CellLog 8 hacking

[dumbass]

can i confirm something mentioned earlier, you can put an on/off switch on the negative wire to cell log to cut all power to it?
anyone done this?

That's the way I've been doing for more then a year with no problems. Some say there is still a slight power draw but it's very small if any.

Bob

[EBJ]

can i confirm something mentioned earlier, you can put an on/off switch on the negative wire to cell log to cut all power to it?
anyone done this?

That's the way I've been doing for more then a year with no problems. Some say there is still a slight power draw but it's very small if any.

Bob

yeah, I have a main on-off switch running to a relay that will interrupt the cell-log negative (turning it off).
I did some real-world testing w/ cell-level voltage readings, and my 20Ah pack would take something like 60 days to drain when left "off" and at about 65%-charge (storage charge).
And I think w/ everything left "on" it would take something like 6 days to drain. But I also have an alarm that would be triggered at LVC... so if you hear the alarm you know you screwed up and left the thing "on". I don't know what the exact #'s were since I did the calculation a while ago, but I do know I found no problem w/ that method of "turning off" the cell-logs.

[whatever]

thanks for confirming the neg off switch works,
another question for da gurus:
the graph that show voltage as bar graphs does anyone know if it represents the high and low voltage settings?
I mean lets say one bar goes right down to the bottom does that represent the low voltage alarm point or does it go right down to zero volts?

[dumbass]

thanks for confirming the neg off switch works,
another question for da gurus:
the graph that show voltage as bar graphs does anyone know if it represents the high and low voltage settings?
I mean lets say one bar goes right down to the bottom does that represent the low voltage alarm point or does it go right down to zero volts?

Good question. It represents the true voltage of each cell but the value (height of the line) at a given voltage will change based on the max voltage setting. But I'm not sure what it does for the low voltage.

Bob

[whatever]

thanks bob, instead of being lazy i'll check in the next day or two about low voltage and post

[EBJ]

ok, I've got a question.
Does the Celllog still trip its LV alarm when it is in "Sleep Mode" ?

I ask because the last 2 times I "thought" the pack was low, I switch OFF / ON the cell-logs and then they started tripping the LVC only moments later.
(It could just be coincidental though...)

[dumbass]

ok, I've got a question.
Does the Celllog still trip its LV alarm when it is in "Sleep Mode" ?

I ask because the last 2 times I "thought" the pack was low, I switch OFF / ON the cell-logs and then they started tripping the LVC only moments later.
(It could just be coincidental though...)

I had a simular problem but on HV alarm. The CellLog would go into sleep mode while charging and I had cells going over 4v (lifepo4). When I would touch a button on the CellLog the screen would light and the alarm would go off. I actually did correct the problem but damn if I can remember how I did it. But it was in the settings that I made a change and it's been fine from then on. In any case I would do a trip test on it by setting the HV just above the highest cell. Then let it go into sleep mode and start the charger. It shouldn't take long for it to activate the alarm if it's working correctly. I've never had a reason for a LV alarm but I'm guessing if it works for HV it will work for LV too.

Bob

[EBJ]

ok, I've got a question.
Does the Celllog still trip its LV alarm when it is in "Sleep Mode" ?

I ask because the last 2 times I "thought" the pack was low, I switch OFF / ON the cell-logs and then they started tripping the LVC only moments later.
(It could just be coincidental though...)

I had a simular problem but on HV alarm. The CellLog would go into sleep mode while charging and I had cells going over 4v (lifepo4). When I would touch a button on the CellLog the screen would light and the alarm would go off. I actually did correct the problem but damn if I can remember how I did it. But it was in the settings that I made a change and it's been fine from then on. In any case I would do a trip test on it by setting the HV just above the highest cell. Then let it go into sleep mode and start the charger. It shouldn't take long for it to activate the alarm if it's working correctly. I've never had a reason for a LV alarm but I'm guessing if it works for HV it will work for LV too.

Bob

Interesting. Maybe someone else can chime in and we can figure out what is happening. I could have sworn they worked fine previously, but now I think I am experiencing the same you were: A lack of detection in "sleep mode".

[dumbass]

Yeah, it was strange because I gave detailed instructions to another guy how to wire the CellLog and set it up. I gave him all my setting to copy which he did. He never had a problem but I did. I remember just resetting everything and making a small change here and there on some odd settings that normally would have nothing to do with the problem. Bottom line it worked on both my units.

Bob

[EBJ]

Yeah, it was strange because I gave detailed instructions to another guy how to wire the CellLog and set it up. I gave him all my setting to copy which he did. He never had a problem but I did. I remember just resetting everything and making a small change here and there on some odd settings that normally would have nothing to do with the problem. Bottom line it worked on both my units.

Bob

Perhaps I will ditch the "sleep mode" and instead turn down the screen brightness.

I also noticed that the cell-logs have a "time-out alarm". I haven't tested it yet, but I imagine it is used to sound an alarm when you leave the thing powered ON for too long? (In which case that could be a nice feature if you have it hooked up to trip a BUZZER.... It could warn you when you leave your ebike ON)

[dumbass]

I will need to take a look at how mine are setup again. I know I have the screen saver activated but I may have turned off the sleep mode. It's hell when ya get old........anybody seen my pants?

Bob

[EBJ]

I just pulled open the battery box and switched over from sleep mode to "backlight off" mode (after 3-min).
I imagine that will solve my problem.

[dumbass]

I just pulled open the battery box and switched over from sleep mode to "backlight off" mode (after 3-min).
I imagine that will solve my problem.

As I mentioned I would recommend setting the HV to just above the highest cell voltage. Allow it to go into the backlight off mode. Then charge it and see if the alarm activates now.

Bob

[auraslip]

anyone have the fun problem where the screen gets flipped upside down? all it takes is a flick to make it go back. it's kinda like in star wars where han solo has to hit something to make it work.

[megacycle]

Yeh what weird kinda firmware issue makes it do that, or the wackiest gremlin, having to do headstands to view the screen.

[whatever]

i've checked the graph function in respect to whether the graph goes down to zero volts or just down to the low voltage alarm point, it seems fairly obvious it goes only down to the low voltage alarm.
below is a pic, its a 7 cell ( lilo - cobalt pack), cell no.7 has a voltage of 3.529v, I adjusted the low voltage alarm point to
3.5v ( normally its 2.5v) and its quite obvious the graph only displays down the the low voltage point.
( the graph high voltage point was 4.1v and the tops of the cells with high voltage has spread up a bit far but no problem)

graphcelllog1.jpg (23.41 KiB) Viewed 708 times

[EBJ]

I just pulled open the battery box and switched over from sleep mode to "backlight off" mode (after 3-min).
I imagine that will solve my problem.

As I mentioned I would recommend setting the HV to just above the highest cell voltage. Allow it to go into the backlight off mode. Then charge it and see if the alarm activates now.

Bob

yeah, that's a good way to test it.
But I just did a "real-world" test because I didn't wan't to open up the battery box twice.
In "backlight off" mode everything functioned as it should. LVC tripped many times w/o any problems.
I guess mine were either (1. not working, or 2. a little delayed) when in "sleep mode".

[dumbass]

Like I said I'm not sure if I have mine set in the screen saver mode or sleep mode. And considering I screwed up my knee again and can't ride for several weeks again I'm not likely to check it for a while. Most likely it's in the screen saver though.

Bob

[EBJ]

ok, new question:

Does anyone know how the "time-out" alarm works ?
Because I was under the impression it would cause the cell log to alarm-out after "x" amount of time.
I tested it today, and I couldn't get it to alarm.

Any ideas ?

[auraslip]

Just a heads up. The even drain mod doesn't appear to work over a period of weeks. After roughly a month the cells are know 100 mv off.

My setup is to leave them constantly plugged in, but to have the 1 ground pin broken on each unit via relays. Works great and is a fantastic setup, but it looks like I'll have to be balancing every month if i want to use them like this

Another issue I'm having is a "phantom cells" that pop up on my 6s pack showing .3v for cell 7 or 8. I believe this is an issue with the even drain mod.

[fechter]

The even drain mod doesn't make the drain truly even, just better than without it. The only way to really make it even is to completely disconnect the CellLog.

The phantom cell readings are interesting. I haven't seen this. There could possibly be some soldering flux residue from the mod causing this. You could try cleaning the board. If you are only using 6 or less cells then the jumper isn't going to help with making the drain even anyway and you could leave it out.

[auraslip]

I figured as much about the even drain mod, I was just surprised by how quickly the balancing went even when I only used the bike 20 minutes a day.
I'm using 22s so it's 2x 8s and 1x 6s celllog.

In any case, the phantom cells haven't showed themselves for a few weeks now and it quickly goes away once I start riding, so I'm not gonna worry.

[megacycle]

Those pesky gremlins again
Mines working ok thanks Fecht'

[Bartimaeus]

I've seen several references to tapping the TTL on this guy before the USB converter chip, but no specifics. Is there a testpad or a through hole that you can solder to or do you have to go directly to the pin on the ATMEGA32L?

[Gregb]

for what it is worth, I blew up the alarm O/P on my cellog and have bought a BVM-8s. very slightly larger but comes with its own alarm unit which consists of a buzzer and ultrabrite red led. Redaind correspond exactly with cellog and my DVM. (and it was a bit cheaper)