Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[togo]

Here is all of the glorious voltage data from my MBB and BMS logs during my test ride and charging. Conditions are a single pack, 2016 FX, riding between 5% SOC dash increments from 100% to 0%, and charging at 18 amps. Note that the dash SOC display does not match the log SOC numbers as can be seen in the stopped points. MBB, BMS and DVM voltages agree closely. This was done in sport mode. I am surprised by the amount of regen shown in the battery amps but the test was done on mountain twisties. I now think that the slight rise in voltage seen at lower SOCs is probably due to the raised pack temperature. So a more accurate voltage to SOC conversion table for off bike stored packs would require controlling the pack temperature for each measurement.

Oh, this is awesome. Thank you so much. Did you use any spreadsheets or scripts to collate log info, and can you share them?

I notice the 90v with still fairly high reported SoC in your riding graph. Did you run out of juice? Is that a case of SoC inaccuracy, do you think?

[Keith]

This is a quick and dirty spreadsheet but why not share https://docs.google.com/spreadsheets/d/1zHsNWRYzPgohigwvZqp1881x9xyLLXoOs1yElWzPCnM/edit?usp=sharing
It has fragments of log files and hand entered measurement numbers but maybe it shows something about how I did it.
The lowest riding voltage I saw was 94.8. I rode about 5 minutes with the dash at 0%. Torque was very limited, very noticeable above 20mph but the bike never stopped. I had never been below 7% before, this was just for the test.

[Keith]

I'm considering changing my pack charging and storage routines based on the revised Zero recommendations and general battery wisdom about not charging fully unless necessary. It is very convenient and easy to just plug in the bike or externals, let them charge to 100% and forget it. But that leaves my extra packs stored at full 100% charge and stressed when they don't need to be. So now that I know how to control the contactor it shouldn't be too hard to design something that can automatically stop charging at some lower preset level, maybe 60 or 80%. Perhaps the same gadget could also check for too low of a charge while stored and alert or even add some charge. I'm still thinking about the possibilities but this seems like it might be worth the effort. Or not, maybe I won't be using these packs at all in 4 years. Manually stopping the charge short of full or draining and not charging isn't too difficult. And a simple AC timer could also do much of this charge management without further attention. But doing it that way would take some thought and planning and remembering every time so back to the gadget idea. Maybe one gadget could manage multiple packs and chargers too. Food for thought...

[motornissen]

Thanks for helping me with a similar problem. I would like to use a single pack in a go cart, but couldn't get the contacter to engage, so I've been bypassing the BMS system while testing. Not something I like doing. So now I have a couple of questions: when I do this, is the Bms still active? Will it disconnect when battery is fully charged/discharged? And where can I buy the Anderson connectors I need to connect the battery to the Go cart? And lastly... is there a pin diagram on the packs? Thanks!


Sent from my iPhone using Tapatalk

[Keith]

Well I can't say for sure, but I think so. The fake enable signal is the same as what comes from the QuiQ chargers. The contactor stays closed when they stop charging but I assume the BMS is involved, it does wake up and show charging activity and charge level on the LEDS, but I'm not sure if it really stops the charging or if that decision is done by the charger. I'm going to do some more investigating to see what happens with charging on the bike and controlling the enable line. I do notice that the BMS logs don't say much about QuiQ charging so there's that.

[togo]

I'm considering changing my pack charging and storage routines based on the revised Zero recommendations and general battery wisdom about not charging fully unless necessary. ... automatically stop charging at some lower preset level, maybe 60 or 80%. ... Food for thought...

For an electronics guru, it might be fairly straightforward to set up a circuit that closes a relay with a momentary switch and opens it when it hits a particular voltage.  I'm thinking a flip flip, a transistor, and a zener diode would be most of the circuit...

[Keith]

it might be fairly straightforward to set up a circuit that closes a relay

I'm sure I could design a circuit like that and it might be the best approach but those components probably cost more than a single board computer today. Since the enable line is just a low current logic signal, no relay is needed. The voltage can be measured very accurately with an ADC and you could easily change the cutoff set point. The tradeoff between sbc computer and a simple hardware design approach might be mostly about design time rather than parts cost, but a software based solution would probably cost less and could certainly do more.

I did some testing today to make sure the contactor can be controlled with packs on the bike charging while externally. I found no surprises, opening the enable line during charging causes the contactors to open and charging stops. The same battery and resistor enable connection works at the external connector. Charging will not resume after stopping the enable unless the key is cycled. The key must always be switched on to close contactors and begin external charging but the key can then be turned off. That differs from the pack behavior off bike since they will close the contactor whenever an enable signal is provided.

[togo]

I'd be tempted to adapt this circuit.

Obviously the 4.7k zener would have to be something like the target voltage, 112v? 114v?  Can you stack zeners to get the value you want?

Also you'd have to get components tolerant of the higher voltages and Power FETs capable of the current.  And some fuses, and probably some LEDs and resisters so you know what's going on.

[Keith]

I have ordered the parts to try this design. Just add a little code, I hope.

[togo]

I happened to see that in the manual just as I was telling a friend about it, thanks Brian. I have to say it is a thrill to light up a 100W bulb from a pack and know it will stay lit for a long time. I also like the work you did on the charging options table, very nice.

A cool test, but I don't recommend seeing how long it will stay lit.  Light bulbs will drain well below safe minimums for lithium batteries.

[Keith]

I ran a 1500W oil radiator heater for a while just for fun. I learned that 110V DC is very different stuff than 110V AC. I noticed an arc when I switched the heater power level, blue spark inside the switch. Interesting! But it still surprised me when I unplugged it while it was drawing 13 amps of DC. BIG ARC! WOW Every time. That little spark you see when disconnecting AC is nothing, this is a BIG SPARK! So that's why a lot of switches and relays are rated for lots of AC current and voltage but much less for DC. I'd seen those ratings but now I've seen the Reason. For those who doubt here's a video

[Keith]

A cool test, but I don't recommend seeing how long it will stay lit.  Light bulbs will drain well below safe minimums for lithium batteries.

I'm doing more testing now. I found that the enable line trick does not work when the pack is not fully charged. I don't know the cutoff point, but a 102V pack charge would only keep the contactor closed for about 30 seconds before opening, while driving that 100 watt bulb. I've got my Arduino running the enable line now and it works fine keeping the contactor closed for about three hours now, dropping from 116 down to 111V. My point is that the BMS is looking out for the pack so it should be safe regardless. But if the voltage gets down to 96V, and if my code is correct, the Arduino will shut it off then. But I'll keep an eye on it! I've got a serial port log so I should get a number for a BMS cut off if it happens.

[Keith]

Contactor opened after about 4 hours closed, 108V, 6 hours since full charge, did a few code edits. After opening, contactor would then not stay closed for a minute on restarts. Put the pack in the bike, read 79% SoC, went for a quick ride, now it is happy driving the 100W bulb again. For how long? kWhs used is about right but of course my "load" is not really constant power. Would I need CAN bus to convince the BMS to stay closed for longer? I'll look at the logs later, wonder what mode the BMS is in, 2 second flash code is for charging according to the Zero owners manual. So it waits 4 hours to get some charge current and then gives up?

Huh, stopped after half an hour. Seems very dependant on SoC.

[Shadow]

30-minute timeout is normal, and is reset by closing throttle. There's a way to signal the bike to stay in charging mode using AUX port signal lines.

[Keith]

I guess you mean using settings by serial com to the OBD connector, right? Maybe this: ov_chg_dis Override charger disable? I could but for now I'm doing my tests with FX packs off the bike so no communication is possible unless I connect directly to the BMS, maybe I will someday, I haven't found a suitable connector for that. I did see 30 minutes mentioned in the unofficial manual contactor section but so far from 80-100% SoC the timeout is 4 hours. I'm now testing with a much heavier load (1600W) to see if that makes a difference. Is it time or SoC that caused the 4 hour open event? Not SoC, it is now down to 104V after 50 minutes and still closed.