You really need the next size down adapter. A rear 140mm should be about right, and if it's too low, use those alignment washers on it, and plain washers as necessary.
£245 Argos Folder Improvements - complete story
- Thread starter saneagle
- Start date
For anyone landing on this thread without seeing the original a word about the controller swap, especially as I plan on one myself.
The new controller at 12A increases the torque over the old one and doubles the heating effect. The OP used calculations, heuristics and a wealth of experience to indicate the bike should be OK with this and has tested to check it worked, successfully.
So can we be sure the battery will be OK with the extra heat? LiNMC is known for exponential runaway meaning they can fail fiercely long after whatever set them off.
Extracting better performance consumes some of the margin of safety, shrinking the envelope of circumstances for safe use. Testing doesn't prove something's repeatable but gathers evidence we've left enough headroom for our circumstances. In other words a successfully uprated bike remains an experiment in progress. So when over-taxing batteries I suggest:
- a prominent label on the handlebars: ** AVOID TAKING INDOORS **
- lots of edge case testing straight after conversion
Testing can show that the outcome is or isn't worth upgrading with a suitably rated battery for safety. Alternatively the bike could live its life as a successful experiment, but this is where the OP's long experimental experience comes into play. For the rest of us there's the insidious risk of forgetting that it remains an experiment and loading it into the car one day, for example.
The new controller at 12A increases the torque over the old one and doubles the heating effect. The OP used calculations, heuristics and a wealth of experience to indicate the bike should be OK with this and has tested to check it worked, successfully.
So can we be sure the battery will be OK with the extra heat? LiNMC is known for exponential runaway meaning they can fail fiercely long after whatever set them off.
Extracting better performance consumes some of the margin of safety, shrinking the envelope of circumstances for safe use. Testing doesn't prove something's repeatable but gathers evidence we've left enough headroom for our circumstances. In other words a successfully uprated bike remains an experiment in progress. So when over-taxing batteries I suggest:
- a prominent label on the handlebars: ** AVOID TAKING INDOORS **
- lots of edge case testing straight after conversion
Testing can show that the outcome is or isn't worth upgrading with a suitably rated battery for safety. Alternatively the bike could live its life as a successful experiment, but this is where the OP's long experimental experience comes into play. For the rest of us there's the insidious risk of forgetting that it remains an experiment and loading it into the car one day, for example.
Last edited:
Personally I'm sure the BMS will limit the current to the maximum allowed for that battery pack but whether that is good for the battery pack long term I don't know. The Bosch battery packs that fail because of use rather than being left discharged seem to be those that are used by Bosch e-mountain bike motors which perhaps draw as much as 25A at 42V at peak times which is over 1000W while climbing hills. However those same battery packs can be used with very weak mid-drive motors designed for city/comfort bikes which only generate about a third of the torque of the e-mountain bike motors. Same battery pack but different application. I think it goes without saying the long distance slower discharge rate of those comfort/city bikes will prolong the lifespan of those battery packs. Bosch fill those battery packs with glue to prevent them being dismantled and that is part of the reason they can cook sometimes when discharged at very high currents.
Did anyone actually discover what cells this Argos ebike actually uses? The fact Argos discounted this ebike so much makes me wonder about the spec a bit and whether Argos discovered fairly low end cells in the battery pack which again might be the reason the controller is so conservative in power output.
I think you're overthinking this. I've bought consignments of "clearance" product from Argos for stupid cheap prices before... they're just churning stock to make room for new season product.
The heat in the battery is caused by the resistance of the cells and the current flowing through them. P=IxIxR. According to Ohms law, the voltage sag V=IxR. The resistance goes up when the battery cannot provide the current, so you can see straight away from the amount of sag whether there's a chance that the cells are getting a hard time and likely to overheat. There is no indication of any noticeable sag when running at 12A, so it's pretty clear that the battery has decent cells in it capable of giving a lot more power. I'd be happy to run at 15A. It's a 30 cell battery, so that's 5A per cell. nearly all cells can give that these days.
I'm quite happy to charge that battery in my conservatory, and even in my kitchen. There's no need to panic.
It could still apply to the manufacturer though, they are using more basic cells perhaps from a Chinese factory so used a very conservative controller. However Argos being UK based they will need full certification for their products so presumably the cells used are certified for sale within Europe and the UK.
Will that change as the cells age with use? Seems like these Argos ebikes will get a very long life out of their battery packs from what you described. As I seem to remember you said the controller was only 8A max so less than 3A per cell peak output. Surely that is going to give those cells a very easy life. Do you have any rough idea about the torque output of the motor with the standard controller compared to your upgraded KT controller? I seem to remember the advert from the Indian manufacturer stated 40Nm but that is likely the max the hub motor was capable of but may not be related to the power output of the controller they fitted. Does the KT controller provide the same amount of power the motor is rated for maximum?
China produces 80% of the world's battery cells and 78% of all lithium batteries, so I suppose bearing that in mind, most ebikers are riding on cells made in a Chinese factory. I suspect that the quality of cells depends more on the price paid for them rather than their country of origin. As I am sure we all know, many companies which are headquartered around the world are doing their actual manufacturing not where their head offices are, but in China, or in other low cost locations.
The charger is pretty easy on the cells too. It takes four hours to put back the electrons of 20 miles of riding. Slow charge, gentle drain, low heat. I think if the cell pack is treated properly and not left to go flat over many months, it will trundle on for years.
By the way, I have fitted a simple, wired cycle, computer to mine and after meticulously measuring the wheel circumference and programming the head unit, I note that my machine in its mid pas setting, gives power up to 11.8mph and then cuts off. As I exceed this speed through gravity on a slope or pedalling, and then slow down, it will cut back in at about 12.5mph. The higher of the three pass levels cuts off at around 15.5 - 15.8 indicated mph.
The gearing on the original bike means that unless running downhill it is pretty hard to exceed 16 miles an hour without the rider's leg's being a blur of thrashing thighs. I don't mind this set up at all. I really like it as a knock about ride.
Re Saneagle's failed battery power switch....
Mine failed in the on position the first time I used it. I bet there are dozens of returned bikes in the Argos warehouses with this problem. I am sure that many users would send them straight back.
I would be quite keen to get my hands on one of these failed batteries as a spare. Maybe they will sell the bikes REALLY cheap too. I hope they won't just be crushed for disposal with such a silly fault.
How could we get our hands on them?
EDIT:
Contact details https://www.argos.co.uk/help/contact-us/
Mine failed in the on position the first time I used it. I bet there are dozens of returned bikes in the Argos warehouses with this problem. I am sure that many users would send them straight back.
I would be quite keen to get my hands on one of these failed batteries as a spare. Maybe they will sell the bikes REALLY cheap too. I hope they won't just be crushed for disposal with such a silly fault.
How could we get our hands on them?
EDIT:
Contact details https://www.argos.co.uk/help/contact-us/
Last edited:
It won't be any different to any other ebike battery. They all get tired eventually, that's if they don't get other problems, like balance issue, leaking cell, wire broken, etc.
Torque is more or less directly proportional to current, so going from 8A to 12A gives a 50% increase in torque. It's a standard medium sized hub-motor, so it would be OK up to about 22A, especially in a 20" wheel, where it gets the chance to spin faster all the time and the benefit of less reaction force.
I reckon the main reason for the original low current is that you'd get a very jerky ride with the cheapo speed control controller if it were any higher. Other ones I've tried at 15A with 20" wheels are like mopeds. As soon as you start to pedal, the motor takes over with maximum power, so you can't keep up with the pedalling. Any more than about 12A needs a current control controller. It's not so bad with 26" or bigger wheels because of the lower torque from the motor. The other Argos bike with 27.5" wheels and 15A speed control controller is verty nice to ride. No improvements are necessary for it because the bigger wheels result in 37.5% lower torque or equivalent to running a 20" wheel at 10.9A.
Last edited:
No panic here, you've the experience to tell when the cells have more to give, to arrive at a fine bike and write a really interesting thread.
Suppose I click my 52V battery into my 36V bike, the display may not survive (see IEC 60384). In dissipation terms running at 15A instead of 8A is like overvolting from 36V to 120V.
Even if that works well the new usage envelope is tighter but how tight is unknown. Cell breakdown begins around 60 to 100 degrees and my conservatory may get hotter than yours, my hills may be steeper or longer, internal wiring hot spots can be invisible to the BMS, and so on.
That's why if I tried this kind of mod I'd be cautious until the battery part of the experiment was resolved one way or the other.
Last edited:
You're right to consider the dangers. When anybody does modifications, they need to be sure about what they're doing. I wouldn't recommend anything unless I was sure it's safe. Anybody, who's unsure should seek advice first.
When I started on ebikes 14 years ago, we were all pioneers. We pushed stuff to limits to see how far we could go, and we established safe boundaries. In those days without knowledge and experience, despite cooked controllers, melted wires, use of unmanaged lipos and weak batteries, nobody had a fire, so I'd conclude that there must be some other factors involved in the bikes that catch fire. Maybe it's something simple, like common sense or luck. I know one guy, who's standard and safely built BBS02 bike randomly and spontaneously started to burn in his garage. Luckily, he saw it smoking, so disconnected the battery. The cause was a badly sealed (by the manufacturer) controller. Water got in and shorted one of the MOSFETs. That one frightened me because it could happen to anyone. It's one of the reasons I've never used a BBS** on any of my bikes.
Last edited:
Whoah, when was that? Can't happen to my BBS01B, it's too well waterproofed - I've stuck the controller to the motor using heat-shrink resistant silicone sealant, and the controller components are potted (was it the potting that was leaky on the BBS02 you've mentioned?) - water would have to get through the silicone sealed gasket first. My motor has had much exposure to water, including complete submersion for 13 seconds with no ill effects. Left it standing in the rain for hours while gardening, then rode it back through rain. It's survived long rides in heavy rain.
I think a BBS01B would be a big improvement for your Argos bargain in a decade or two, when the hub motor doesn't provide enough help.
With a better battery (@Nealh's P42A? Can't recall which those high contuinuous discharge current Molicells he uses are), it'll devour hills like my Dahon conversion, and probably more voraciously. Of course you'll never do it. Cadence sensored BBS01B would suit you better than the TSDZ2 did.
Last edited:
It was about 7 years ago. The controller had been potted, but not well enough. There was a bit missing just where the MOSFET was. The guy had been riding in the rain. The dirty or salty water got in and shorted the MOSFET, which is right at the bottom, where the water would collect. The guy showed pictures on Endless-sphere. I've referenced it a couple of times on this forum.Whoah, when was that? Can't happen to my BBS01B, it's too well waterproofed - I've stuck the controller to the motor using heat-shrink resistant silicone sealant, and the controller components are potted (was it the potting that was leaky on the BBS02 you've mentioned?) - water would have to get through the silicone sealed gasket first. My motor has had much exposure to water, including complete submersion for 13 seconds with no ill effects. Left it standing in the rain for hours while gardening, then rode it back through rain. It's survived long rides in heavy rain.
I think a BBS01B would be a big improvement for your Argos bargain in a decade or two, when the hub motor doesn't provide enough help.
Alex and Bill buy new stems and being an engineer Bill lightens his. After a couple of months of daily rides Bill's twin Ben is convinced and buys the same stem and does the identical mod. Another two months go by and Ben face plants when his stem snaps only to be told "You must have screwed up your mod, ours are rock solid". What's really happened is that Ben's identical stem had a 40% chance of snapping and has done, while Bill's stem used twice as long had a 2 in 3 chance but hasn't. The mod increased the original failure rate of 0.0001 per ride 80 fold but Bill's stem continues to perform the same as Alex's, building up confidence, and there remains a 5% chance he'll still be singing its praises a year later (1-(1-P)^n).
Most of us can't test for go/no-go failures at a meaningful scale and what testing we can do doesn't guarantee repeatability, it mainly reveals any gross weakness. Batteries too have a brittle (exponential) failure mode with a potentially nasty outcome that makes this infeasible testing essential. So the take home for the most part has to be Don't abuse your battery.
Equally it's obvious battery safety isn't keeping up with modern cells and that embedding today's designs into standards to meet commercial ends won't help *. Threads like this, the folk who admin and sponsor the forum despite some iffy posts, the parts access and the enquiring minds to push the limits occasionally, are all essential - trace how mobiles and the internet got here and you can see why. Oops, tangent.
I'm sure you're right about the reason for the weak controller and I get as confident as the next guy in the stuff I build. However if a mod could carry this type of unquantifiable risk to others the engineer in me can't properly refute that, and the responsible thing is to keep any risk to myself regardless of how I feel, hence "Remain cautious with it". I hope you'll do loads more like this but come to see the battery as a special case for most mortals.
(*) At least 18% of ebike fires in 2023 were in manufactured ebikes according to Govt analysis.
Last edited:
That's an interesting statistic what is the source for this, that's quite interesting that 82% of ebike fires are from ebike kits? I saw some statistics somewhere and they were mixing in e-scooters with ebikes and they also said a percentage were modified ebikes, i.e. replacement battery packs plus there were others that had used the wrong chargers on their ebikes. I felt at the time the statistics were really messy and hard to get a handle on. When the New York fire brigade created a pile of burnt ebikes and e-scooters it was more e-scooters than ebikes it looked like and those of course were manufactured units. I don't think there are scooter to e-scooter kits yet.
The one common theme to many ebike fires I've seen in the UK is gig economy riders who typically ride huge miles so those ebikes both ebike kits and manufactured ebikes seem to have a high failure rate due to excessive miles. They don't typically use mid-drive motor ebikes as they aren't as reliable and would wear down the drivetrain much quicker, some do but it seems very rare to me.
Personally I still don't think the full picture of ebike fires has been presented yet. I'd like to see clear information on every ebike fire, which model, spec, modified? Amount of use etc. Enough data that you can form an accurate picture of what is going on.
I very much doubt that many of those came from bikes with Bosch / Yamaha / Shimano etc systems which have higher level communications between the battery and charger / motor than a lot of the cheaper options.
Related Articles
-
MTF Enterprises announces acquisition of EMU Electric Bikes- Started by: Pedelecs
-
Wisper 806T folding bike wins Which? ‘Best Buy’- Started by: Pedelecs
-
Sustrans calls for protected cycle lanes- Started by: Pedelecs
-
Amazon launch their first UK e-cargo micromobility hub- Started by: Pedelecs
