Top speed limited or torque-speed trade off?

E-HP

10 GW
Joined
Nov 1, 2018
Messages
6,180
Location
USA
Hello folks,

First time poster here. I can't say I'm a long time lurker here, but I got bit by the e-bike bug about a year ago and reading about them casually, but got more serious a couple months ago, and really trying to research this forum and others and watching every YouTube video I could find.

Anyway, as far as background goes, I'm pretty handy with DIY stuff, which I prefer for all of my hobbies, and fabricating a solution for problems for me in a fun challenge. I have a technical background and like anything mechanical or electrical.

I'm 90% done with my first e-bike conversion, converting my 25 year old steel frame mountain bike. I gave it the first couple of test rides last weekend and I'm pretty happy with it, and actually surprised how good it does climbing the hills (~10% grade) by my house given it's a rear hub motor. But, I was expecting more speed since I'm topping out at 24 mph on the flats.

I guess being my first e-bike, my real question is if I'm getting the performance that I should expect from the components I've chosen; granted I went pretty cheap, while putting a little more investment in the battery and controller (or so I thought). I did searches on the components and only found one on my motor from a year ago, and it appears that the member was having a similar issue. Here's what I have; the wheel came in a kit, but I chucked the controller and throttle, and added a KT-LCD3 display:

Motor:
26” Rear Hub - MXUS model: XF 40 1000W 1610 0020
48V, 1000W

Controller:
Sine Wave - RisunMotor model: KT36/48SVPRK-SLSO2G
Maximum current: 45±1A
Rated voltage: DC36V/48V
Rated current: 22A
18 MOSFET

Display:
KT-LCD3

Battery:
Li-Ion – Unit Power Pack - 52V, 20Ah, 30A BMS

Total Load Weight (including rider):
Under 250 pounds

Any help or insights would be greatly appreciated.

Thanks ahead of time!
E-HP

EDIT: adding a pic; this spot is at around 700 ft, my house is at 150 ft. I didn't take this street up (17.5%), maybe next e-bike, but are more gradual route up the hill.
 
First, I recommend editing your first post and changing the Subject: line to something relating to the issue you're having, rather than the motor model (which doesn't have much to do with your questions, and will keep some people from even looking at the thread, that might otherwise have come to help). ;)


Easy first test is take the wheel off ground and use max throttle, and see if it will go faster than when loaded. If not, it's speed limited in the controller, and there may or may not be a way around that in the LCD menus. If there isn't, you'd need a different controller without a limit. Or you could check out Casainho & Stancecoke's threads about alternate firmware for the KT series.

If it only goes a little faster, then you may not have enough voltage to go faster with the particular winding pattern of the motor you're using. If so, you'd need either a different motor, or higher voltage battery (which may also require a different controller depending on the voltage limits of yours).


But most likley, you simply don't have enough power to go faster under the conditions. (weight, terrain, wind, etc).

It takes a certain amount of power to go a certain speed even on the flats, and with more headwind or more of an uphill slope, it'll take more power. The faster you want to go the more it will take, and the curve goes up more steeply past around 15-20MPH depending on the aerodynamics of the particular bike/rider. THere's websites out there that have watts-at-speed calculators for various bicycle/etc stuff; I think http://kreuzotter.de is one. (there's a thread around here titled "calculators" or similar that has lists of them).

Do you have a wattmeter? Or an ammeter and voltmeter? Those, placed in series with the battery to controller connection, would let you know what actual power is being drawn from the battery, which would give you an idea if it is simply insufficient power to go beyond the speed you're after, or something else. You'd multiply the current drawn by the efficiency of the system, which is probably around 75% or so, sometimes more, to get the actual power the bike is putting out to the ground.

If it is insufficient power, the problem is most likely the current limit of the controller, which sometimes can be safely modified via changing the shunt value one way or another to "lie" to the controller about how much is actually being pulled. Sometimes the extra power blows up the controller.

Usually the motors can handle up to half-again the power they're rated for, for a while; more if you modify them to shed heat better (lots of threads about that, like Definitive Testing on Heating and Cooling of Hubmotors).

Sometimes the battery can't handle the extra current, and sags in voltage so much it shuts down to protect itself. or gets hot inside and causes problems. Adding another of the same in parallel will often fix that.
 
Thank you sir for the great info. I've edited the subject line as you suggested. Hopefully it's more in line with what I'm trying to determine. I like the torque, so if it's winding related and everything else is working right, then I'd be completely satisfied, and leave speed to my next project. Here's more info:

  • The motor spins at 26 mph with no load, with the battery charged to 57 volts.

    24 mph was achieved on a flat paved bike trail, tires slightly over inflated, and battery at 53 volts (LCD display)

    Cruising at 24 mph on the flat trail , power consumption per the LCD display is around 240 watts

    Hill climbing, I glanced down and saw 1700+ Watts on the LCD but it's probably higher on the steep parts (forgot to check)

    I set the current limit on the LCD to max (10), and it seems to be allowing max current to flow, at least on hills

I don't know how to describe how well it pulls on the hills, without other e-bike experience, but if it's a winding issue, I'm OK with trading torque for speed, but I want to make sure everything else is optimized. Thanks!

PS. It's pretty cools to see the regen braking working going down that big hill, watching the voltage going up on the way down.

PPS. Another clue: I've been watching videos on programming the KT-LCD3 and my recollection is that the maximum speed setting on the videos tops out at something like 72, while my unit tops out at 99. Does anyone see this as well or has the controller or LCD firmware been updated?
 
E-HP said:
The motor spins at 26 mph with no load, with the battery charged to 57 volts.
Then 24MPH is probably about right for a loaded system, on the flats. Usually you'll get about 80%+ of the unloaded speed.

The 53v I'm assuming is it sagging down under load, from the full charge voltage.

Cruising at 24 mph on the flat trail , power consumption per the LCD display is around 240 watts
If you're not pedalling, then that watt display isn't calibrated very well. ;) Or else it's actually slightly downhill, and not really flat. I'd expect closer to 500w or more required for that speed, depending on conditions and your aerodynamics.


Hill climbing, I glanced down and saw 1700+ Watts on the LCD but it's probably higher on the steep parts (forgot to check)

How fast was it going, on what slope? The 10%? You can check about how much power it should take if you go to one of the calculators or simulators; that will then tell you if the display is calibrated better than it sounds like it is.

If your controller is actually outputting the 45A max it's rated for, and the battery is saggng to around 52v or so, then up that hill you'd be drawing close to 2500w.

As long as you're keeping the speed up then that's not that bad for short periods, but if you are going slow up the hill and drawing that kind of power, that motor is likely to overheat. If the motor is listed in the http://ebikes.ca/simulator or there's one similar enough, you can determine how long it would take to overheat using a similar power level from similar battery/controller combo.
 
If you're not pedalling, then that watt display isn't calibrated very well. ;) Or else it's actually slightly downhill, and not really flat. I'd expect closer to 500w or more required for that speed, depending on conditions and your aerodynamics.

I'll need to be more scientific on my next run and record the power and speeds with more discipline. The trail is pretty flat, but the data should confirm what's going on.

How fast was it going, on what slope? The 10%? You can check about how much power it should take if you go to one of the calculators or simulators; that will then tell you if the display is calibrated better than it sounds like it is.

Sorry, I should have mentioned that I was pedaling up the hill, but not too hard, and about 18 mph and 16 mph on the steeper parts. I didn't look at the LCD when I was on the steeper grades, so I'm not sure how much of a draw it had at its peak. I only mentioned it, because the controller is definitely capable to supplying more current than on the flats at max speed.

I'll be back with more data on my next test run this weekend. Too dark after work to ride until I get a good set of lights for night riding with the time change. I have a bunch of little accessory stuff to mount, plus I'm changing out a couple of cables and I'm relocating my shifter and throttle now that I've ridden, which will keep me busy. Also bought a little pack of neodymium magnets convert my 5 magnet PAS disc to 10, just to see if it will have any effect (and setting the controller to 10 magnets).
 
E-HP said:
Also bought a little pack of neodymium magnets convert my 5 magnet PAS disc to 10, just to see if it will have any effect (and setting the controller to 10 magnets).

It might make the PAS quicker to respond to start and stop of pedalling, but since the PAS is not torque sensing, it doesn't really affect how it reacts to your pedalling.


As noted in the previous reply, the on-the-flats speed you're seeing is about right based on the no-load speed you saw.

So most likely, unless you up the voltage of the system (probably requiring a different controller, or modifying the one you have with new higher voltage components, but possibly not), the speed you have is the best it will give.
 
amberwolf said:
So most likely, unless you up the voltage of the system (probably requiring a different controller, or modifying the one you have with new higher voltage components, but possibly not), the speed you have is the best it will give.

I guess it's a good thing I got the 52V battery instead of the 48V then :) . So I see 750W rear hubs doing 25 mph on 48V on some videos. Does that mean my "1000W" motor is likely wound for torque, since I'm supplying more current and voltage?

I see hub motors on some sites that have RPM ratings, so I'm wondering if that info is of value when selecting a motor instead of luck of the draw, or maybe just sticking with the ones modeled in those online calculators?
 
E-HP said:
I guess it's a good thing I got the 52V battery instead of the 48V then :) . So I see 750W rear hubs doing 25 mph on 48V on some videos. Does that mean my "1000W" motor is likely wound for torque, since I'm supplying more current and voltage?

It's likely that's just the speed that motor goes at that voltage.

If it were a speed-limiter thing, the wheel should spin the same loaded or not.

One more test you could do (that would be a PITA) is to try the other controller that came with it and see what you get for speed. I dont' think it's necessary, but unless they both happen to be speed limited the same, you'd get different results if one was and one wasn't.

BTW, unless you're using a separate speedometer of some type to verify, it's possible the speedo on the LCD isn't accurate (but most likely it's close enough).


I see hub motors on some sites that have RPM ratings, so I'm wondering if that info is of value when selecting a motor instead of luck of the draw, or maybe just sticking with the ones modeled in those online calculators?

If they have an RPM rating, then you can use that to compute the wheel speed with a particular wheel and tire size (larger diameter = faster speed, smaller = higher torque). The rating will be for a particular voltage, so if it doesn't give a voltage as well, then the RPM rating is useless.

Then, as long as it's supplied with enough voltage and current to reach and keep that speed based on other conditions, it'll do that.



The http://ebikes.ca/simulator has a bunch of motors that were tested to create it, so if you want to be "certain" of a particular outcome, you can find the one there that does what you want, then get one of them, and supply it with the same voltage and current that you used in the simulator to make it do that.


In either case, there's still no guarantee that the seller will actually send the version of the motor you ordered (numerous examples of that problem here on ES), but it's more certain than trying various unpec'd motors "randomly". ;)
 
amberwolf said:
It's likely that's just the speed that motor goes at that voltage.

If it were a speed-limiter thing, the wheel should spin the same loaded or not.

Thanks, makes sense. Actually also helpful to revive some old memory cells. I took a motors/generators class at school about 40 years ago. It was fun, because we got to mess around with the school's electric car, that had won a bunch of national competitions. All of us newbies were ooing and awing when the professor described regenerative breaking, and we all thought it was genius at the time. Anyway, we touched on things like the counter EMF stuff, so now I see why my motor is hitting a wall at 25.

I just ordered a digital battery monitor to get more accurate consumption data, which should get here on Friday. That should confirm more on the supply end. You mentioned a watt meter earlier. Are you talking about the clamp on ones that measure AC for the phase current or a DC unit? I have one of those somewhere, if I can find it. Here's the unit I ordered, that uses a shunt rather than being inline, so I can mount the display up front; Plus I like that it retains the watt hour data even if you lose battery power (the others I looked at reset when the power is disconnected):

https://www.amazon.com/gp/product/B0752FDN99/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1
 
E-HP said:
You mentioned a watt meter earlier. Are you talking about the clamp on ones that measure AC for the phase current or a DC unit?
Basically what you linked to; things like the WattsUp, Cycle Analyst, Turnigy Watt Meter, or many others. (I've got multiples of all three of those; but the Cycle Analyst is most useful, at least three of those still work; the TWMs I think are both dead now; I think only one of the WU still works).

The WU and TWM and similar units use internal shunts, so if you want them on the bars you have to run your battery wires up there and then back to the controller. :/ There are some newer ones (like the panelmount one you link to) that have external shunts, I think even a BT unit was posted about recently. Most of them (including WU/TWM) dont' save data when power is removed, unless you install your own external backup battery.

The CA uses either an external shunt that comes with it (SA model), or the shunt inside your controller (DP model; requires compatible plug on the controller that you can add if not yet present). I use the DP versions. The CA also can monitor speed, and so can give wh/mile (efficiency) etc. It always saves the data on powerdown. But its' a lot more expensive, partly because it has many more functions than just a power meter, if you want to use them (I don't, usually).



Here's the unit I ordered, that uses a shunt rather than being inline, so I can mount the display up front; Plus I like that it retains the watt hour data even if you lose battery power (the others I looked at reset when the power is disconnected):
Just keep in mind it's on the large side (about as big as the large-screen Cycle Analyst, maybe a little bigger). "1.96 x 3.54 x 0.95 in" according to the page.
 
amberwolf said:
Just keep in mind it's on the large side (about as big as the large-screen Cycle Analyst, maybe a little bigger). "1.96 x 3.54 x 0.95 in" according to the page.

Measurements are accurate, but doesn't seem so big, but the shunt is huge.
 

Attachments

  • cockpit.jpg
    cockpit.jpg
    170.3 KB · Views: 1,659
  • pas 10.jpg
    pas 10.jpg
    72 KB · Views: 1,659
Well the wattmeter was a great investment (thanks for the suggestion), and provides more accurate data to look at. The air quality here has been hazardous by my house, so the weekend was shot for doing any riding/testing. I spent the time in my garage, re-doing some of the cable routing, trading out my throttle, and generally getting the "cockpit" organized (mounting the new meter and phone holder). I also added 5 magnets to my PAS sensor ring,for a total of 10.

The air cleared up enough at ground level to take my bike out for a test run on Monday afternoon, so I did a quick 15 mile loop, first taking some time to calibrate things like the LCD3 speedometer (using external speed sensor), which is now only +/- a couple tenths of a mph compared to my GPS. Wattmeter is spot on between the new meter and what LCD3 shows. With the battery at 57v and instruments calibrated, I did two runs, one in each direction, on the flattest section of trail. Both runs hit the exact same top speed of 23 mph, with me riding in a tucked position. No load speed is exactly 26 mph at the same voltage. Power draw from the battery on both runs was ~475 watts. Punching the throttle to full when barely rolling and output hit about 1600-1700 watts, and from a stop the bike tends to wheelie, so feels like decent torque. My conclusion after looking at the data is that the motor is likely wound for 25 mph @ 48v (no load), which is good enough for my first conversion, and the torque seems to work out great given the hills in my area.

I also did a couple of steeper climbs and feeling the controller and motor afterwards, both were cool to the touch, so it doesn't feel like I'm stressing the system much, but I may add some Statorade just for insurance. Oh, and the wattmeter is great. It definitely keeps the watthours in memory even when powered off, so that's cool if I want to test the battery capacity over several days. The only problem with the meter is that I had to actually read the instructions to figure out how to reset it :roll:
 
E-HP said:
With the battery at 57v and instruments calibrated, I did two runs, one in each direction, on the flattest section of trail. Both runs hit the exact same top speed of 23 mph, with me riding in a tucked position. No load speed is exactly 26 mph at the same voltage. Power draw from the battery on both runs was ~475 watts. Punching the throttle to full when barely rolling and output hit about 1600-1700 watts, and from a stop the bike tends to wheelie, so feels like decent torque. My conclusion after looking at the data is that the motor is likely wound for 25 mph @ 48v (no load), which is good enough for my first conversion, and the torque seems to work out great given the hills in my area.

If the speed is controller limited, you will hear/feel the motor power cut when you reach the speed limit. I reconfigured my system to respect the 28 mph motor assist limit law in AZ and I can hear and feel when I hit 28 mph. You should also see the power draw drop when you hit the limit.

Your system definitely has enough power to go well past 23 mph. I'm running less power and was able to go over 30 mph on a flat section and my no-load speed is (I think) over 40 mph.
 
wturber said:
If the speed is controller limited, you will hear/feel the motor power cut when you reach the speed limit. I reconfigured my system to respect the 28 mph motor assist limit law in AZ and I can hear and feel when I hit 28 mph. You should also see the power draw drop when you hit the limit.

Your system definitely has enough power to go well past 23 mph. I'm running less power and was able to go over 30 mph on a flat section and my no-load speed is (I think) over 40 mph.

Yup. I think doing the math, it's a 350 rpm @ 48V motor. 350 rpm with a wheel diameter of 26" I think comes out to 27 mph, no-load. If I end up adding some Statorade, I think I may see some thick copper windings inside. The torque makes it fun to ride, but now I'm thinking I need a real torque arm instead of the cheesy ebay one I have.
 
Your no load speed tells me you most likely have a slower wound motor. 350 at 48v is not killer slow, but its not the fast motor for sure. Sounds to me like a 7 turn, vs the 30 mph 6 turn. Maybe even an 8 turn, though they are rare.

More speed at 48v is unlikely IMO. But the good news is it will make a bit less heat going up the killer hills.


you will get more speed out of that motor with 72v, but much experimenting by me and others tells us that with that motor size and winding, 72v though faster, is not better than a different faster, and larger motor on 48v. Its just to easily overheated when fed over 2000w.

Enjoy what you got, its not broke, and works fine. You can buy a faster wind motor if you really want more speed, vs less waste heat on the hill. Often the cores swap out easy, even with other brands. So you can just buy a fast wind, bare motor and not have to lace a wheel.


Best way to get the most for both, is bigger motor, faster wind, and smaller wheel. Then you will haul ass at 72v, and climb those hills like a motorcycle, all with a cool motor. Next build go that way, often modifying a frame longer, and to take the tiny rear wheel.

Your current rig WILL perform better with a stronger battery. If you lose 5v under load, you just lost about 3-5 mph top speed. A stout battery that sags only a volt or less will of course result in a higher voltage under load, and more top speed under load, up to the maximum of 27 mph.
 
Back
Top