Electric bike project (complete/questions)

Excellent post.
Off the top of my head I'd have thought reving the motor faster would be more efficient.
Surely the way to find out is to run some tests?
Run at a known road speed in the various gears whilst measuring the current.
(You may need to build a simple rolling road for this)
It's great to see a post from some one who has actually done something.
Kudos...I shall watch the thread with interest.
(BTW Narrow profile racing type tyres inflated hard may help considerably)
Del

Your question really boils down to "at what speed is the electric motor most efficient".

If this is a BLDC motor then it has a wide range of RPM where the efficiency is the same.

One of the best ways to answer the question is to try to obtain some current v/s power (torque and RPM) graphs.

Since you can change gear ratios and are able to monitor current, you are probably in the best position of all to determine what the best gear ratio is to minimise current drain for a given speed.

This motor is a brushed one, so it probably has a narrower efficiency range. It seems to me to use less amperage with lower rpms and higher gears...but then it lacks torque. I try to ride it conscientiously and minimize the amperage based on how I twist the throttle, but still I only get about a half-hour high performance ride out of it.

The guys that play with brushed DC motors a lot fiddle around with brush timing. By advancing the brush holders along the direction of rotation, you move the performance curve up the RPM graph. I don't know whether this is possible with your motor or even if you want to go down this path. A brushless permanent magnet motor would give you higher efficiency than the brushed one if that is your goal.

I may go with a brushless eventually, but for now I'm going to stick with what I have...since the money has run out Lol

Beautiful work. You do not mention your operating voltage, which is necessary to calculate the actual power you are consuming. Your maximum current draw will occur when you are accelerating (either from a standstill, or when climbing a hill), and will depend on the total weight you are accelerating (you and the bike). Your current draw while maintaining speed will be a function of the energy losses (friction, wind resistance, etc). If you want to use less current, go slower, or reduce the weight of the system.

To determine the frictional losses, find a flat stretch of road, accelerate to speed, then time how long it takes to coast with no power to a standstill and measure the distance. You now have a change in velocity with respect to time, which is acceleration (negative, in this case). From Newton's Law, F=ma, you can determine the losses (except total wind resistance, which is a whole other ball game, unless you only ride in still air. Some of the losses in this calculation will actually result from wind resistance, but determining the variation due to wind resistance is beyond this quick tutorial). Now multiply the force calculated from Newton's Law by the distance it took to stop, and this is how much "work" you have to do to keep the bike moving at a constant velocity. If you worked this out in the metric system, you know have a number of joules that you have to add just to maintain speed. One joule per second is one watt, and one watt is volts times amps.

You can use pretty much the same calculation to give you an idea of how much more energy you will need to accelerate from a standstill to your cruising speed. This should tell you how much current you should be drawing.

Rather than thinking about changing the RPM's of the motor, you should be looking at ways to reduce the weight of the system, or improving the bearings you are using, or reducing the contact area between road and tire (i.e., skinny tires inflated hard, as per Dell). Changing the motor speed is going to have very limited effect on the actual power consumption- unless the chain drive is a bigger source of loss than I think it is...

Thanks for the tip, and I think its probably best for me to just simplify my "assumptions" about the system and do a translational analysis, like you said, instead of a full rotational vector analysis. It makes more sense that way. Luckily I do have a flat stretch of road and if the wind is dead out...I will lump that factor in with systemic friction.

The operating voltage is 24v nominal. The system usually charges to 27.4 and is considered drained at about 25.1 or so. I don't like to drain it below that because that is what the "battery university" suggests to stay above. Most of the time I draw from 10 to 30 amps depending on if I am accelerating or not.

I don't think the chain drive uses too much... I tested the motor's running amperage just alone with no load, and then I have ran it while just driving the crankset. The amperage went from 2.0 to about 2.5-3 amps at most. This is a brushed motor by the way.

As far as reducing the weight goes...I don't think there is much I can do. The mounting system is about the minimum possible without actually going carbon fiber or something. Plus I am running SLA batteries...which total to about 26.4 pounds. I am actually losing weight off myself to make up for that :)

Since I've for 3 months a bicycle with an assistant electric motor, my questi0n is if you are pedaling while you bike?

An important role is the ability to listen to the motor and to your body,then you can have long distance biking including hill climbing.

Not much. I have to install some chain guides before I can pedal assist in the higher gears...otherwise the chain comes off! Up hills it works well...and actually uses a little less than 10 amps most of the time. I can't wait to be able to pedal in the higher gears...but right now its not really an option.

Good work,

Which chain falls off?

I like these No More Flat inner tubes

Wait for it, all the bicycle guys sneering in unison...

Use the next larger size, use lots of baby powder to install.

The ride is kind of harsh, but the rolling resistance low & never having a flat is worth it.

Around here there are lots of thorns that make short work of even the puncture resistant tubes.

Right now it draws from 15 to 30 amps while riding and has a top speed of 22 mph despite my ability to shift gears using the drive motor. I am thinking now that if I reduce the RPMs of the motor more, that I will get more torque, better acceleration, and perhaps use less amperage.

If the bike has the same top speed in two different gears, then you have shown that the top speed is limited by motor power. For the motor to produce 900 watts you'd need 36 volts at 25 amps. Depending upon the motor's current limits, you could also use 24 volts at higher amperage... or, depending upon its voltage limits, lower amperage (18.75) at 48 volts.

Changing gearing so that the motor will turn slower for a given road speed will do the opposite of what you want -- it will increase amperage, not reduce it. It will also give you less rear wheel torque, and poorer acceleration. This assumes that your motor has fairly typical dc motor characteristics. If your motor is AC, (which I doubt, unless you made your own controller) then the torque curve is such that there is substantially more torque at low motor speeds than at high (and finding the ideal gearing is a little trickier). But for a typical DC motor, torque is a function of current (double current to double torque) and speed is a function of voltage (double voltage to double speed). DC motors (especially permanent magnet ones) will often have published torque constants (e.g. lb-in /amp) and speed constants (e.g. rpm per bolt).

Very nice job putting this all together.

The motor is 900W but rated for 24v and a top amperage draw of 34 amps. I know I can increase the voltage but I think that will reduce the motor life since it is rated for 2600 RPM at 24v. Also the controller I purchased is rated at 24v and I don't have a schematic for the circuit so I don't really want to try to analyze the circuit for 36v. I would have to purchase another controller. When I say more torque I meant to say more mechanical advantage earlier too... but what I want is better acceleration so I will look into reducing the motor's rpm more.

Basically I estimated that the typical biker pedals at just under 100 rpm at the crank...and since a human being typically can produce about 100W...applying 900watts of power at the same mechanical advantage would give some very nice acceleration!

I eventually plan to string together some nickel-metal hydride cells to power this thing, and that should reduce the weight quite a bit...so that will help. I'm afraid to make my tires any smaller because they may not support the weight of the batteries and myself...esp. if there is any shock by bumps in the road.

Thanks for the info! Your post was really useful.

Another question for you...

Using the junction rule...wouldn't a bunch of smaller lead acid batteries hooked in parallel also reduce the stress of high amperage draw? If I were to have 2 pairs of 12v batteries hooked in series for the 24v, and then hook those together in parallel, since each pair (if I am drawing 20 amps at speed say...) would only be giving 10 amps instead of one pair giving 20? Do you think that would reduce the effect of high amperage draw?

Yes, greater AH capacity will reduce the undesirable effect of high amperage draws. The amperage draw is spoken of as "xC", ( e.g. .5C. 1C, 2C etc.). A 10 amp hour battery, discharging at a rate of 10 amps is said to be discharging a 1C. If it is discharging at 20 amps, then the rate is 2C, etc. For all chemistries, is is good to avoid high C rates, but some batteries are much better at dealing with high rates.

In general, the best way to get a particular voltage and AH capacity from lead acid batteries is with the smallest number of batteries. For most electric vehicles, batteries are simply hooked up in series. So if you wanted a 24 v 60AH pack (for 1440 watt-hours) the best approach is to use two 12V, 60AH batteries in series (rather than, for example, using four 30AH batteries with a parallel/series arrangement. Other things being equal, the two packs will behave identically, under normal conditions. However, when the pack becomes unbalanced, a low voltage battery in parallel with another will drain the better battery. In the same way, if one battery in parallel is not just low voltage but has, for example, a shorted cell, the better battery can be permanently damaged by the paralleled bad battery. (Keeping batteries in balance is important when they are in series, too -- you can read up on battery management systems, which serve this function.)

Space constraints may dictate four smaller batteries in a series/parallel arrangement. In that case, there is no catastrophe brewing, just a higher likelihood of possible problems occurring. Of course, the fewer the connections the better also.

With some other chemistries, it can be hard to achieve desired goals with a small number of cells... and there is a lot of variation in the design intent of the cells, with some being great for high power output but not so great for high energy capacity, and vice versa. (In lead acid, the first type is usually called a starting battery, and the second is called deep cycle.)

Ok so by hooking 4 12v 10ah batteries up together ...2 pairs in parallel to achieve 24v and 20ah... I am gaining the advantage of each set of batteries having less amperage drawn from them...but I am also adding the risk of the batteries damaging each other if I were unable to detect the faulty cell in time. Wow batteries are totally catch 22.

Yep, you've got it*. But if you have a monitoring system, or simply have calibrated voltmeters for each battery, or a single voltmeter with a selector switch, you can keep track of them. If you go to the EVDL (electric vehicle discussion list) and search the archives for battery balancing, you will find lots of info -- probably too much for your purposes.

*At least I think you do... your setup would then act exactly like 2) 20 AH batteries in series (but have the slightly greater potential for the problems with balancing I mentioned before). But don't panic, rebalancing is just a matter of keeping your eyes on things and selectively charging to keep all the batteries at equal voltage. The EVDL does give suggestions for battery balancing and monitoring systems that you can make pretty easily.

Imbalance actually occurs at the cell level, and almost noone actually tries to balance lead acid batteries at the cell level -- even though it can be done with flooded batteries. (2 20AH batteries have a total of 12 cells in series.) EV people can be obsessive about making sure that all the batteries in their pack were manufactured on the same day or week. That is a little over the top, for your purposes, I think.

I have an electrical fundamentals class coming up this year in which we will have to build a "final project." Perhaps I will be able to build a monitoring/protection circuit that will shut down a pair of batteries if they are showing a significant imbalance.

My setup (desired) would be something like this:

I probably won't switch up my configuration until I build a monitoring/protection circuit for this bike. Right now I simply have 2 18AH batteries in parallel running the system. If I do no pedaling I can go about 7-10 miles on flat ground...driving normally (not conservatively).

Hi Blink,

I wanted to ask you again about the parallel-series arrangement. Diodes. Do you think I could find a set of diodes heavy-duty enough to prevent one set of SLAs in a parallel arrangement from charging the other set if there were an imbalance? My intent would be to install a monitoring system to keep an eye on the pack balances.

However, if I'm out on a trip and riding along, and I want to keep going without dealing with it right away, I would want to keep one set of batteries from being drained just to charge the "worse off" set. At the same time I would still want to use up what energy I could from the imbalanced pair.

After getting home I would be able to analyze the imbalance and replace a bad battery or the pair entirely. But I was thinking diodes would be the simplest solution to a parallel setup. This would enable me to "travel lighter" if I wanted to only go 10 miles or so vs a 25 mile trip. I think that a parallel arrangement would extend the life of the batteries also by reducing the average level of draw through each battery from about 15A down to about 7.5A. That would in turn reduce the capacity-reducing effect of high amp-draw.

Let me know if there is any merit to this thinking... also I haven't priced diodes and I don't know enough about them to know if I can find an inexpensive pair that will handles up to 30A draw through them. I also don't know what the resistance characteristics of a P and N semiconductor material arrangement are...but further research into on wiki or something should also help clear that up for me.

Try Digikey

Hi

in connection with your 30 Amp Diods requirement,

You can easily use ready made diode rectifier bridges.

You can use KBPC35 bridge for 35 Amps and KBPC50 for 50 Amp. Rectifiers.

They are universal and cheaper than any specific power diodes. Square shape and smaller than a matchbox.

They are very suitable for easy assembly, no need any specific insulation. They are ready to use. Their polarization voltages are quite low, of which brings you some benefits, such lower voltage on diodes and higher voltage to your motor.

Regards

Good work. Excellent.

Chain drive at the output of such an electric motor of turning very fast?

Chains are made of steel, having heavy masses. They are Ok. at low speeds. However at high speeds, they try to enlarge due the centrifuge force. Therefore trying to resist for the power, of which trying to turn them.

I advise, leave the other transmission sections as chain drive, because they turn slower. but only replace the drive from motor to the other gear with a pulley and belt system. Even Moped's are working like this. They design their transmission chain as to be High speed section with pulley and belt, and low speed with chain and gears.

Since this design I have revised it and reduced the effective motor rpm even more. The primary jackshaft sprocket was a 40 tooth before, and is now an 80 tooth. So now the end result is that the RPM at the pedals has gone from 360 RPM down to 180 RPM. The net effect of doing this isn't exactly as powerful as I had hoped. Honestly I think my motor is a lot weaker than advertised.

I think it just uses 900w, but doesn't give that much work in the end output.

Good work.

However Pls. bear in mind yhat Chinese electro bikes have one direct Hubmotor at the center of the rear wheel. They are rated to 250 Watts and reaching to 30 Km/h. The total travel time is approx 5 hours. This is what they say in their advertissements.

Possibly your chain and gear transmission consumes more power. Chinese ones are direct driven without chain.

Yeah I found that out after the full build was completed. I thought that by using a chain drive I could increase my top speed by having the ability to shift with the motor drive...but in the end it would have been much easier and much more efficient to just use a hub motor and up the voltage for a higher top-end speed.

This build was definitely a learning experience. Sometime when I am not busy with engineering school I think it would be fun to go back and calculate losses due to the chain drive setup. I could probably do that by measuring the increase in amperage draw from having the motor run with no load, and then just the chain drive system hooked up.

I appreciate your experiments. Lovely.

I always imagined if power transmission from electric motor to the rear wheel could be achieved hydraulically.

A hydro pump, direct coupled to an electric motor, and the batteries in the rear basket. A hydro motor at the rear wheel. Two flexible pressure hoses from the pump to the motor.

Speed, load as well as the ampere current, could spontaneously adjusted by varying the flow rate at the pump. Electric motor runs at a pre determined constant speed, of which maximum effectiveness is achieved.

I am sure you can minimize your losses of chain and gears

Regards

Hydraulics? really

Maybe if the hoses were integrated into the frame & you used some very low friction pumps & motors, all & all adding one more conversion is bound to reduce the efficiency I suppose the interface with the pedals could be done...

I just saw this toothed belt set up from gates

Hoses, integrated in the frame is genious.

Maybe Copper pipes, led through the frame, could be an another solution.

They are strong enough for pressure transmission, since flexibility is not needed.

I impressed of the hyraulic pumps used in the cars for power steering. they are compact and strong enough.

However I have no idea abt. the dimensions for hydro motors.

Regards