"True assist": is anyone else doing this?

[-dg]

But I do have the battery pack I got with my original DD hub motor and it should still have some life in it. It was originally rated at 10 Ah and 48 volts. I could probably use it for such a test and extrapolate from there.

That will work, just measure the actual battery voltage and use that, not the rated voltage in your calculations. But, if you do decide to use the Bafang SWXK or K5, I've already done this test for you, use my numbers for this motor in a 16" wheel from above and you should be good to go.

 

[wayover13]

But, if you do decide to use the Bafang SWXK or K5, I've already done this test for you, use my numbers for this motor in a 16" wheel from above and you should be good to go.

And, earlier in this thread:

In a 16" wheel it would have no-load of 16 mph at 48 volts or 12 mph at 36 volts which is exactly what you want.

And yet earlier in this thread:

Get the motor first and measure the speed with a known voltage and then order the battery with the right voltage to get the no load speed around 12-13 mph. A Ping would be a good fit for this as he can make custom sizes, eg 33 volts last I checked. Since it's a tandem you can afford the space and weight. But anything 20+ah around 30-ish volts will work. This application is not going to demand much from the battery as long as it is big enough to be useful.

Just to make sure I understand what you're recommending. You're recommending a 36 volt, 20 Ah battery if I go with the Bafang motor you've referenced, correct? Or maybe with a 33 volt 20 Ah Ping battery if I want a slightly slower climbing speed, or a 39 volt 20 Ah Ping battery if I want a slightly higher climbing speed. Have I understood correctly?

Another thing I've wondered about along these lines. Can the CA limit battery voltage and thus motor speed? In other words, say one had the motor you're recommending and a 48 volt battery. Could the CA cause the battery to feed the motor 36 of the 48 volts? And, conversely, were a higher speed wanted at some points, be switched so that it feeds the motor all 48 volts? I do not understand very well how the CA works, nor do I understand much better how varying voltages affect the motor, so this may be a daft question. But answers to daft questions can nonetheless sometimes aid in learning.

 

[dogman dan]

With a no load speed of 12 mph, that would be a good motor choice. It should play well with your existing battery too, where you would need more battery to run a bigger motor, If that is, you ran a big controller on it.

The main advantage of the big motor, even when still run on the same low amps controller as the little motor, would be it's ability to soak up heat when you rode up a really long hill, or many short ones. Big covers are big radiators. That's the only real reason, not that he's going to use the power of the big motor.

I just think this way, since I melt a motor fairly regularly. The ones I melted easiest and fastest were tiny gearmotors.

He won't melt one on a "normal" ride. He'll do it on that day he gets a stomach virus or the flu while out on the road. That will be the day his lady is the only one with legs. Time to rely on the motor, and it could melt climbing a large hill.

People here always calculate on a perfect ride. Rarely happens out on the road. Suppose they have a chain break or derailur fall apart? I've had both happen at one time or another.

 

[dogman dan]

With a no load speed of 12 mph, that would be a good motor choice. It should play well with your existing battery too, where you would need more battery to run a bigger motor, If that is, you ran a big controller on it.

The main advantage of the big motor, even when still run on the same low amps controller as the little motor, would be it's ability to soak up heat when you rode up a really long hill, or many short ones. Big covers are big radiators. That's the only real reason, not that he's going to use the power of the big motor.

I just think this way, since I melt a motor fairly regularly. The ones I melted easiest and fastest were tiny gearmotors.

He won't melt one on a "normal" ride. He'll do it on that day he gets a stomach virus or the flu while out on the road. That will be the day his lady is the only one with legs. Time to rely on the motor, and it could melt climbing a large hill.

People here always calculate on a perfect ride. Rarely happens out on the road. Suppose they have a chain break or derailur fall apart? I've had both happen at one time or another.

 

[chas58]

Good point. With a Tandem, strength is more important than weight. Shoot, my pedal tandem loaded for touring is easily over 500lbs.

I tend to go with this. If the bike is going to be used mostly without the motor, a small 2.2kg Cute Q100 with a 14amp controller and a no-load speed of 12mph would be just right.

I think the Q100 is going too far in this direction, the motor I'm suggesting is 3 kg, the BPM which would be my next choice is over 4 kg. I also don't like the Q100 stub axle construction for a tandem, I'm concerned it might not handle the weight.

 

[wayover13]

. . . if you do decide to use the Bafang SWXK or K5, I've already done this test for you, use my numbers for this motor in a 16" wheel from above and you should be good to go.

As I'm looking at sites that sell this Bafang motor, I note that models with differing windings seem to be made. The one listed on the greenbikekit site, for example, says it's rated for 201 RPM's. I've looked at other sites listing Bafang motors and seen this SWXK listed at 300-some RPM's. So, which of these models/windings do you have?

 

[-dg]

As I'm looking at sites that sell this Bafang motor, I note that models with differing windings seem to be made. The one listed on the greenbikekit site, for example, says it's rated for 201 RPM's. I've looked at other sites listing Bafang motors and seen this SWXK listed at 300-some RPM's. So, which of these models/windings do you have?

Good point. The inability of the Chinese vendors to specify at least part numbers or winding codes or some consistent nomenclature for these motors and the various speed grades is hugely frustrating.

The BPM comes with different winding codes that determine the motor speed, and the code is printed on the motor. But, none of the vendors will sell you a "code 10" motor, they will sell you a "36V 26" motor. But the actual motor they ship appears to have an element of luck in it.

The SWXK and SWXK5 and SWXH as far as I know come in a slow wind (the one I think I have), and a fast wind (but I have never seen one). The numbers quoted on the vendor websites are either cut and paste from the Bafang site, or are wildly useless, eg 20kph-35kph for speed. They also don't seem to track the observed speeds very well. Some of them offer different speed grades in a drop down box usually tied to wheel size. Hugely annoying.

However, for your purposes I think it is pretty simple. Order the slow wind if offered an rpm choice, or the 26" wheel if offered a size choice. I assume you will build your own wheel. Since you only want power on the steep hills, the worst case is that it is slower and climbs better than expected and you can always up the voltage if it is too slow. For what it is worth, the motors I used for my numbers are an SWXK from greenbikekit, and and SWXH from elifebikes. Order sensorless if offered a choice but either will work as the controllers will run either way. Get an extra controller and throttle they are cheap and you don't want to wait for a replacement if you short one out or it has a defect. The thumb throttle is ok.

 

[wayover13]

I guess 201 RPM would be the slow wind, then? Some quick and dirty calculations I did--if they're correct--indicate that 201 RPM's in a 16" wheel works out to about 9.5 MPH. I'm not quite sure what the 201 number represents: motor's optimal RPM range? maximum RPM's? Could it represent the motor's RPM's prior to the gear reduction accomplished by the planetary gears inside the hub? Seems unlikely.

 

[teklektik]

At the end of my two-year experimentation with the DD hub motor, this is exactly the sort of system I envisioned as likely most congenial for my needs. I decided at that point that what would be ideal for me would be a mid-drive motor that would match, via various types of gear reduction, my typical pedal RPM range (ca. 95 RPM's) to the motor's most efficient RPM range.
...The system would essentially have, as you suggest, a simple on/off switch; for safety reasons, such as I was envisioning it, some kind of spring-loaded paddle or button switch.
...When assist is wanted, the switch is pressed and helps the rider keep the desired RPM level. When assist is not wanted/needed, the switch is released. Simple but perfectly effective for my uses.
...It could be enhanced somewhat by causing the motor to do what I think they call "ramping," i.e., bringing the RPM's up gradually rather than instantaneously. Of course no such system as this exists, ....

Well, actually, it does - it's called a Cycle Analyst V3.

Another thing I've wondered about along these lines. Can the CA limit battery voltage and thus motor speed? In other words, say one had the motor you're recommending and a 48 volt battery. Could the CA cause the battery to feed the motor 36 of the 48 volts? And, conversely, were a higher speed wanted at some points, be switched so that it feeds the motor all 48 volts? I do not understand very well how the CA works, nor do I understand much better how varying voltages affect the motor, so this may be a daft question.

Please download the V3 Unofficial User Guide and read the overview section and the sections on Pedal Assist in the Advanced Section and in the Tips and Tricks appendix. That thread is a good place to ask details to implement the kind of controls you prefer.

An example of V3 PAS:

• My bike uses gear motors and a simple PAS wheel with an Assist Level knob. The PAS wheel is not as sophisticated as a true torque sensor (which the CA supports) but does vary the assist with pedal rpm. The CA measures my effort in Human Watts and multiplies by the assist knob setting - so if I put out 50W, the CA can be adjusted to add 25W, 100W, or any wattage proportionate to my effort. Set the knob to zero and no assist. The CA ramps up the motors to apply power smoothly according to user configuration. The CA manages the power scaled against Human Watts regardless of battery voltage. I also have a throttle -- if the assist is not enough to make a little rise, hill, or headwind comfortably, I can just tweak the throttle to override the PAS and overcome the brief obstacle before closing the throttle to return to straight 'no hands' PAS operation.

This bike runs flat out under throttle control at 3300W in traffic but with the V3 is easily adjusted to add just 50W of PAS for 'no controls' enjoyment putzing along bike paths at 10mph. As with Kepler, the underlying power of the bike or the way it is used are an entirely separate from the control mechanism than can afford assist proportionate to human effort or the exact proportion (5% or 500%) for which it is configured to assist. The fact that the bike has other modes of operation should also not dissuade you from considering the technology to fulfill your particular needs. Kepler focused on stealth with PAS to retain a bike-like riding experience - my bike is a big moose 40mph grocery-getter that I also use for enjoyment on wooded trails - in some fashion, these both fail to fulfill your vision of 'True Assist', but the underlying CA V3 solution for us is quite likely a solution for you as well...

As others have recommended - get a gear motor. Then hang a V3 on it with a Thun bottom bracket or just a simple PAS wheel and you will have something virtually identical to what you have described above, regardless of the terminology you care to use.

 

[-dg]

I guess 201 RPM would be the slow wind, then? Some quick and dirty calculations I did--if they're correct--indicate that 201 RPM's in a 16" wheel works out to about 9.5 MPH. I'm not quite sure what the 201 number represents: motor's optimal RPM range? maximum RPM's? Could it represent the motor's RPM's prior to the gear reduction accomplished by the planetary gears inside the hub? Seems unlikely.

The 201 is what I have. I think the 201 is supposed to be the loaded speed for the rated voltage, ie 9.5 mph on level ground with 36 volts exactly. Which is about right, I was calculating 12 mph no-load for nominal 36 V (really 37 to 39 depending on battery chemistry and state of charge) and typically unloaded speed is about 15 % to 20% higher then level running speed, so 201 works out to be 11 mph no-load then.

But, all the above is full of assumptions based on the website specifications. There are also some variables on your end, like tire size that will affect speed too. Anyway, I think the 201 with a 36V nominal battery will be in the ballpark. To be sure, order the motor and controller and throttle and maybe a couple spares and then measure it with your battery and order a battery of the right voltage. Or depending on the voltage of your existing battery (what is that anyway), just use it to get a feel for whether the setup meets your needs. If so, get the right battery, if not, sell it on here and move on sadder but wiser.

Personally, I think you are on the right track here and that this will be a success.

 

[wayover13]

I just wanted to post here that I finally put together the kit I bought in late spring, put it on the bike, and have a taken a few test rides. I have to say that -dg really hit the nail on the head with his recommendations about what kit to buy. I bought the 250 watt bafang motor he suggested and the 36 volt, 20 Ah Ping battery. The motor got spoked into my 16" wheel. It performs almost exactly as advertised: on steeper hills, full throttle with a fair amount of rider assistance keeps our pace at 10 mph, give or take. On more shallow grades, full throttle with liberal rider assist keeps us going 12 mph, give or take. This is exactly what we wanted--assist on hills--and it's made our riding, while still strenuous, much more enjoyable. So far I've used the motor to maintain a steady 12 mph pace also when, say, we have a strong headwind. So the net effect is that the kit will give us the help we want up those hills, and help us to keep a very reasonable touring pace that won't drop much below a 12 mph average. We proved this just the other day by going on a 50 mile test ride and doing the distance at a better than 12 mph average: I think we might have used one quarter of the battery capacity on that ride (still learning how to use the CA v3). Thanks loads, -dg, and to all others who contributed to this thread.

We're going on a short tour in the coming days, and I'll be sure to stop in and post further real-world results regarding our kit and bike.

 

[WhatcomRider]

Sweet, WayOver13! Your thread generated a lot of interesting discussions and I am glad to know you got good advice that is working out for you. Looking forward to reading more about your riding experiences.

 

[-dg]

I just wanted to post here that I finally put together the kit I bought in late spring, put it on the bike, and have a taken a few test rides. I have to say that -dg really hit the nail on the head with his recommendations about what kit to buy. I bought the 250 watt bafang motor he suggested and the 36 volt, 20 Ah Ping battery. The motor got spoked into my 16" wheel. It performs almost exactly as advertised: on steeper hills, full throttle with a fair amount of rider assistance keeps our pace at 10 mph, give or take. On more shallow grades, full throttle with liberal rider assist keeps us going 12 mph, give or take.
...
Thanks loads, -dg, and to all others who contributed to this thread.

You should really thank Justin, I just used the nifty simulator tool he provides at ebikes.ca. I'm really glad to hear that this worked out as predicted, I was pretty sure, but it is always nice to see practical results agree with theory. Please do update the thread when you get back from your tour.

 

[wayover13]

Here's a report on our first test tour with the new electric assist system. Once again, I added to my recumbent tandem delta trike a Bafang SWXK 250 watt geared hub motor and a Ping LiFePO 20 Ah, 36 volt battery. I've also got the CA version 3 to help me keep track of battery usage and so forth. The hub motor is laced into a 16 inch front rim. I use a thumb throttle to engage assist when needed.

Our trip was 257 miles over 8 days (2 days off, toward the middle, while my wife attended a conference along our route). Overall, the tour was a huge success. The bike carrier I cobbled together for our ungainly rig (8.5 foot wheelbase, 33" width)--which consists in a strap-down roof rack on which the back wheels sit and a standard trailer hitch bike rack behind, which I modified to hold the front wheel--performed flawlessly. The electric assist kit was a huge improvement over our former, DD motor/kit. The terrain we covered was quite flat (southern Ontario, CA, northeastern OH, USA, and southeast MI, USA), but I'm quite optimistic that even in more hilly or even mountainous terrain, the kit would perform well also. We drove the bike from our home in WI across MI and had our friend drop us off in southern Canada, from whence we biked (and ferried) back to MI.

As mentioned, the assist is mainly for help getting up hills, but it has the added benefit of helping us keep a slightly faster pace when, for example, riding into unfavorable winds. Our longest day was a mere 57 miles, while the shortest was about 37. The CA showed that we used 24.8 amp hours of battery over the 257 mile journey. So the average per mile was about .096--close to .1 amp hour per mile or 1 amp hour over 10 miles.

There was a fair amount of variation by day in how many amp hours we used, dependent, I think, on wind/hill conditions. On the final day, which was our shortest day, for example, we used around 6 amp hours to go 37 miles. By contrast, on our 57 mile day, we used 6.5. On that final day we did have a headwind, so I was using the assist quite a lot to help us maintain about a 12 mph pace. That was also the day on which we encountered the hilliest portion of our route (hilly by flat, midwestern standards).

That said, the trip was really a huge success, and I was quite gratified that everything came together so well (and that the weather was pretty agreeable). Thanks loads to all who contributed to this thread and who have advised me in other threads on this forum. I couldn'a done it without ya!

A final query: just how many amp hours can I drain from my battery before I run the risk of damaging it unless I recharge? Or, in general, how do I tell when it's reached that point?

 

[-dg]

A final query: just how many amp hours can I drain from my battery before I run the risk of damaging it unless I recharge? Or, in general, how do I tell when it's reached that point?

That trip sounds like fun. Best thing I ever did was ride from Berkeley, CA to NYC back in the 80s when I was young(er) and fit(er). Glad everything is working. Nice rig too.

The usual rule of thumb is not to routinely discharge over 80% of capacity to get the longest possible life. So, ideally about 16Ah from your 20Ah pack. But, since the Ping is LiFePO, it should be good for a couple thousand cycles. Unless you plan on touring an awful lot it will probably last effectively forever, so I think the occasional full 20Ah discharge is acceptable. The BMS will cut the battery off before you actually damage it.

If it were me, I'd try to keep the battery about 2/3 charged most of the time and not store it fully charged or especially not mostly discharged. It's a good idea to check the voltage and possibly charge it a little every couple months if you are not using it. On the road, i'd charge as much as convenient whenever I was near a plug and on a day with significant mountains maybe plan a break for charging somewhere if you anticipate a lot of climbing later in the day. For planning I'd propose 16Ah or less, but if I had to use 18 or 19, or even all of it due to wind, tired, whatever I would not worry about it. As the battery ages it will lose a little capacity, so you will eventually need to de-rate to allow for that, but if you are only touring a few weeks a year, that could be many years from now.

At the risk of puffing myself up till I explode, I'm really thrilled that you took my advice and that it worked out as predicted.

 

[arkmundi]

I just wanted to post here that I finally put together the kit I bought in late spring, put it on the bike, and have a taken a few test rides. I have to say that -dg really hit the nail on the head with his recommendations about what kit to buy. I bought the 250 watt bafang motor he suggested and the 36 volt, 20 Ah Ping battery. The motor got spoked into my 16" wheel. It performs almost exactly as advertised: on steeper hills, full throttle with a fair amount of rider assistance keeps our pace at 10 mph, give or take. On more shallow grades, full throttle with liberal rider assist keeps us going 12 mph, give or take.
...
Thanks loads, -dg, and to all others who contributed to this thread.

You should really thank Justin, I just used the nifty simulator tool he provides at ebikes.ca. I'm really glad to hear that this worked out as predicted, I was pretty sure, but it is always nice to see practical results agree with theory. Please do update the thread when you get back from your tour.

Thanks Justin!

Here's a report on our first test tour with the new electric assist system. Once again, I added to my recumbent tandem delta trike a Bafang SWXK 250 watt geared hub motor and a Ping LiFePO 20 Ah, 36 volt battery... As mentioned, the assist is mainly for help getting up hills, but it has the added benefit of helping us keep a slightly faster pace when, for example, riding into unfavorable winds. Our longest day was a mere 57 miles, while the shortest was about 37. The CA showed that we used 24.8 amp hours of battery over the 257 mile journey. So the average per mile was about .096--close to .1 amp hour per mile or 1 amp hour over 10 miles.

That's a fairly extraordinary result. Appreciate the hard data! Long distance touring builders take note!

A final query: just how many amp hours can I drain from my battery before I run the risk of damaging it unless I recharge? Or, in general, how do I tell when it's reached that point?

Generally, no more than 80% discharge, ~16ah. Doing so consistently will extend the life of your battery. What was your charge history on the trip? Top it off every day? If so, you're well in the safe range.

 

[wayover13]

Thanks for the tips on acceptable discharge levels and related topics, -dg and arkmundi. And, yes, thanks to Justin too. Yeah, I did cross-country, unassisted touring in the 80's as well, -dg: I'm sort of trying to recover my youth a bit by having another go at it--though with some technological assistance--later in life.

As to the extraordinary result you mention, arkmundi, bear in mind that, as I've mentioned in other postings on this forum, I try to use the motor as little as possible: when I get on a bike, I want to pedal and I routinely work hard enough to break a pretty good sweat. Ideally, I would only use the motor to help get up hills in such a way that I'm not completely spent at the crest--the motor has been a huge improvement in ride quality in that respect. There was at least one day on our trip, when we had a favorable tailwind, that I managed to use the motor pretty much like that. On most of the other days, though, the wind was either coming right at us, or was an oblique headwind. When I'd see our unassisted speed on flat stretches drop below 12 mph on those days, I'd bring the motor into play. So, though I ended up riding hard on those days too, I found myself using the motor a lot more than I'd have liked to.

And yeah, I did top off the battery each day at the end of our ride. I initially wondered whether I'd need to charge during the day--say, over lunch. But I never used more than 7 amp hours on any given day and, though I was unaware of what safe discharge levels would be, I decided it was unnecessary to recharge with only 3 or 4 amp hours drained from the battery.

Anyway, it's great to know I could probably do a 100-mile-plus day without the need to recharge during the trip. I'm definitely up for that, though the wife favors more like 60-70 miles maximum for the day.