alcatraz Posted March 20, 2020 Share Posted March 20, 2020 Since the original bms doesn't appear to have a charge terminal, but charges through the output leads. The only way for it to shut off charging at 4.2V is to communicate to the control board to shut down charging. If your serially added pack for any reason charges faster than the main pack (higher IR/lower capacity) then you will end up overcharging those cells. This is my theory... Quote Link to comment Share on other sites More sharing options...
alcatraz Posted March 20, 2020 Share Posted March 20, 2020 Also if you haven't connected the extra pack's bms properly for EUC usage you could trip a protection circuit and faceplant. Quote Link to comment Share on other sites More sharing options...
Planemo Posted March 20, 2020 Author Share Posted March 20, 2020 14 minutes ago, alcatraz said: If your serially added pack for any reason charges faster than the main pack (higher IR/lower capacity) then you will end up overcharging those cells. My theory is that when the serial pack BMS reaches 4.2v it shuts down, whether the main pack BMS has reached full charge or not. 11 minutes ago, alcatraz said: Also if you haven't connected the extra pack's bms properly for EUC usage you could trip a protection circuit and faceplant. I didn't build the packs or BMS's. The sole reason why I obtained them from an EUC specialist who has done many of these conversions was so that I didn't have to think about what BMS' would be suitable. As I said, I believe that every 100v MSX out there advertised as 1845~1860wh has been modified in exactly the same way Quote Link to comment Share on other sites More sharing options...
alcatraz Posted March 20, 2020 Share Posted March 20, 2020 Sounds like you are using gotway compatible parts, and that the small packs can shut down charging of the wholep system = good. Quote Link to comment Share on other sites More sharing options...
WI_Hedgehog Posted March 27, 2020 Share Posted March 27, 2020 It does work, though it's not a good idea. The primary battery inside the unicycle will have a certain charge/discharge profile. The secondary battery outside the unicycle should contain the same cells as inside the unicycle so they have the same profile (via cell chemistry). The secondary battery will most likely not have the same charge/discharge profile because it is newer than the OEM pack. The OEM pack will therefore discharge to a greater Depth-Of-Discharge and wear out [even] sooner. The way around this is to not discharge the 100V combo-pack below 20% (if you're conservative 30%). @meepmeepmayer is correct, a singular BMS would normally be used to check the voltage on each parallel bank of cells, which is why a BMS is rated for the series voltage of the entire pack. Without the BMS knowing what one or more banks of cells is doing, the BMS doesn't know how to regulate the output of the pack. With two separate Battery Management Systems, one sub-pack can start to over-discharge, but that's hidden because the other sub-pack doesn't know to idle-down, meaning the more discharged sub-pack is at 100% load even though the total combo-pack is not. This is complicated by the fact the two OEM main packs are in parallel, which would normally shift the burden to the less-discharge main-pack, however, since the combo-packs don't have a BMS reading all the cells, the less-discharged sub-pack is masked by its tandem sub-pack (which boosts over-all pack voltage), and the combo-pack on the "discharged side" looks more fully charged than it is, putting even more load on the most discharged sub-pack. The opposite could also happen, the newer sub-pack could over-discharge if the low-voltage cutoff in the new sub-pack was set lower than the OEM pack (or the voltage monitor was off--we're talking hundredths of a volt accuracy for good voltage monitors). The risk of fire is minimal due to the BMS monitoring the sub-pack temperature and charge, and the same with sub-pack over-voltage. The real issue is excessive wear on a sub-pack. In summary, as @meepmeepmayer says, for the time & expense it's better to upgrade to a new wheel, or at least wait until the current battery expires, then buy a 100V battery and sell off the usable 84V parts you'll be replacing with 100V parts. 1 Quote Link to comment Share on other sites More sharing options...
meepmeepmayer Posted March 27, 2020 Share Posted March 27, 2020 3 hours ago, WI_Hedgehog said: The way around this is to not discharge the 100V combo-pack below 20% (if you're conservative 30%). That's easy. With the voltage drop during riding, 30% no-load battery is where the ride ends anyways (battery beeps and slowdown under load). In addition, if you charge to 100% regularly (including keeping the charger in for balancing), shouldn't it be (mostly) ok? It appears that a lot of 100V mods are actually this 84+16 build. I'm wondering if the real-life disadvantages of this approach are overblown (like a lot of battery worries) and it is more legit than I initially thought. Maybe the future will tell more. 1 Quote Link to comment Share on other sites More sharing options...
Planemo Posted March 28, 2020 Author Share Posted March 28, 2020 19 hours ago, WI_Hedgehog said: The primary battery inside the unicycle will have a certain charge/discharge profile. The secondary battery outside the unicycle should contain the same cells as inside the unicycle so they have the same profile (via cell chemistry). The sub packs do indeed have exactly the same make and model of cells as the main packs. 19 hours ago, WI_Hedgehog said: The way around this is to not discharge the 100V combo-pack below 20% (if you're conservative 30%). My wheels never go below 20% indicated whilst riding so likely 30%+ at rest. 19 hours ago, WI_Hedgehog said: the newer sub-pack could over-discharge if the low-voltage cutoff in the new sub-pack was set lower than the OEM pack (or the voltage monitor was off--we're talking hundredths of a volt accuracy for good voltage monitors). The 'newer' sub packs are only less than 400 miles newer than the main packs but yes you're right I wouldn't have done this conversion on a wheel that already had 1000's of miles on it. I was however satisfied that the profiles on my main packs with only around 7 full charge cycles on them would not be significantly different (if at all) to the new cells in the sub packs. 19 hours ago, WI_Hedgehog said: In summary, as @meepmeepmayer says, for the time & expense it's better to upgrade to a new wheel, or at least wait until the current battery expires, then buy a 100V battery and sell off the usable 84V parts you'll be replacing with 100V parts. With the amount of riding I do, it would take years for the packs to become 'expired' to the point where it would affecting the range significantly. I also did this conversion for several hundred pounds less than it would cost me to sell the 84v and buy a 100v. I also didn't have to wait 2 months for delivery. And, crucially, I am not convinced that even if I did buy a new 100v 1850wh MSX, the battery setup would be any different to what I have done to mine. 15 hours ago, meepmeepmayer said: It appears that a lot of 100V mods are actually this 84+16 build. Thats part of my point. Its all well and good people saying 'just buy a 100v already done' but unless I am proved otherwise, I am convinced that ALL 100v 1850wh MSX's are done the same way. So why waste 100's of pounds and wait 2 months to end up with what I already have 15 hours ago, meepmeepmayer said: I'm wondering if the real-life disadvantages of this approach are overblown (like a lot of battery worries) and it is more legit than I initially thought. Maybe the future will tell more. And that is the million dollar question. No one quite knows. All I do know (if my suspicions are correct) is that there are many, many 100v MSX's out there with the same setup and I haven't heard of issues yet. I will make it clear, this conversion isn't for everyone. I did the math and it worked for me, but it also took a lot of my time and I didn't even build the sub packs. There was a lot of checking and re-checking, plus even relatively small jobs like simply changing the charge port from 4 pin to 5 pin wasn't quick as not only does the shell need to be split but the wires and plug, being caked in hot glue took some time to remove. Admittedly the extra time I spent in wiring in an XT90 and upgrading the wiring from it to where the packs in each half of the wheel come together probably wasn't needed but it made sense to me. 1 Quote Link to comment Share on other sites More sharing options...
meepmeepmayer Posted March 28, 2020 Share Posted March 28, 2020 3 hours ago, Planemo said: unless I am proved otherwise, I am convinced that ALL 100v 1850wh MSX's are done the same way. I agree that probably all 1845Wh MSXes are this 84+16 - doesn't make much sense geometrically otherwise (unlike for the Nikola, for example, where it's just three regular 100V battery packs). 3 hours ago, Planemo said: Its all well and good people saying 'just buy a 100v already done' To be fair, the biggest argument to "just buy a 100V" is the time, knowledge, skill, and courage (do not underestimate) needed to do the modification yourself. Which you clearly have, but most people don't have (me included) Otherwise, one might just argue people can spot-weld their own packs from the individual cells, so any conversion would be easy peasy. So you certainly have my respects, and the result seems to work out great, enjoy your 100V speeds - Extra question (food for thought, because why not make everything worse?): Would it be a good idea or not to mix "real" 100V packs and 84+16 packs? 2 Quote Link to comment Share on other sites More sharing options...
alcatraz Posted March 29, 2020 Share Posted March 29, 2020 Anyone that's buying an electric vehicle should expect to receive BMSs that handle the whole voltage of the pack. Anything else is just unnecessary risks. Also they should also expect to get cells that are closely matched for each pack. That does not mean the same brand and model. DIY is fine for someone that's knowledgeable. Not for the regular consumer. Quote Link to comment Share on other sites More sharing options...
WI_Hedgehog Posted March 29, 2020 Share Posted March 29, 2020 I agree with @alcatraz 2 BMS in an OEM pack is rubbish. They may piggyback cells, but Gotway is probably using one BMS. Thanks for the feedback @Planemo, you did an excellent job and I would also think you should be way ahead for less money. It sounds like a fun and challenging project! 1 Quote Link to comment Share on other sites More sharing options...
Planemo Posted March 29, 2020 Author Share Posted March 29, 2020 4 hours ago, WI_Hedgehog said: I agree with @alcatraz 2 BMS in an OEM pack is rubbish. Interesting view. Do you have anything to substantiate it? 4 hours ago, WI_Hedgehog said: They may piggyback cells, but Gotway is probably using one BMS. Well the ones that the respected ewheels were/are selling aren't because Jason pulled a sub pack apart to prove it had a BMS. Gotway never officially sold an 1850wh 100v MSX either. 1 Quote Link to comment Share on other sites More sharing options...
Planemo Posted March 29, 2020 Author Share Posted March 29, 2020 6 hours ago, alcatraz said: Also they should also expect to get cells that are closely matched for each pack. That does not mean the same brand and model. Do you seriously think that all the wheel manfacturers and modders go to the length of matching cells? Whilst I agree with your view on matching in a perfect world, I am sceptical that the manufacturers doing anything beyond making sure they are the same make and model is a pipe dream. 6 hours ago, alcatraz said: DIY is fine for someone that's knowledgeable. Not for the regular consumer. Ageed, as I have said on this thread. 1 Quote Link to comment Share on other sites More sharing options...
WI_Hedgehog Posted March 29, 2020 Share Posted March 29, 2020 (edited) 4 hours ago, Planemo said: Interesting view. Do you have anything to substantiate it? I built and flew R/C helies, and learned way more about battery tech in general than I intended. When you have $5K+ in the air dependent on lithium, you tend to take interest. If an OEM piggybacked packs (84V+16V) and didn't use a common BMS that's asking for trouble as the packs age. In flying we have to put 2 packs in series to get extreme voltage (so a similar situation), but we monitor each before, during, and after flight (or have a bright blase, whichever...). It's now common for each pack to have RFID and the charge unit a RFID scanner with laptop connection and logging software to track the usage and charge profile in order to predict impending failure with strong certainty (hopefully). This is all based around how important discharge profiles and cell health are. (Tesla ultrasonically welds a fusible link on each cell so if one goes sub-critical it permanently disconnects before going critical--smart.) I'm not seeing an "official" Gotway 100V 1850Wh unit either (@Planemo). (Though really, is there any support/concern beyond Jason of eWheels?) Edited March 29, 2020 by WI_Hedgehog Quote Link to comment Share on other sites More sharing options...
Planemo Posted March 29, 2020 Author Share Posted March 29, 2020 I have been working lithium with RC most of my life as well, although it was only Ni-Cd/Ni-Mh in my earlier days lol. Jason seems to do a pretty good job of customer support so if he was having issues with the MSX's in question I think we would all know about it. And lets not forget that he is far from the only person who has sold them. To summarise, yes of course it would be nice to run a single bms for the entire wheel but at the moment we are where we are. I and many others are happy to ride the 100v 1850wh msx, nobody is saying anyone else has to 1 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.