CellLog 8S battery monitor and logger - any reviews?

[evblazer]

I just ordered the soundtraxx and tcs one I'll report back with what I find on fit and quality with my celllog 8s.

 

[GGoodrum]

These are quite handy little widgets.

I've been testing these as a replacement for the LVC and HVC detection parts of the new BMS we've been working on. What I did was use an opto, plus a few parts, on each unit, so that the non-isolated alarm outputs could be "ganged" together to generate the same sort of opto LVC/HVC signal. I'm testing this with an external "balancer" version of the new BMS. Actually, the charge control circuit is on a standalone board that goes in a small 1" x 2" x 3" box. This goes inline, between the charger/supply, and the pack's main charge connections. There is also a two-wire connection to the optos on the CellLog interface board, to pick up the generated HVC signal. Separate from the CellLogs and the charge control unit, is a another box with just the shunt circuits (24 channels...), and four fans. The reason for doing it this way is because with Lipos especially, the cells stay pretty close, so I don't always need to balance, but when charging, I still want the overvolt protection that the charge control circuit provides.

What I'm going to try next is doing a board for the three CellLog units that has all the wiring for connecting the three units in either an 18s, or 24s configuration, selected via some jumpers. This way I can use the same three CellLog units on both my 24s and 18s setups.

-- Gary

 

[j3tch1u]

These are quite handy little widgets.

I've been testing these as a replacement for the LVC and HVC detection parts of the new BMS we've been working on. What I did was use an opto, plus a few parts, on each unit, so that the non-isolated alarm outputs could be "ganged" together to generate the same sort of opto LVC/HVC signal. I'm testing this with an external "balancer" version of the new BMS. Actually, the charge control circuit is on a standalone board that goes in a small 1" x 2" x 3" box. This goes inline, between the charger/supply, and the pack's main charge connections. There is also a two-wire connection to the optos on the CellLog interface board, to pick up the generated HVC signal. Separate from the CellLogs and the charge control unit, is a another box with just the shunt circuits (24 channels...), and four fans. The reason for doing it this way is because with Lipos especially, the cells stay pretty close, so I don't always need to balance, but when charging, I still want the overvolt protection that the charge control circuit provides.

What I'm going to try next is doing a board for the three CellLog units that has all the wiring for connecting the three units in either an 18s, or 24s configuration, selected via some jumpers. This way I can use the same three CellLog units on both my 24s and 18s setups.

-- Gary

integrating the cell-logs and further simplifying/shrinking The BMS would be killer! are you going to rip the cases off and mount them directly to the board?

 

[monster]

so have they made a 45s version or 16s veersion yet?

 

[GGoodrum]

integrating the cell-logs and further simplifying/shrinking The BMS would be killer! are you going to rip the cases off and mount them directly to the board?

So far, all I've done is mounted them to a PCB that has the opto-isolation for the alarm outputs, but I'm looking to do a new board that integrates the charge control part of the BMS onto the same board. Here's my current test setup:

View attachment 24s CellLog Balancer-BMS-v3.9.9-01.jpg

 

[nwmtnbiker]

So, how many of you using the CellLog 8 have made the modification so that power for the device is drawn from batteries 1-8 and not just the first 6? I got my CellLog today and should be getting my TS cells mid-week. I guess I'll be opening up the CellLog and soldering so that I don't begin to get my cells out of alignment. I hope that the company makes a hardware improvement and corrects this issue this year.

 

[evblazer]

Argh.. Ok took me a while reading through lots of posts but Post 123 http://www.rcgroups.com/forums/showthread.php?t=1055820&page=9 starts off with the directions for putting in a resistor to allow all 8 cells to power the thing I think? Since I have 8S thundersky that I hope not to have to charge individually I may try this mod.

Also I got my 9pin JST connectors from the train shop. I think it's too small it must be another series of JST connectors that is also 9 pins but a difference size. Wires are 24awg of smaller it looks like and the ones in all these pics look alot larger then that.

 

[scoot]

....Also I got my 9pin JST connectors from the train shop. I think it's too small it must be another series of JST connectors that is also 9 pins but a difference size. Wires are 24awg of smaller it looks like and the ones in all these pics look alot larger then that.

You'll want the JST-xh series I believe. You can get the housing & pins from didgikey but crimping would be challenging. There is a fella' on the RCG forum that makes custom ones using high strand 22awg servo wire and he has the proper crimper. I have been planning to have him make some for me some day soon.

 

[SpeedEBikes]

So, how many of you using the CellLog 8 have made the modification so that power for the device is drawn from batteries 1-8 and not just the first 6? I got my CellLog today and should be getting my TS cells mid-week. I guess I'll be opening up the CellLog and soldering so that I don't begin to get my cells out of alignment. I hope that the company makes a hardware improvement and corrects this issue this year.

I've done the mod on half a dozen units. It's worked ok on all of them. I don't use a resistor but just bridge the points with a single strand of fine copper wire. It requires a fine tip on your soldering iron and one must be careful as there is another pad close by that you don't want to accidentally bridge. A magnifier is a big help.

One small downside is that after the mod you can't use the unit on a 2 cell pack.

 

[nwmtnbiker]

SpeedeBikes, thanks for the guidance on soldering. I haven't opened up my unit yet--I'm waiting on some electronic parts to be able to build my battery case. I have a Currie conversion kit and want to build a case that will slide into the Currie rack. FYI--for those wanting to monitor individual cells with the Celllog, you need to have it plugged into USB power, an individual cell doesn't seem to provide enough power to power on the Celllog. Once you have the unit plugged into USB power, you can read individual cell voltage using the alligator clip cable provided.

Evblazer, you can order a ready-made 9-pin cable off ebay for $13 (with free shipping). I'm patiently awaiting mine--it sat in the Hong Kong post for 4 days before they moved it (don't know if they had a backlog due to the month-long February holiday in China). Using the tracking number, it's now supposedly shipped out of Hong Kong.

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=270470253072&ssPageName=ADME:L:OU:US:1123

 

[scoot]

.... you can order a ready-made 9-pin cable off ebay for $13 (with free shipping). I'm patiently awaiting mine--it sat in the Hong Kong post for 4 days before they moved it (don't know if they had a backlog due to the month-long February holiday in China). Using the tracking number, it's now supposedly shipped out of Hong Kong.

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=270470253072&ssPageName=ADME:L:OU:US:1123

Please let us know nwmtnbiker how that DB8 connector ends up fitting onto the CellLog pins. That's a nice ready made harness for headway cell/pack attachment. There might be a bunch of Gopedders interested in those soon 8) .

 

[nwmtnbiker]

Scoot,

The cable arrived today. It's very nicely made and the 9-pin JST fits the Celllog perfectly. I will need to modify it since it has only 1 ring connector for each cable and, from looking at the schematic in the CellLog manual, it looks like cables for pins numbered 2-8 should have two since each cable will have to connect to positive and negative terminals on the battery cells.

Here's a photo of the cable (I'll be cutting off the extra connectors on the positive and negative cables. I'm glad I ordered it and once the Hong Kong post put it in international mail, it arrived quickly.

 

[scoot]

Awesome....

All you need is a single ring connector at each serial connection of the cells... so don't let that diagram confuse you. You're good to go from what I can see. The extra read and black jumpers with the banana tips are for plugging into a charger, like a 208b icharger. In fact it is clear that is what this harness was designed for (a balancing hobby charger). If it were me, I would hack off the banana pins and replace them with anderson power poles or something that can't cause a short.

Are you hooking it up to a set of Headways? That's what I am wanting to use that harness on. Are the ring connectors 1/4" ID?

EDIT: BTW, I researched the "DB8" designation today and found that it is a hobby balancing charger from Chargery, and looks like a clone more or less of an iCharger 208b

 

[nwmtnbiker]

Scoot, thanks. I should have realized that the serial connection was all I needed to connect to. I attached the harness to one of my 2 12V 20AH TS packs, which have 4 cells each. Everything works well. I will need to wire these two packs in series as well as build my battery case. The ring terminals don't have any ID markings. They're just the right size for the bolts in the TS cells. Measuring them, they're about 1/4"--as I said, the harness is very well made.

Now to get the CellLog opened up to modify it so that cells 7 and 8 don't go out-of-balance quickly. SpeedEBikes, how do you get your CellLog units opened without damaging them? I've tried to gently open mine with a jewelry screw driver and it doesn't seem to want to open. I've not pried too hard because I don't want to damage it.

 

[mwkeefer]

Now to get the CellLog opened up to modify it so that cells 7 and 8 don't go out-of-balance quickly. SpeedEBikes, how do you get your CellLog units opened without damaging them? I've tried to gently open mine with a jewelry screw driver and it doesn't seem to want to open. I've not pried too hard because I don't want to damage it.

Sorry to jump with an answer but just grab the parts protecting the pins and pull - mine pop open easier than anything I ever ordered... be careful you ensure to put the little grey buttons back in the proper way.

-Mike

 

[SpeedEBikes]

Scoot, thanks. I should have realized that the serial connection was all I needed to connect to. I attached the harness to one of my 2 12V 20AH TS packs, which have 4 cells each. Everything works well. I will need to wire these two packs in series as well as build my battery case. The ring terminals don't have any ID markings. They're just the right size for the bolts in the TS cells. Measuring them, they're about 1/4"--as I said, the harness is very well made.

Now to get the CellLog opened up to modify it so that cells 7 and 8 don't go out-of-balance quickly. SpeedEBikes, how do you get your CellLog units opened without damaging them? I've tried to gently open mine with a jewelry screw driver and it doesn't seem to want to open. I've not pried too hard because I don't want to damage it.

I just gently run my thumb nail down the crack on each side of the case. There are tabs on one half that squeeze into the other half. When you catch one with you nail it should depress allowing the case to be easily spread.

 

[SpeedEBikes]

One other thing. I've had pretty bad luck with the alarm cables for the cell logs. They are very delicate. I've broken several and was making my own replacements with new connectors and pins bought from Digikey. But the tiny pins are tedious to crimp and they can only accept rather thin wire. I've tried a couple different methods of making strain reliefs, but have only had limited success. I put Y connectors on my packs balance leads so that I can leave the cell-logs connected while using the second connector for balance charging. It's very nice having the cell-logs hooked up while charging as it provides an additional safety factor, especially when I do quick charging without balancing of multiple packs in series. I use adhesive backed velcro to stick the cell-logs and the external piezo alarms in place on the tops of my packs. But even with all that, I still manage to sometimes break an alarm wire.

A few days ago I tried a new approach. Instead of making a cable to plug into the alarm connector, I soldered two leads of heavier wire (22 awg I think) and ran them out a hole I made by cutting out one of the bars of the adjacent speaker grill. I had to trim a bit of plastic from the mating case to get proper closure after routing the wires out the hole. The snug fit of the wires exiting the case and the tight routing internally creates a sort of strain relief in that tugging on the wires doesn't transmit force to the solder joints. I put a bigger connector on the new alarm leads and so far I've had no trouble.

 

[garygid]

Instead of contracting for a 24s or 48s development, adding a proper daisy-chain feature to make an 8D model might be much more economical, and a lot more flexible (bigger market).

Adding a 2-way opto-isolated (or fiber-optic) connection (to connect additional modules) would allow easy expansion to any number of cells.

Firmware changes would be needed to sync log-writing and log reading, and maybe enter sleep mode and wake-up.

Reading the log(s) now seems to require manual interaction with the Cell-Log 8S to select the log file (there could be a default) and to initiate "sending" the log-file's data to the PC. A USB command from the PC should be able to initiate the transfer(s).

Perhaps the current USB port is only a "simple" one-way data-out design?

Thoughts?
I have yet to get mine yet.

Is there any schematic of the Cell-Log "guts"?

With that, I could tell a lot more about how difficult this "D" re-design would be.

Those who have contacted the maker of the 8S might want to propose this "8D" product.

Cheers, Gary

 

[GGoodrum]

One thing I'd like to see changed is a reduction of the amount, or elimination, of the hysteresis built into the alarm output. that way I could use these as HVC triggers with the charge controller unit. I've added optos to these outputs, so that I could use two or three of these together, to drive the charge controller, but the hysteresis is just too great. Even adding a big capacitor, to smooth out the voltage swings, isn't enough to make these usable for this application. Still handy little units, though.

-- Gary

 

[scoot]

One thing I'd like to see changed is a reduction of the amount, or elimination, of the hysteresis built into the alarm output. that way I could use these as HVC triggers with the charge controller unit. I've added optos to these outputs, so that I could use two or three of these together, to drive the charge controller, but the hysteresis is just too great. Even adding a big capacitor, to smooth out the voltage swings, isn't enough to make these usable for this application. Still handy little units, though.

-- Gary

Gary, could you elaborate on this your observations just a little. We were planning on using the CellLog LV for triggering the brake circuit on a controller we have in our Goped ESR scooters. Since we haven't gotten to the actual testing, we are not quite sure what to expect. Any idea what we might run into there?

 

[GGoodrum]

Gary, could you elaborate on this your observations just a little. We were planning on using the CellLog LV for triggering the brake circuit on a controller we have in our Goped ESR scooters. Since we haven't gotten to the actual testing, we are not quite sure what to expect. Any idea what we might run into there?

I haven't seen what the hysteresis would do to the LVC signal, other than to verify that it was there, but I'm guessing it will be the same sort of problem, but in reverse. With our LVC signal, which has no hysteresis, if you connect it to the brake input, or use it to pull down the throttle signal, it will cause the throttle to "oscillate", at about a 2Hz rate. With the CellLog, I'm guessing the throttle could be cut for as much as a second, but I don't know for sure.

-- Gary

 

[SpeedEBikes]

So hobby city now is stocking a new 'M' version of the cellog8 that is 1/2 the price. The difference is the new unit doesn't do logging. But for those mainly interested in the monitor/alarm function this is great news.

I've been considering making a feature request of user selectable configurable alarms. Currently the triggers are configurable but they all result in the same alarm. I use the alarm output to trigger a big piezo buzzer. Because I mount the cellogs inside my battery cases I can't see the displays in motion. Sometimes it can be difficult to know whether an alarm is tripping because of low voltage or due to voltage differential. In the past I resolved this using the logging feature and then analyzing the logs. But with the new units without logs even that won't work.

So what I was thinking was that if the unit could be configured to strobe the alarm outputs, say with user specified on & off durations in milliseconds one ought be able to generate 2 or 3 easily distinguishable alarms regardless of whether one is using it to drive a piezo or an led. Another thought I had was that alarms ought to have a timeout/chirp option. So for example if a low voltage alarm has triggered for more than say a minute, the unit could then switch to just chirping the alarm for a second every couple minutes. Ideally the timeout and chirp frequency would be configurable too.

What do others think? Any better way to design such features?

 

[shinyballs]

Having problems wiring the alarm outputs of 3 CellLogs to a single piezo buzzer, powered from a 12v dc-dc converter. The alarm output wire(black) of each Celllog are combined together and connects to the converter's(+) and the other alarm wires(red) are again combined and wired to the buzzer's(+). The ground connects directly between buzzer and 12v converter.

This fried a CellLog when they're connected to the pack - this occurred even when the alarm outputs are not connected to the 12v power source Does each CellLog has to to have its own buzzer and needs to be isolated? A wiring diagram will be very helpful!

 

[webfootguy]

Having problems wiring the alarm outputs of 3 CellLogs to a single piezo buzzer, powered from a 12v dc-dc converter. The alarm output wire(black) of each Celllog are combined together and connects to the converter's(+) and the other alarm wires(red) are again combined and wired to the buzzer's(+). The ground connects directly between buzzer and 12v converter.

This fried a CellLog when they're connected to the pack - this occurred even when the alarm outputs are not connected to the 12v power source Does each CellLog has to to have its own buzzer and needs to be isolated? A wiring diagram will be very helpful!

Read the post from "jrickard" in this thread http://www.diyelectriccar.com/forums/showthread.php/lithium-battery-monitor-39708p2.html.

Basically, you need one relay per Cellog with NC (Normally Closed) contacts. This forms a loop that can be broken when the Celllog activates (program it properly to do this).

 

[Doctorbass]

For those who would like to activate or not their permanently installed celllog8, Mike on this forum have found a great relay that can switch all contacts at the same time so they dont stay always ON when connected to the battery!!

8PST for 2.50$ !!!

PC Mount Sealed Relay - 8PST
(KO) 0490-1385

KV PC mount sealed relay. 8PST, normally open double contacts. 13.8 vdc coil. 7/16" x 1-3/8" x 1-3/16" high. Alternate P/N: DM723.



here is the link: http://www.surplussales.com/Relays/sealed_pc.html

Very great find!!!


Doc