My Swytch kit battery (Mk1 - larger battery) has without warning stopped working. The bike still functions ok with alternative replica Bosch type lawnmower battery. Remove battery from Swytch casing to establish make and connections. I suspect the problem is in the BMS which incorporates an isolating battery switch circuit. The battery and BMS are 'encased' in a strong blue plastic wrap which I am hesitant to cut off until I know a little more about these batteries. It is a 36V ~10Ah LiIon battery with 6 leads - 2 thin white (switch sensor leads), 2 heavy power out leads, and 2 (thinner red + thinner white) charging leads. Any info would be gratefully recd - cnthpower.com website incomplete and doesn't appear particularly useful. Thanks Peter
Info required: Chinese TFI battery (ex Swytch Mk 1)
- Thread starter peterjd
- Start date
For info I took the plunge and cut-off the outer wrap so as to be able to examine battery cells and BMS. Since full battery voltage is available at the inputs to the BMS does look like a problem in the BMS. Not up to trouble-shooting that device unless it was simply a badly soldered connection eg to external on/off switch, so have ordered an almost identical looking battery management/protection module from Banggood for ~£9 delivered (April). Spec seems to be appropriate for the cells (18650). I will let you know if it works when it arrives.
Update. Ordered Banggood BMS but could take a month to come (tends to vary I guess because of shipping delays). In the meantime I had contacted Swytch and the battery manufacturer, both discouraging disassembly of battery. There is an understandable reason for that - lithium batteries can be dangerous particularly if cells or groups of cells are shorted or inappropriately charged. Being a retired industrial R&D chemist I can appreciate the risk and safety concern of the companies (and probably litigation concern). However unless as a world we don't find ways of effective recycling and repurposing the resources we are rapidly depleting, future generations (and perhaps our own) are going to be in real trouble. A start would be to resurrect a cost-effective repair industry - oh for the days of the knowledgeable repair man.
The knowledgeable repair man still exists but now a days they frequent the various forums such as this, but like the switch and battery company there are nay sayers on the forums who are scared of their own shadow who try to prevail at times.
The German pedelec forum and the Endless sphere forums are two where a lot of 'repair men' frequent also a few on here, plus there are many other niche forums where one can find good advice.
The companies don't like 'repair men' not because of litigation but because of one thing , profits and resales. Repairs ae no good for there bank balances and share holders.
The German pedelec forum and the Endless sphere forums are two where a lot of 'repair men' frequent also a few on here, plus there are many other niche forums where one can find good advice.
The companies don't like 'repair men' not because of litigation but because of one thing , profits and resales. Repairs ae no good for there bank balances and share holders.
Thought you might be interested in an update. Whilst waiting for a replacement BMS, I've established that there is a problem in the Li Ion cells of the battery. One bank of 4 parallel wired cells has short-circuited (at least as far as measuring voltages at each of the BMS sampling points go). So all the other banks are consistently 4.05 -4.09V but that bank is 0V. So total voltage shows 36V rather than ~40V without load. Pondering what do with the battery pack since it is practically impossible to (safely) replace the bank which is in the middle of the battery pack.
In the meantime I bought a Yose 10Ah battery which seems to be working well.
Peter
Another Update and query
So have now received the replacement BMS for my ex-Swytch ~10Ah battery - I think functionally identical to the original. However in the meantime I had disconnected the one parallel bank of 4 faulty cells (shorted 0V) and the original BMS and shown that the remaining 9 banks (~36V charged) could be successfully connected to the controller for a 24/36V friction motor which I am using as a booster on my other bikes and the motor was spinning ok.
So now I have some queries about using the old battery for my booster friction motor:
1. Can I safely leave the faulty 4 batteries disconnected in the pack?
2. Can I use the new(or ? old) BMS and the original charger to safely charge the recovered 9S4P nominal 31V (36V fully charged) battery. With or without modifications?
3. If (2) will be problematic how about charging the parallel banks of cells separately using the very cheap 4096 charger modules?
4. Am I right in thinking that a BMS will always be required for safe use of the battery pack even if only for the use (ie discharge) function? Or could I use method (2) for charging the battery and dispense with the BMS altogether?
Would welcome any comments. Peter
PS A fascinating retirement activity learning about rechargeable batteries
)
1. Yes if the BMS doesn't shut down.
If the motor works then yes one can use the BMS as 9s, though one will have to monitor the discharge voltage. Depending on the BMS lvc you may not need to monitor voltage (but it is better to do so), If the display doesn't have a voltage read out then wire in a cheap one as a live voltage reading is essential over having LED's which tell very little except they are working. . Usually a BMS for 36v battery will cut out at about 3.0v - 3.2v per cell +/- 0.1v, the 9s will be 3.2v at an overall voltage of 29.8.
At 9s the motor will have a bit more torque.
2. Yes if 1. doesn't occur and if one can do as mentioned below.
Sans brand chargers I have all have a voltage pot for adjustment and another China brand one I have also is the same.
As for charging one will need to see if the charger is adjustable with a voltage potentiometer inside, if it has one it is usually set with a blob of silicone on the top of the screw adjuster. Remove the blob and see if it will adjust down to 37.4 - 37.8v max for 9s, if so one can use the charger but just double check the charger output each time.
3. Not used a 4096.
4. No.
One can manually do both discharge and charge without BMS but needs a bit more work.
I use a 36v charger to charge my 12s/ 44v battery, the adjustment lets me charge to 50v so 4.1666v per cell bank without issue. Good cells will stay pretty well balanced.
If the motor works then yes one can use the BMS as 9s, though one will have to monitor the discharge voltage. Depending on the BMS lvc you may not need to monitor voltage (but it is better to do so), If the display doesn't have a voltage read out then wire in a cheap one as a live voltage reading is essential over having LED's which tell very little except they are working. . Usually a BMS for 36v battery will cut out at about 3.0v - 3.2v per cell +/- 0.1v, the 9s will be 3.2v at an overall voltage of 29.8.
At 9s the motor will have a bit more torque.
2. Yes if 1. doesn't occur and if one can do as mentioned below.
Sans brand chargers I have all have a voltage pot for adjustment and another China brand one I have also is the same.
As for charging one will need to see if the charger is adjustable with a voltage potentiometer inside, if it has one it is usually set with a blob of silicone on the top of the screw adjuster. Remove the blob and see if it will adjust down to 37.4 - 37.8v max for 9s, if so one can use the charger but just double check the charger output each time.
3. Not used a 4096.
4. No.
One can manually do both discharge and charge without BMS but needs a bit more work.
I use a 36v charger to charge my 12s/ 44v battery, the adjustment lets me charge to 50v so 4.1666v per cell bank without issue. Good cells will stay pretty well balanced.
Last edited:
Hi Nealh, Thank you for that rapid and authorative reply. I feel confident that I can go to the next stage. Peter
Most chargers have a screw x4 at the base feet location, the pots are easy to ID or if unsure show a pic. Some chargers use a solid state pot so not adjustable.
The 4096 modules for each cell group will be a faff if one was to go down that route as each cell group would need it's own jst connection lead.
The 4096 modules for each cell group will be a faff if one was to go down that route as each cell group would need it's own jst connection lead.
Thanks again. No screws on charger but with a little brute force.... Is this likely to be the required pot marked VR? Blobs of buff plastic scattered everywhere! Output voltage (no load) 41.6V
Yes, that's the one Peter.
Remove the silicone blob and turn it ACW and one might be able to use the meter to set it lower .
I think we can see the screws at the corners, one expects underneath they are covered by a little grommet that needs prying out with a thin pointy or flat driver.
An older thread where I modified mine for better heat control as they get very hot without a fan
Basic Sans charger mods/adjustments. | Pedelecs - Electric Bike Community
P.s.
Just read thru my old thread and realised I wasn't able to go much lower then 40.7v on my one, so one may have to find another charger.
An adjustable meanwell led driver/charger would work, 100HLG - 36a has the correct voltage range to adjust but the price is a bit high at some £50.
Or the 80hlg - 36a is a tad over £38.
HLG-80H-36A is a 82.8W 36V 2.3A LED Lighting Power Supply | Power Supplies Online
Remove the silicone blob and turn it ACW and one might be able to use the meter to set it lower .
I think we can see the screws at the corners, one expects underneath they are covered by a little grommet that needs prying out with a thin pointy or flat driver.
An older thread where I modified mine for better heat control as they get very hot without a fan
Basic Sans charger mods/adjustments. | Pedelecs - Electric Bike Community
P.s.
Just read thru my old thread and realised I wasn't able to go much lower then 40.7v on my one, so one may have to find another charger.
An adjustable meanwell led driver/charger would work, 100HLG - 36a has the correct voltage range to adjust but the price is a bit high at some £50.
Or the 80hlg - 36a is a tad over £38.
HLG-80H-36A is a 82.8W 36V 2.3A LED Lighting Power Supply | Power Supplies Online
Last edited:
The mean well range of led drivers are good for a battery charger for li-on as one can adjust the current and voltage range manually via ext pot grommets.
The 80h or 100h range ending in xxA are the ones required as the A indicates adjustment can be carried out, a xxB model is no good.
The spec/data sheets give the models and adjustment range min/max.
The 80h or 100h range ending in xxA are the ones required as the A indicates adjustment can be carried out, a xxB model is no good.
The spec/data sheets give the models and adjustment range min/max.
Last edited:
Interesting.
This is pretty much the problem I have had with a Swytch battery.
It only charges to 37.6V and the charger switches off. The battery has very little capacity.
The voltage is 37.6 when connected to the charger and the charger is showing a green light.
I assume there must be some sort of current sense in the charger as well as voltage and it switches off when the battery stops taking much current,
This is pretty much the problem I have had with a Swytch battery.
It only charges to 37.6V and the charger switches off. The battery has very little capacity.
The voltage is 37.6 when connected to the charger and the charger is showing a green light.
I assume there must be some sort of current sense in the charger as well as voltage and it switches off when the battery stops taking much current,
I don't think it is a problem with the charger the O.C. voltage is 42.6 and it charges my other battery OK.
It probably does have a duff cell group. I might unwrap the shrink-wrap coating and have a look. I am reluctant because I would not be able to replace the cells myself and it does work well enough for short trips at the moment.
The charger output has nothing to do with cell balance if it is outputting the correct voltage, cell balance can be due to several factors and very few from the charger unless it is out putting to low a voltage.
Cell balance can occur to a faulty balance wire, and ageing cell, one of a higher internal resistance or a bms bleed resistor not doing it's job correctly. Like wise it isn't uncommon for the first or last cell group to be lower then the others, this is mainly down to the poor discharge wire connection on the last or first cell in the group.
The only way to know for sure if the battery pack does have an issue is to manually check for an issue by opening.
42.6v charger output is high, it might be the error accuracy of the meter or a badly adjusted voltage potentiometer.
Cell balance can occur to a faulty balance wire, and ageing cell, one of a higher internal resistance or a bms bleed resistor not doing it's job correctly. Like wise it isn't uncommon for the first or last cell group to be lower then the others, this is mainly down to the poor discharge wire connection on the last or first cell in the group.
The only way to know for sure if the battery pack does have an issue is to manually check for an issue by opening.
42.6v charger output is high, it might be the error accuracy of the meter or a badly adjusted voltage potentiometer.
Turning into an interesting thread. I've taken note of all of your comments Nealth about chargers etc. - thank you. Whilst I have the naked battery pack I am experimenting with partially discharging the pack (using 3x12V 20W halogen bulbs in series and monitoring the voltage of each set of 4 parallel batteries (9 sets) until one of them reaches something like 3.1V and then monitor their recharge using the 4056 modules if necessary (I haven't checked the pot yet on the standard charger and I already have some 4056 modules). Not much to lose if it all goes pear-shaped.
PS The charger case was obviously designed for screws but in the one I have it uses case edge clips hence the 'brute force'. I will eventually use screws to put it together.
Well battery worked well powering the light but haven't run it down too far since I decided that with the remaining 9 series groups looking good and well balanced plus my existing purchase of a replacement BMS , learning about balancing/matching replacement cells, and the cost of specialised chargers for a 9 pack battery, it was worth lashing out a further £18 or so for 4 (Samsung) 18650s and hopefully reconstituting the original battery spec. I'll let you know the outcome. Peter
Related Articles
-
Swytch announce new conversion kit with ‘pocket-sized’ battery- Started by: Pedelecs
-
New Swytch launches on Indiegogo, raises £100k in first hour- Started by: Pedelecs
-
Swytch to unveil 70% smaller, 50% lighter conversion kit- Started by: Pedelecs