ATV conversion

Hello, I have an old 50cc chinese ATV, fairly small thing. I am converting it to electric using this Vevor controller/motor combo:

https://www.vevor.com.au/brushless-...w-kit-go-kart-bicycle-powerful-p_010194063736

Basically it's a 2kw brushless motor with a 48v 35A controller.

I hooked it all up and although the wheels spin fast, there doesn't seem to be much torque. It can't get me up a small hill. I am using 8awg wire for the power, connecting 4 12v SLA batteries in series - same ones you get from the powerwheels cars, like this:

https://www.ebay.com.au/itm/1541970...kw9uTECSWL9EH_Dubd_TLpYi31bxAE5BoCXV0QAvD_BwE

I tested the amps. I just sit the ATV before a hill to force it to use maximum current (stall current). When I put a clamp metre around one of the cables going to the batteries, it goes to about 23A for a second or so then drops down to about 15A. After a few seconds, all three phase wires going to the motor get fairly warm to touch. I know they're connected the right way as the motor and controller have colour coded wires, which are connected through a 60A connection block.

So my question is, do you think this could be the batteries not good enough to provide 35A, or the controller? I can provide pictures if it helps, but it looks like it's all connected fine using proper thick cabling.

Thanks for any help!

I did a very similar conversion last year and had the same experience. 2000w with 35A controller

https://endless-sphere.com/forums/viewtopic.php?f=34&t=109518

You have to look at the gearing of your set up. I ended up putting a huge rear sprocket and jackshaft in between to get the torque to a point it can climb those modest hills. Another challenge is the fixed rear axle. Overcoming the additional torque you need for turning can overwhelm this basic configuration just as much as a hill.

My build suits my purpose use now but it is overall underwhelming in performance.

Cheers

Ah that's very interesting. I haven't had time to read it all yet, as I'm about to sleep, but I will read it all tomorrow. But you're using a 56v battery... so, a 56v controller? Or can you limit the voltage on those EGO batteries? I know some can be limited but can't remember to what.

I am beginning to think that my batteries are playing a big role to the problem I'm having. Trying to find "C" ratings for these batteries are next to impossible, and finding out how much current you can pull from any battery is not easy. Do you know how much current you can pull from those EGO batteries?

Yeah the sprockets were a pain. The one that came with the motor was small, so had to weld a proper 420 sprocket to match the one on the axel. After testing and how crap it was, I decided to buy the biggest rear sprocket I could fit on it which was only 48T. I've also ordered a replacement front sprocket (10T) which will give it a 1:4.8 ratio. Nowhere near a 1:10 like yours though. If yours still struggles with 1:10 and a li-ion battery then mine will probably still be too slow to go up hills. And I think yours is a 45A controller although you're only drawing around 25A from the battery?

Ok I read your thread. Just realised you have a 35A controller but I think you mentioned in one of your other posts you were using a 45A one. Anyway, I'm going to try and focus on delivering 35A of current by using some batteries in parallel and then go from there.

You can add another gearing stage, by adding a jackshaft between the motor and wheel, and using two chains.

The end of the jackshaft that accepts the chain from the motor gets as large a sprocket as you need to get the right gearing for torque, then the other end gets a small sprocket to feed a second chain that then drives the wheel's large sprocket.

You can calculate out the sprocket sizes you need, by measuring the motor's free-spinning RPM at the average (half-charged) voltage of the battery, at full throttle.

If you measure the wheel, then figure out how fast you need it to go, vs how fast it is going, and then the ratio of the two minus a small percentage for loading down later, is the ratio of the gearing change you need.

If it's a pretty small ratio, then it's likely you'll have to accept a slower than desired speed in order to get the torque you're after, and use a (much) higher ratio.


If you go to the http://ebikes.ca/tools/simulator.html, you can experiment to determine how much power it will take to do the job you want to do under the conditions you need to do it in.

Regarding the batteries, the size of SLA you've chosen are not intended to provide that much current. Even if they could, then with only 7Ah, assuming you could get all of it (you'll really only get about half, or less, from SLA), then at 35A you'll probably get about 5 minutes of runtime, at best.

To do what you want, you'll probably need 20Ah+ sized SLA, and would be better off with 35Ah+ sized units (then at full load you'd at least get 20-30 minutes of runtime, and you wouldn't be hammering the batteries so hard).

One problem with SLA is that while they *can* deliver high currents, they drop in voltage tremendously while doing so, when undersized for the application. They also have a large Peukert loss factor, so unless you use them at their designed rating of 1/20th C (1/20th of their Ah rating, in amps), you won't get nearly their full capacity (in this type of application, maybe half of that).

Another thread with issues using the vevor kit
https://endless-sphere.com/forums/viewtopic.php?f=10&t=109552

I built a 14s5p battery with Sony VCT4s (10A continuous) and tried to see if it was current that was limiting the performance and it had no affect to torque over the EGO batteries originally designed in the build. I am swapping out controllers to a FarDriver now but suspect ultimately these inexpensive BLDC motors are the root. I’ve seen builds with these motors that have crazy torque but suspect there is a range of performance that matches their manufacturing quality control.

This was my first build and have/am learning a lot. Some times I keep getting the same lesson and in this case I think it’s the ol’ - you get what you pay for.

I’ll keep tweaking this as it is a hobby after all and see if I can get that torque challenge sorted.

Thanks for the replies.

Yeah not sure if I want to go down the route of putting a jackshaft in. But then again I am kind of at the limit with the rear sprocket because any bigger then there wouldn't be much ground clearance, and I'd probably have to move the sprocket over closer to the wheel (which then I'd need a jackshaft for anyway to align with the motor).

I am not really after speed as it's not for me but for my son who is only five (!) but he will use it at my parent's farm. He's a pretty good driver for his age as he has a bunch of other random modified cars that he spins around in. But this ATV as it is would be way too fast, so converting some of that for torque would be ideal. I wouldn't mind having a burn around in it though every now and again...

Regarding the batteries, I'm still working out how they all work together. My understanding is that the aH is irrelevant to how much current can be drawn but generally, the higher the amp hours, equals more current that can be pulled (?). But it's hard to find an official rating of batteries. Cells on the other hand seem to be much easier to find that kind of info, so is easier to plan a battery build.

Speaking of a battery build - the SLAs that are in it were really only temporary as it's all I had lying around. What I really want(ed) to do is test 48v with high current to see if it's my controller or batteries that is the weakest link here (I never assumed it was the motor). I don't have 48v li-ion batteries though (or anything that can total that many volts) so my next thing I was going to do was to take 3 car batteries and run them in series to get 48v. Seeing that car batteries can handle a lot of current, this would be plenty to see if it would make a difference to torque. I'd put these big ass heavy batteries on a trolley and connect them to the ATV and get someone to run along side me to test. It would look pretty funny...

But the fact that you (H2Guy) already tried with 14s5p battery and it made virtually no difference to the torque is a bit disheartening, but I'm sure they're better than my crappy SLA setup. So I'll try with the car batteries, work out if that's good enough, then build my own battery I suppose. Those cells you used would cost me almost $1000 for the same setup as yours over here. Might be able to find something cheaper though...

I will have a look at that simulator site. Thanks.

Sorry for the long reply.

justaperson76 said:

My understanding is that the aH is irrelevant to how much current can be drawn but generally, the higher the amp hours, equals more current that can be pulled (?).

Click to expand...

Generally, the latter statement is true, for any particular battery made of a particular cell. Put more cells in parallel, and the more current you can draw, vs less of the same cells.

But just because one random battery has more Ah than another random battery, doesn't mean it can supply more current. That capability depends on the specific abilities of the cells the battery is made of, and is usually given as the C-rate, which is a multiplier of the Ah number that gives you the Amps the cell can provide.

A 10Ah 1C battery can provide 10A. A 10Ah 5C battery can provide 50A.

There are also Max/Peak (a few seconds at a time, intermittently) and Continuous (till it is empty) ratings.

Different chemistries (lithium, nickel, lead, etc, and each of their several flavors) have different characteristics, so some are better at some things than others.

But it's hard to find an official rating of batteries.

Click to expand...

Wherever you buy from, should have the basic ratings for the things they sell. When they don't, you can check the manufacturer of the actual battery being sold, and they will have them. If they don't, don't buy those; they're probably garbage if they can't be bothered to even rate them. If the seller doesn't specify the manufacturer and model, don't buy from them, becuase you won't even know if you get the same thing you paid for.

Particularly with lithium batteries, the cheaper the battery is, vs a similar one from somewhere else, the less likely it is to be capable of what they claim, and/or the greater the likelihood it is to be made poorly, and of counterfeit cells, recycled garbage cells, or random cells.

There are different kinds of BMSs, as well, and the cheaper the battery is the less likely it is to have the balancing type, but the more likely it is to need that. If it doesn't say it has a balancing BMS, I'd pass on it and get one that does. (even if it's a known-good battery, it's still a handy feature especially if you run the pack hard, drain it down far a lot, etc).

Right, got it All makes sense, thanks.

The batteries I have are cheap Chinese ones and I can't find any info about them at all.

Anyway I'm going to put a few car batteries in series, test to see if the thing can get up a hill and if so, that means the limiting factor is the battery, so I'll find a battery that can supply the required amps, otherwise build one myself.

I have another related question to ask. If I can get four car batteries, is it safe to run these in series? All the car batteries would be 12v obviously but different brands/types/capacity most likely. I only want to do this for a few seconds just to see if it solves my issue. I'm assuming that nothing bad could happen just for a few seconds...?

Actually doesn't matter. I think I'll just pick up four 12v SLA batteries and run them in parallel with the other four I have. It's pretty easy to find those powerwheels cars out in the rubbish collection so I'll snag their batteries.

If they're regular FLA (flooded lead acid) then sure, you can run them in series fine. Just don't drop them or knock them over, or the refill covers can come off or the cases can crack, and then you have sulfuric acid everywhere.

I suspect you'll find they'll help a little with the performance, as they won't sag nearly as much as the tiny 7Ah SLA do under the load you have, so you'll get more watts to the motor (v x a = w, if volts drops because amps are high, watts drop too).

But I don't expect they'll completely solve your problem; it depends on hte speed your wheel system is geared at vs the top speed and torque you actually want, and you have to do the loaded vs unloaded RPM test to find out if you need to change gearing.

It's really easy to do an offground wheelspeed test, even just eyeballing it you can see if it goes way faster offground than it does under load (riding on the flats), and if it does it means you need to gear the system down to get the torque you need.


If you need a jackshaft it can be as simple as an old dual-disc bicycle hub with dropouts (and some frame) cut off an old junk steel frame, with sprockets for the size chain you are using bolted to the hub. Weld the frame/dropouts to the ATV or to a plate that you then bolt to the ATV (easier for alignment purposes). Then offsett the motor to one side enough to line up with one side of the hub while the other side of the hub lines up with the wheel's sprocket.

If a dual-disc bike hub isn't availalbe, but an old steel hub (ancient ten-speed bike, etc) with large enough flanges is, you could bolt the sprockets to new holes drilled in the flanges, or you could weld them to the flanges directly.

Yeah regarding the batteries, I will just build it myself. I can get INR18650-25R Samsung Lithium Ion 3.6V 2.5Ah with 20A max discharge rate for about $5 each (about $3.65 USD), so 14s2p will give me 40A max current at 48v roughly fairly cheap. Hopefully they'll be ok. Might need to add another 13 batteries to get longer lasting time but that's a fairly easy addition I can do.

With the speed of the ATV, it's definitely wayyyy too fast. If I lift the back wheels up and do full throttle, it looks as if the wheels are going to fly off. I haven't really properly tested it on flat though for too long but seems to go good enough, but too fast obviously for a kid.

The more I think and read about it, a jackshaft is probably the way to go. You make it sound easy but I'm just an IT guy that can't do any welding...! Any kind of structure additions or modifications are pretty hard for me. My father on the other hand is good with all that. He did the engine mount and can do basically anything else in that department but I need to learn how to do this stuff myself

justaperson76 said:

Yeah regarding the batteries, I will just build it myself. I can get INR18650-25R Samsung Lithium Ion 3.6V 2.5Ah with 20A max discharge rate for about $5 each (about $3.65 USD), so 14s2p will give me 40A max current at 48v roughly fairly cheap. Hopefully they'll be ok. Might need to add another 13 batteries to get longer lasting time but that's a fairly easy addition I can do.

Click to expand...

You'd have to add 14.

Some test info for those cells, if it's helpful:
https://lygte-info.dk/review/batteries2012/Samsung%20INR18650-25R%202500mAh%20(Green)%20UK.html
Pretty heavy voltage sag (nearly a volt per cell, -ish) at the 20A current. (true of most cells if you are pulling high C-rates from them).


Might not matter for your purposes, but with only a 5Ah pack, it's only going to last a few minutes max, at anything like full power levels. If it was only pulling 5A continuously, it would last an hour. If it was 10A, then half an hour. 20A, 14-15m. 40A, 5-7m. (you get less out of the cells at the higher rates, so it's not a straightforward linear thing, plus you probably wouldn't want to actually drain them empty (to help them last longer), so if you are using a conservative LVC, you might get 3 or 4 minutes out of the pack at the higher rate (more if it is just doing that for longer bursts rather than continuous).

The more I think and read about it, a jackshaft is probably the way to go. You make it sound easy but I'm just an IT guy that can't do any welding...! Any kind of structure additions or modifications are pretty hard for me. My father on the other hand is good with all that. He did the engine mount and can do basically anything else in that department but I need to learn how to do this stuff myself

Click to expand...

This particular kind of thing is fairly easy to do, with bicycle parts, because as long as everything is steel, you basically cut and splice things together, and it doesnt' take a lot of designing. It's also fairly forgiving in that if you mess up you can cut and reweld pretty easily. Not like having to redesign and remachine aluminum/etc stuff.

If even I can learn how to do it (using crappy harborfreight welder stuff), I think most anybody can do it. Just takes practice.

If you need inspiration, just look at my CrazyBike2 thread, and my SB Cruiser thread. If stuff that looks like that can do what I do with it without disintegrating....

Oh yes I meant 14 in series, not 13. Yep will definitely need more in parallel. I haven't looked too much in to voltage sag. I might look in to some other batteries, probably the ones I mentioned will do for now. They're cheap... and the whole battery thing isn't complicated for me, compared to....... the jackshaft(!).

So I've been looking in to this and working out how I'd do it. Using an old bike hub, I'm not sure if I can do that as I'd need the axle to probably be a bit longer than what the bike's ones are, as one sprocket (that connects to the motor) would need to go on the outside of the frame, and the other sprocket (that connects to the back axle/wheel sprocket) to be on the other side of the frame.

The best I can think of would be to mount a steel bracket inside the engine bay somewhere and screw some pillow blocks to it, then put an axle through it with sprockets mounted on each side. You can get some kits off ebay that have the pilot sprocket on it, such as this kit: https://www.ebay.com.au/itm/325042453651?hash=item4bae0a3c93:g:sWIAAOSwxDdiBKy8&frcectupt=true - then all I'd need is the big sprocket. I'm not sure how that would connect to it because all the big sprockets have big bores, so I guess you'd need to weld a hub to the jack shaft axle and screw the big sprocket to it. Don't know... tomorrow I'll go to the market and see what I can find. I don't mind mixing and matching and welding myself if I can find all the stuff I need (axle, bearings, hubs, etc.) that can match up. Most of the stuff at the market is cheap.

Your other builds are impressive...! I would love to be able to have the skills to do all that.

Here’s the kit I installed from OMB Warehouse. Am sure you can get them from all sorts of sources local to you …

Yeah that looks good. I could work with something like that.