Voltage drop down converter to 12v on Bike battery pack

[eHomer]

The parts have now arrived, and I've tested the setup.

I apologise for waffling on about batteries for canoes, not very electric bike oriented, but it may be interesting for anyone like me who's into lightweight electric waterborne propulsion, as well as bikes.

So, the object of the exercise was to enable the wider use of my lightweight 36 volt bike lithium iron battery packs to power my 12v electric outboard motor for my kayaks and canoes.

The basic advantages of lithium are vital for canoes as well as bikes, ie, maximum usable capacity at minimum weight, with as long a recharge cycle life as possible.

I have two proprietary aluminum cased 36 volt 10 amp/hour lithium iron packs, with built in BMS, switch and fuse. Each one weighs 3.8 kg. Each is very simple to automatically charge, with the bike's dedicated 36v "block" charger.

I bought the 25amp max Buck voltage dropper via Ebay for £25, (though it actually arrived direct from Amazon, who sell it for £20 with free delivery, I keep forgetting to check with them as well...).

My lithium packs need some protection from splashing and laying in wet, so I adapted a transparent 12 litre "really useful box" where I installed the buck, a wattmeter, and two 30 amp blade fuses, for the input and output.

One or both 36v 10ah battery packs fit snugly inside, connected by a modified Bulgin C14 Right Angle Cable Mount IEC Connector Plug from RS components. One of the photos shows the before and after, where the hood has been cut off, the earth pin removed, and the two pins shortened by 2mm. It fits the battery packs perfectly, although is it's not physically handed, so I've marked the pos and neg very clearly, to avoid inadvertently reversing the polarity.

The box closing clips have been backed up by a surrounding 25mm luggage strap which makes a convenient handle and tie off point when in the kayak. The output is via two 6mm stainless steel studs through the plastic box with wingnuts, that my outboard's crocodile clips attach to.

The other pics show the tank testing (the outboard has to be run in water to get accurate consumption figures from the prop resistance).

The inbuilt watt meter (amazing value still at less than a tenner, even from UK sellers), permanently shows the full 36v power usage direct from the pack, and I temporarily hooked up another watt meter to show the losses through the Buck after conversion to 12 volts.

I have three electric outboards, a Sevylor 18lbs thrust, a watersnake 24lbs thrust, and a Minn Kota Turbo 50 32 lbs thrust.

This table shows the current drawn on two of the motors at each speed, with readings from both watt meters.

Watersnake T24 24ilbs thrust
speed one 2.78a 39.8v 110.5w 36v output
8.85a 12.0v 105.0w 12v input

speed two 7.20a 38.9v 270.0w 36v output
20.0a 11.8v 240.0w 12v input

Minn Kota Turbo 50 28lbs thrust
speed one 2.25a 39.3v 90.0w 36v output
7.2a 12.02v 86.0w 12v input

speed two 2.7a 39.7v 110.0w 36v output
8.7a 12.03v 105.0w 12v input

speed three 3.9a 39.5v 157.0w 36v output
12.3a 11.94v 147.0w 12v input

speed four 5.7a 39.4v 226.0w 36v output
17.9a 11.8v 210.0w 12v input

speed five 31a (test stopped to avoid overloading the Buck)

reverse one 7.8a at 12v
reverse two 9.1a at 12v
reverse three 25.0a at 12v

I think it shows the efficiency losses are quite minimal, and the 25 amp buck is sufficient for all of the settings apart from the maximum speed setting 5 on the Minn Kota. The Buck heatsink remained almost stone cold throughout the tests.

So each of my bike's 10ah batteries should give me around three hours at speed one on my T24, or about an hour and a half at speed two.

Speed one is normal paddling speed, around 4mph, and speed two is about 6mph.

 

[Deleted member 4366]

The motor uses resistors to control the power. Level one has the highest resistor and so on until full power has no resistor. That means that the efficiency is very poor on anything but maximum setting.

Instead of the buck converter, you could use a 36v brushed motor controller (with reverse switch) that has the correct maximum current directly on the supply wires, which would make your present inefficient speed control system redundant. You would need to leave your present motor control set to maximum, or remove it completely. The motor can manage 18v and lets say 15 amps, which is 270 watts. At 36v (actually about 40v) you'd need 7 amps to get 270w, so you need a 36v controller that limits the current to 7 amps.

If the controller has a potentiometer, you can limit its output by adding a resistor in series with it on the high voltage side. Say it were a 10K pot. A 10K pot in series would half the maximum signal.

Alternatively, you can cut down the current of a higher current controller, by crimping or shaving the shunt. A wattmeter would help determine the actual output current.

http://www.ebay.co.uk/itm/DC-12V-24V-36V-48V-PWM-Motor-Speed-Control-Switch-Controller-/232090559885?hash=item3609ace58d:g:qwkAAOSwubRXKaz4

 

[Geebee]

The motor uses resistors to control the power. Level one has the highest resistor and so on until full power has no resistor. That means that the efficiency is very poor on anything but maximum setting.

Instead of the buck converter, you could use a 36v brushed motor controller (with reverse switch) that has the correct maximum current directly on the supply wires, which would make your present inefficient speed control system redundant. You would need to leave your present motor control set to maximum, or remove it completely. The motor can manage 18v and lets say 15 amps, which is 270 watts. At 36v (actually about 40v) you'd need 7 amps to get 270w, so you need a 36v controller that limits the current to 7 amps.

If the controller has a potentiometer, you can limit its output by adding a resistor in series with it on the high voltage side. Say it were a 10K pot. A 10K pot in series would half the maximum signal.

Alternatively, you can cut down the current of a higher current controller, by crimping or shaving the shunt. A wattmeter would help determine the actual output current.

http://www.ebay.co.uk/itm/DC-12V-24V-36V-48V-PWM-Motor-Speed-Control-Switch-Controller-/232090559885?hash=item3609ace58d:g:qwkAAOSwubRXKaz4

d8veh, any issues with the motor seeing 36v instead of 12v from the PWM?

 

[eHomer]

The motor uses resistors to control the power. Level one has the highest resistor and so on until full power has no resistor. That means that the efficiency is very poor on anything but maximum setting.

Instead of the buck converter, you could use a 36v brushed motor controller (with reverse switch) that has the correct maximum current directly on the supply wires, which would make your present inefficient speed control system redundant. You would need to leave your present motor control set to maximum, or remove it completely. The motor can manage 18v and lets say 15 amps, which is 270 watts. At 36v (actually about 40v) you'd need 7 amps to get 270w, so you need a 36v controller that limits the current to 7 amps.

If the controller has a potentiometer, you can limit its output by adding a resistor in series with it on the high voltage side. Say it were a 10K pot. A 10K pot in series would half the maximum signal.

Alternatively, you can cut down the current of a higher current controller, by crimping or shaving the shunt. A wattmeter would help determine the actual output current.

http://www.ebay.co.uk/itm/DC-12V-24V-36V-48V-PWM-Motor-Speed-Control-Switch-Controller-/232090559885?hash=item3609ace58d:g:qwkAAOSwubRXKaz4

Thanks for the constructive post d8veh, I understand most of what you've said, and certainly do want maximum efficiency with no wasted stored battery power.

IR722 had suggested a PWM controller too, and I did get one of those to experiment with too, http://www.ebay.co.uk/itm/Motor-Speed-Control-2000W-10-50V-40A-MAX-DC-PWM-HHO-RC-Controller-12V-24V-48V-UK-/161975055276?hash=item25b67733ac:g:sGYAAOSwDuJWvFZ2

That one was rated at 40 amps max, so I could use it with my Minn Kota at max speed 5 too, rather than the 16 amp PWM.

I like your idea of placing a fixed value resistor in series with the pot to fix the maximum output voltage, rather than a rotary position marking or physical rotary stop.

As regards the resistors built in to the standard outboard motor speed switches, I'm not sure, but I believe that some do use PWMs already, which Minn Kota market as their "maximiser" circuitry, which I gleaned from a US boating forum at https://www.glen-l.com/phpBB2/viewtopic.php?t=7680

I also read elsewhere, though I can't find it again now, that my two small canoe outboards don't use resistors either. They each only have the two fixed speeds, and I think they're achieved by two sets of motor windings. When I looked inside at the wiring, the speed switch has two live wires that both drop down the tube to the motor, with no resistors in the head casing.

 

[Geebee]

with no resistors in the head casing.

I think they tend to put them in the motor casing for cooling purposes.

 

[Deleted member 4366]

d8veh, any issues with the motor seeing 36v instead of 12v from the PWM?

No. It sees blocks of charge coming at it. It doesn't matter whether they're tall and thin (high volts and low pwm) or short and fat (low volts and high pwm). It's the power that counts, which is the total area contained in the blocks.

 

[Deleted member 4366]

This is how I see it. The cheapo electric outboards use resistors and the throttle clicks into stops so you get about 4 fixed power levels and reverse. The expensive ones use a pwm controller that has no indexing and the power is always proportional throughout the throttle range like a petrol motor.

 

[IR772]

The parts have now arrived, and I've tested the setup.

I apologise for waffling on about batteries for canoes, not very electric bike oriented, but it may be interesting for anyone like me who's into lightweight electric waterborne propulsion, as well as bikes.

So, the object of the exercise was to enable the wider use of my lightweight 36 volt bike lithium iron battery packs to power my 12v electric outboard motor for my kayaks and canoes.

The basic advantages of lithium are vital for canoes as well as bikes, ie, maximum usable capacity at minimum weight, with as long a recharge cycle life as possible.

I have two proprietary aluminum cased 36 volt 10 amp/hour lithium iron packs, with built in BMS, switch and fuse. Each one weighs 3.8 kg. Each is very simple to automatically charge, with the bike's dedicated 36v "block" charger.

I bought the 25amp max Buck voltage dropper via Ebay for £25, (though it actually arrived direct from Amazon, who sell it for £20 with free delivery, I keep forgetting to check with them as well...).

My lithium packs need some protection from splashing and laying in wet, so I adapted a transparent 12 litre "really useful box" where I installed the buck, a wattmeter, and two 30 amp blade fuses, for the input and output.

One or both 36v 10ah battery packs fit snugly inside, connected by a modified Bulgin C14 Right Angle Cable Mount IEC Connector Plug from RS components. One of the photos shows the before and after, where the hood has been cut off, the earth pin removed, and the two pins shortened by 2mm. It fits the battery packs perfectly, although is it's not physically handed, so I've marked the pos and neg very clearly, to avoid inadvertently reversing the polarity.

The box closing clips have been backed up by a surrounding 25mm luggage strap which makes a convenient handle and tie off point when in the kayak. The output is via two 6mm stainless steel studs through the plastic box with wingnuts, that my outboard's crocodile clips attach to.

The other pics show the tank testing (the outboard has to be run in water to get accurate consumption figures from the prop resistance).

The inbuilt watt meter (amazing value still at less than a tenner, even from UK sellers), permanently shows the full 36v power usage direct from the pack, and I temporarily hooked up another watt meter to show the losses through the Buck after conversion to 12 volts.

I have three electric outboards, a Sevylor 18lbs thrust, a watersnake 24lbs thrust, and a Minn Kota Turbo 50 32 lbs thrust.

This table shows the current drawn on two of the motors at each speed, with readings from both watt meters.

Watersnake T24 24ilbs thrust
speed one 2.78a 39.8v 110.5w 36v output
8.85a 12.0v 105.0w 12v input

speed two 7.20a 38.9v 270.0w 36v output
20.0a 11.8v 240.0w 12v input

Minn Kota Turbo 50 28lbs thrust
speed one 2.25a 39.3v 90.0w 36v output
7.2a 12.02v 86.0w 12v input

speed two 2.7a 39.7v 110.0w 36v output
8.7a 12.03v 105.0w 12v input

speed three 3.9a 39.5v 157.0w 36v output
12.3a 11.94v 147.0w 12v input

speed four 5.7a 39.4v 226.0w 36v output
17.9a 11.8v 210.0w 12v input

speed five 31a (test stopped to avoid overloading the Buck)

reverse one 7.8a at 12v
reverse two 9.1a at 12v
reverse three 25.0a at 12v

I think it shows the efficiency losses are quite minimal, and the 25 amp buck is sufficient for all of the settings apart from the maximum speed setting 5 on the Minn Kota. The Buck heatsink remained almost stone cold throughout the tests.

So each of my bike's 10ah batteries should give me around three hours at speed one on my T24, or about an hour and a half at speed two.

Speed one is normal paddling speed, around 4mph, and speed two is about 6mph.

That is a very tidy home engineered project.

You have used materials to suit your purpose and set out everything in an ordered safe and usable manner.

Really well done. Pat on back.

 

[eHomer]

That is a very tidy home engineered project.

You have used materials to suit your purpose and set out everything in an ordered safe and usable manner.

Really well done. Pat on back.

Thanks IR772.

 

[eHomer]

I think they tend to put them in the motor casing for cooling purposes.

Thanks Geebee, that sounds logical.

 

[eHomer]

This is how I see it. The cheapo electric outboards use resistors and the throttle clicks into stops so you get about 4 fixed power levels and reverse. The expensive ones use a pwm controller that has no indexing and the power is always proportional throughout the throttle range like a petrol motor.

Yes, I guess my cheapo Watersnake t24 does have a resistor for the low speed one setting, probably mounted at the bottom of the leg in the motor housing, as Geebee suggested.

I'm still mulling over all the possibilities now, and I like to keep my kit as versatile as possible. My new "bike battery" box with reducer gives me a dedicated 12v output at 95% efficiency with any of my outboards, and my three outboards are completely unmodified, so each could still be used with that new bike battery box or with my 12v 85 amp lead acid leisure battery, (when used in the Canadian canoe or rubber dinghy, which have a much greater weight carrying capacity than the kayak).

If I did fit the PWM to get greater low speed efficiency, I guess I could permanently fit one in each of my outboards, inside the head unit, replacing the switches, making each of them still self-contained 12v units, with continuously variable PWM speed controls.

I know it maybe sounds like overkill, with the 36v bike battery then passing through both the Buck dropper and the PWM, but it would mean that I could still use any motor with any battery.

Thanks for all the help and ideas, it's really handy to tap into wider knowledge and opinions.

 

[eHomer]

I've just had a worrying thought that may render my project useless.

Here I've been, confidently citing my ready-made lithium iron battery pack's BMS as automatically monitoring the discharge, but I may be quite wrong.

Was I right in assuming that the battery pack itself has got a low voltage cutoff safeguard ?

I'm confident that it has some sort of inbuilt BMS for charging purposes, because the included charger had just a simple coax power plug, with no balance lead connections.

It's now occurred to me that I've always used it with BMS Battery sinewave controllers, which I understand have their own LVC function, to avoid over-discharging the lithium cells.

If I'm wrong, and the battery casing itself does not cutoff power at the safe LV for the lithium iron cells, then my Buck voltage dropper could suck my 36v battery all the way down to 12v before I notice anything.

I guess the safest way to test it would be to go for a long run on the bike until the controller cuts out at LVC, then see if it lets any power through to my buck dropper when connected in the box.

 

[IR772]

It would be unusual for a large battery pack not to have its own BMS built in. This charges each battery individually then uses them in a series and parallel arrangement to get 36 volts, depending how many are in there.

If the pack were batteries only, you could discharge the batteries below V min when using your bike and they would not normally recharge, expensive.

So unless its the nastiest pack of batteries available everything will be fine.

I would ditch the buck and use the PWM when it comes, 5 % is quite a loss to then use PWM as well.

If you are concerned about the voltage stop not working, then measure the pot at the defined points and replace it with fixed resistors on a two position switch. Easier for mounting and waterproofing compared to the pot.

But if you do not want to start de-soldering etc then carry on as before.

I would use the pot as that gives you a variable throttle and you can overclock it to, would 18 volts lift the nose out of the water?

Also if you do not tinker with the PWM controller that will still work on 12 or 24 volt Lead acid batteries.

 

[Deleted member 4366]

Out of all the batteries that I've tested, the BMS didn't shut off in about 20% of them, so you should check it. The wattmeter will tell you the voltage won't it?

 

[eHomer]

Ok, thanks IR772 and d8veh, I'll run another test with the outboard in the tank. This time I'll run it until it either cuts out automatically (hopefully ), or until the pack voltage drops to minimum safe level, by keeping a carefull eye on the wattmeter, as you say d8veh.

What would you say is the safe lower combined voltage limit for the 36v Lithium ion pack ?

Going off on another tack completely, I'm still also tempted with the Hobbyking multistar lipos that are still on big reductions.

The 1600mah 4s at £38.99 are tempting (damnit).

https://www.hobbyking.com/hobbyking/store/uh_viewitem.asp?idproduct=68821

Three of those have a volume of only 173 x 74 x 135, and weigh only 3.9 kg,

....and in parallel they would give me 48 ah at 4s for the kayak motor, (with the slightly extra voltage and no converter losses).

...and in series they would give me 16 ah of 12s for my bike.

I could easily make up the two XT90 wiring looms, one in series and one parallel.

I know it contradicts a lot of my justification for adapting the Lithium ion packs, but that's so much versatile energy packed in so small a space and weight. It would fit very neatly in a 2.3 litre "lock & lock" sandwich box too....

 

[IR772]

I think 2.7 volts per cell is as low as you want to go so 10.8 v for the hobby kings!

Lock & Lock seems a useful limiting factor to me as well.

Grab the hobby king batteries as they are good value.

At 14 volts PWM all the way with no tinkering, good bye bike batteries.

As d8veh says there are no golden rules.

 

[Deleted member 4366]

I think lipos are yesterday's technology. For Ebikes, the modern cells are much smaller and lighter. A 50 cell 36v battery would weigh about 2.5kg and give 17Ah and 50 amps. Although more expensive than lipos, they last about three times as long. Have you seen the 52v Mighty Mini battery from Luna? It can nearly fit in your pocket, and it can take you 30 miles with its 350 watt-hours.

When draining down your battery, the BMS should trip somewhere between 30v and 28v. It will be going down rapidly then, so watch it carefully. Don't let it go any lower.

 

[eHomer]

Prompt and very useful feedback again, IR772 & d8veh, thanks.

I'll time the discharge test and watch it like a hawk when it gets down near 28v if it hasn't already shut off.

On the Lipo front, each time I think I've nearly got it cracked, I then find out I'm outdated again...

I really thought Lipos were the current best compromise choice for power,weight,cost, albeit with strict observance of safety when charging and discharging.

I googled "mighty mini battery",

https://www.google.co.uk/search?q=52v+Mighty+Mini+battery&client=firefox-b&source=lnms&tbm=isch&sa=X&ved=0ahUKEwjoqqPqmKrPAhVHK8AKHWRtCVEQ_AUICSgC&biw=1195&bih=641#imgrc=_

....and was somewhat stunned to see my sandwich box 16ah lipos idea (which I really hadn't seen before ) , in the results about half way down the page, at https://electricbike-blog.com/2016/04/18/use-your-cordless-power-tool-batteries-to-power-your-ebike-what-could-possibly-go-wrong/

...anyhow, back to the mighty mini, I read this:

https://electricbike-blog.com/2016/05/08/the-most-power-youll-ever-be-able-to-fit-in-your-pocket-6ah-3-3-lb-30q-luna-mighty-mini-cube-pack-reviewed/

So it's made out of 18650 Polymer Lithium Ion cells.

I believe my ready-made 36v 10ah lithium-ion batteries contain 18650 ion cells, but presumably of an older, less efficient type ?

My limited understanding up to now that Lipos were more efficient weight and cost-wise than Lithium ion, so does this mean that ion has been improved further ?

I still need something like an "idiot buyer's guide" to lithium battery technology, even though I've now grasped some of it, like 4s,2p, balance leads, LVC etc...

 

[IR772]

A good thing for any project is to keep your eye on the cost.

Value wise your battery choice is spot on, its going in a boat! As they are a product for the modelling world there are stacks of accessories for them.

Sit down, decide what you want to do with the project, then draw a line and build it.

There will always be something new, its just if you want to pay for it or not.

I think you have the way forward, value, practically and reliability are all included. Plus a lovely sandwich box, those under hand cheats to steal your idea. Next thing they will be using empty ice cream tubs.

 

[Deleted member 4366]

Five years ago, lipos were a cheap way to get a light high power battery, but the Samsung 22P cell changed all that. Now every manufacturer of 18650s has high power versions. Each year, the power and capacity goes up a step, so we now have cells that weigh 45g capable of 10 amps and 3.5Ah, which means that a typical 36v 40 cell battery will weigh 2kg, give 14Ah and be able to supply 40A continuous. The equivalent lipo, like your Multistars will weigh double that at half the price and half the life.

Apparently, all the battery makers are confident that within a couple of years, batteries will be double the capacity for the same size and weight. At the moment, an astronomic amount of money is going in to R&D for automotive applications, so we have lots to look forward to.