active when you started connecting it will power-up when you connect the 3rd. wire (#0, #1 and #2). It will read the voltage of the first 2 or 3 cells. Most people that have a problem connecting the CellLog8 find that they connected it 180 degrees backward. I've worked with people on the phone on a few occasion helping them connect their CellLogs just to find they started with the wrong cell. Just remember the "0" and "+" go on the terminals that are used to messure the total pack voltage.

Bob

Bob you wrote that your unit is numbered 0 through 7. Mine is marked - then 1-8 and +. This is ten markings but only 9 pins I assume the first pin marked - is pin 1 and pin 9 is the highest cell tap voltage marked +.
I stated I had connected to 1 and 9 in my post which is - and + on the cell log itself. I was attempting to use it to read the complete pack voltage of 40 volts or so. I am certain the polarity was correct. I did not have any of the other pins connected at the time. I did see that one of the diagrams in the manual seems to suggest that pins - (first pin) and 2 can be used to read from 4 to 43 volts or are they indicating that this is how to power it up?

Ken

[dumbass]

Bob

Bob you wrote that your unit is numbered 0 through 7. Mine is marked - then 1-8 and +. This is ten markings but only 9 pins I assume the first pin marked - is pin 1 and pin 9 is the highest cell tap voltage marked +.
I stated I had connected to 1 and 9 in my post which is - and + on the cell log itself. I was attempting to use it to read the complete pack voltage of 40 volts or so. I am certain the polarity was correct. I did not have any of the other pins connected at the time. I did see that one of the diagrams in the manual seems to suggest that pins - and 1 one can be used to read from 4 to 43 volts or are they indicating that this is how to power it up?

Ken

OK, I think I understand what your doing now. As mentioned you can only read up to 8 cells but you are trying to read the total pack voltage only. Sorry but I've never tried doing what your thinking of so I honestly can't advise you on it. I know you can connect in the normal way and read the first 8 cells. A lot of people will use multiple CellLogs on packs with more then 8 cells. In this manner they are connecting as if they were just several packs of 8 in the normal manner.

Bob

[pm_dawn]

I thought I understood how a Cell Log 8M connected to a battery pack (40 Volts) from reading the rather vague PDF manual but while connecting one using pins 1 and 9 I heard a spark and the cell log would not power up. The polarity was + on pin 9. The manual for the CD appears to show the full pack voltage connected between pins 1 and 9 so I do not know what caused this. The unit will power up with the USB power connected but when I opened it up it looks like a component near the pins has blown open. I have another unit but I am reluctant to try it until I understand the cause of the problem. Perhaps I misunderstand how it connects. The Cell log was new. Any ideas?

If you really look in the Manual.
On page -4- there is a pretty nice picture.
It clearly shows that when measureing a pack you should connect it to pins 1 and 2.

Also the part CW-C220 "One pack voltage measurements line with clips" pictured on Page -3- in the Manual and part of the package should give you a pretty good clue on how to use the cellog when measuring a full pack of any sort.......

To bad about the cellog.....

Regards
/Per

[EBJ]

ok, So I was out testing the CellLog LVC today. (more like abusing it)
And I was able to get the voltage below the LVC without continuing to trip it, here is how:

First, when the cell-log hits the LVC, it would pull-down the throttle line for about 1 full second (just enough time to let off the throttle).
I then proceeded to hit the LVC, wondering what would happen if I got to the "point of no return" (where you can't even accelerate at all any more).
After hitting it about 20 times, I then gradually accelerated to build up some speed, then coasted for a little bit, then "whalla", no more LVC function.
I was able to accelerate full-throttle without the LVC tripping. and I did multiple full-accelerate passes to confirm the LVC stopped working.

So, after I "tricked" the LVC to stop working (somehow).
I then took a reading and confirmed each cell was bellow the set LVC of 3.70V (they are all at 3.65ish).
I then turned off / on the main-power (and cell-logs) and the LVC function seemed to start working again,
this time cutting the throttle almost all the time (even with the most minimal acceleration).

Anyhow, I was surprised I was able to do this.

My best guesses as to WHY I was able to do this:
1. the cell-log only trips the LVC X-number of times ?
2. the cell-log got confused and really only trips when it goes from ABOVE LVC to BELLOW LVC ? (in which case I must have gotten my pack to always be bellow the LVC and that was why it stopped cutting the throttle)
3. the relay stopped working (seems unlikely, especially since the LVC function started working again after I cycled the main power-switch)




Anyone else ever tried (and succeeded) in fooling a cell-log LVC ?

[itchynackers]

This may help you. I connect a cellog 8 (setup for 8s) to my 5s pack just to get some quick measurements. Obviously the LVC trips, making the beeping noises for a while. However, after a period of time (maybe 5 minutes, didn't really track it) the beeping will just stop. Then after another period of time (maybe 10 minutes) it will start beeping again, for no apparent reason. I wasn't discharging the pack. It was just sitting on the table with the cellog on it. I haven't tested it, but it seems like the cellog will beep a certain number of times, then wait for a while, then start beeping again.

[EBJ]

thanks for the info.
That could be what I am experiencing.
but it would be more about the number of times it goes off instead of how long, because technically i am above LVC, then below and LVC triggers, then repeat.

another cell-log related issue I may be having:
I was having strange (what seems like) motor-controller issues,
but they only happened periodically. Mostly when I leave the pack for a while, or when I am at the low-end of the pack.
I had set my "difference-in-voltage" alarm on the cell-logs to something like 0.025V , which I know now is too strict towards the end of the pack battery-life when it easily hits a v-difference of .03 or .04 V.

Another question: Does the cell-log "Low-Voltage" alarm act differently than the cell-log "difference in voltage" alarm ??

example: when LOW voltage is detected, the throttle is cut for what seems to be at least a full second ("one one thousand"), this tells me the cell-log must "alarm-out" for at least one second upon even the slightest 'blip" of LOW Voltage detection, but does it do the same for the slightest "blip" in "voltage-difference", from what I am experiencing, I would guess not.
If this really is my problem then I would guess the "voltage-difference" alarm only sounds as a "blip" when it detects a "blip" in voltage difference between cells (in other words: it alarms for less than 1 full second) ?

I'll pull open my battery-pack and change the "voltage difference" alarm setting to something higher (like .05V) , but I'm waiting for the problem to happen again so I can take cell-level voltage readings and confirm it is my problem.

[itchynackers]

Maybe I'm wrong, but I always thought the "difference in voltage" was the amount of voltage drop you want the total pack to experience. Example, my parallel pack is 20.5V hot (4.1v/cell). If I want the pack to drop only to 18.0V (3.6v/cell), then the difference alarm should be set to 2.5V. That being said, I don't enable the difference alarm, so I can't say for sure. Anyone else?

[jonescg]

So guys, what's your advice in my situation - I will have 170 cells, split up into four female D-Sub44s with 42 or 43 cells wired up to them. I'm thinking I would take 6 cell-log 8s and wire them up into a male D-sub 44 plug so I can get an instant diagnostic on one quarter of the pack. Alternatively, I can get 7 or 8 battery medics to tell me the same thing. I don't know which is cheaper or more reliable, but some kind of instant readout for 43 cells at once would be great.

Which do you think would be a better option?