Q100 stealth assist fail <solved, was LCD3 not connected>

Edit:
< SOLVED! Problem was actually just that LCD3 was not connected to S12S controller, when connector is jumpered to get things moving the controller output is limited. S12S controller must be used with LCD3!>


Hi folks,
I've just completed the build of my first bike, and sadly i didn't hit my own specs so need some help.

My target is:
Stealth look at top speed around 45 km/h / 28 mph with me pedalling full blast as assist to motor.
I want it to be a bike, will always be pedalling, and am fairly fit. I can output 30km/h / 19mph / 150w average on my own. My commuter ride is flat and this will be 99% of my riding on the bike.

I have this kit now from BMSbattery.com:
Q100 328 motor 36V
S12s controller, max amps: 23
48V 9Ah LiFePO4 battery, max current output 15A (limited by BMS)

During my first testride my max speed was 37 km/h without pedalling which was expected but what was not expected was that when pedalling i managed only to increase speed to 39 km/h! It seemed that the motor lowered the assist when pedalling so net effect was minimal. I need a longer ride to check more but from the ebike simulator the motor should not be at its kV limit at this speed so what's going on?

What is the good way to get my spec fulfilled? My first idea is overvolting to 60V, maybe this works in my usage even with the Q100 small size and nominal effect.
Will the S12S controller be able to take higher Voltage than the speced 48V?

Your expectations are unreasonable and your system is somewhat mis-matched.
Your first incorrect assumpsion is that somehow, your leg power will sum on top of the motor power.
Doesn't work that way.
If you look at the Ebike ca sim., your motor only top speed exactly matches your realized top speed.
Now if you add the 150 Watts that your legs can generate, you will notice at the top speed, the power curve(red) is falling like a rock, while the load curve(green) is shooting up like a rocket. Now if you add you 150 Watts on the graph, what do you get?
About 37 to 38 Kph. Again, more or less what you have observed.
To see how your motor is mis-matched to the size wheel, look at two things on the graph.
The power curve has peeked long before the no-load(top) speed.
And, the "meat" of the efficiency curve is at speeds far beyond your top speed.
This is what happens when a sm. high-speed motor is in a big wheel.
The motor is current limited, but due to it's small size, you cannot add any more amps do to your batt. BMS(and it won't add much speed anyhow).
To have a useful assist at top speed, you want the power curve peak at the no-load speed, with the curve SLOWLY falling away. That way, you can contribute more with your legs.
Now, as to your hoped for top speed, I can tell you this;
I have run TWO 328 Cutes in 24 inch wheels and it takes 15S Lipo(over 50V) to reach 28 mph.
And, at that speed you really can't add much with your legs. Firstly, you need serious road racing gears(not easy to add on most bikes)and due to wind resistance, leg power does not mean that much. Remember, the power required, goes up expotentialy with speed.
Another mis-match, is using a 23 A controller with a 15 A BMS. I have used a S12S and it is huge!
With an 15 A limit, you could use one of the tiny 6-FET controllers, although personally, I prefer the 17 or 19 A, non-sine wave 9-FET controllers from Elifebike for use with the Q100.
Right now, your motor is so far out of it's efficiency band, that your power consumpsion is higher than it needs to be and no doubt putting stress on your batter BMS. I assume that you are in Sweden(profile please)and you aren't climbing steep hills, so I guess the system can live.
As what you can do now, well here are some options.
1(Add another complete identical system(including battery) for two wheel drive and you should be able to hit 40 Kph(47.5 Kph with (2) 48 V batts). The advantage to doing this aside from the higher top speed, is that the bike could climb all but the steepest hills and you will have added range. Downside, money, batt. space and more complicated.
2)Change for a 48V battery. 40 Kph, a little more efficient(but not great), but care with hills still needed.
3)If your Q100 is not a CST model, you could buy a 260 RPM Cute(motor only) and swap it in your wheel/motor housing(it is very easy to do). 30 Kph with the best efficiency.
4)260 RPM cute with 48V battery. 36.5 Kph and fair efficiency. Best option I.M.O.

Hi,
I found your double Q100 build thread yesterday, very interesting thread (although a bit too late for me as i already have my kit I must say this forum is great!
I agree with your points below, although i think i might have confused you in my data, the motor is 328rpm@36V but i am running it at 48V.

In the simulator i get this data:

It shows some power left until 45 km/h and with my added pedalling i should be able to reach about 42 Km/h. Charged voltage was 52 when i did the trial run so speed could have been a bit higher

Gearing the bike for this speed is not hard to do, just add a road crankset with 53 teeth on front, 11t rear. (Pedalling cadence of 80 gives you 47,5 km/h at this gearing if you can output the power)
I know that most folks here will wonder why get small motor and run it at it's limit, could be done easier and with more solid motor. For me it's about the speed AND fitness, i don't want the option to just turn the throttle

I am still wondering if there could there be some other cause to the behaviour, could you see any?

Could you check your motor's maximum speed with the wheel off the ground? It's possible that your motor is not a 328 rpm one. Has it got 328 rpm on the label?

Hello,
No 328 label, i'll check the speed off ground and get back in a couple of hours (going for a swim with the kids )

I posted this back in Sept.;'

"Then I ran the S12S without a display, the cruise control worked, but The controller defaulted to a lower range of the speed assist, maybe #2.
I did not use it long enough to know if any other features worked."

It's bmsb, you will have the wrong motor.

15 amp battery and 23 amp controller won't be a good combination though. I have to wonder what cells are in there. You could get a very short lifespan from them.

I used a 270rpm 36v motor (500w cst) at 48v with a 15 amp controller and it did 24mph sat upright, 27mph tucked or 28mph assisting. At that speed the wind resistance is so great when pedaling that tucking in and using motor power alone was almost as fast. The cst/bpm is double the weight of your motor at 4kg but with your 23 amp would be a nice choice. Though the 15 amp battery concerns me.

So i checked the no load speed, its either 38 or 48 km/h. If i accelerate slowly then i can get the higher speed, corresponds to 382 rpm.
Strange, could the BMS start limiting current due to fast acceleration and cause this behaviour? But then why does it not progress until the top speed when the wheel is up to speed and current has decreased?

Anyway, comparable no load speed at 36V is 260 RPM since my battery is at 53V (382*36/53). So maybe i did get the wrong motor. BMSbattery has a "Q100H" with 260 RPM so probably I got this one.
I guess I cannot be sure until the LCD3 display is connected as Motomechs post indicates.

The 260RPM simulation for comparison:


260 motor would explain the behaviour i see.. and it would be impossible to reach my goal with this setup.

Thanks everyone for your inputs, any further thoughts you have are more than welcome!

larsb said:

So i checked the no load speed, its either 38 or 48 km/h. If i accelerate slowly then i can get the higher speed, corresponds to 382 rpm.
Strange, could the BMS start limiting current due to fast acceleration and cause this behaviour? But then why does it not progress until the top speed when the wheel is up to speed and current has decreased?

Anyway, comparable no load speed at 36V is 260 RPM since my battery is at 53V (382*36/53). So maybe i did get the wrong motor. BMSbattery has a "Q100H" with 260 RPM so probably I got this one.
I guess I cannot be sure until the LCD3 display is connected as Motomechs post indicates.

The 260RPM simulation for comparison:


260 motor would explain the behaviour i see.. and it would be impossible to reach my goal with this setup.

Thanks everyone for your inputs, any further thoughts you have are more than welcome!

Click to expand...

That sounds like the 328 to me.
It could be, that by ramping up the motor slowly without a load, the speed limiting feature is not activating.
I am recalling my experience with the S12S/328 combo.
At the time I tried it, I was not using PAS and I already had a Cycle Analyst, so I didn't want to use the display.
Once I realized that it was in intermediate speed mode, I re-connected the display and reset the speed to 5.
Then, when I jumped it again, it automaticaly defaulted to a lower speed again.

As a side note, with the display connected and in speed range 5, my bike would get to 23 mph(12S, no pedal). With the right gears and a fit rider, I suspect an addional 2 to 3 Mph could be gained.
The combo accelerated strongly(for a Cute) and was fun. The controller was very smooth and quiet.
By constrast, using a 25A Infineon made the Cute rough and noisy.

In the end, I thought the S12S was a little to much for the Q100.
That, and the fact I didn't have the room for the S12S, led me to go with the 9-FET controller.
I really think the S12S controller work the best when combined with PAS, where the current limiting feature can be used.

motomech said:

That sounds like the 328 to me.
..
I really think the S12S controller work the best when combined with PAS, where the current limiting feature can be used.

Click to expand...

Wouldn't the 328 have a no load of 328*53/36=483RPM at 53V? If so, something else must be off to explain my motor speed.
When you say PAS, do you mean a true PAS (torque sensor) or would it be good even with the simple rotation sensing PAS?

I guess a way of being 100% certain is opening the hub and checking the reduction in the gears, i've googled the q100 to 8.2 and q100h to 12.6

Not sure if i have the energy tonight though..
But who orders stuff from China without wanting to be surprised

It's starting to make a bit of sense to me now. It's nothing to do with ordering stuff from China.

I didn't realise that you were running without the display, which is not a good idea. How are you limiting current to 15A? I suspect that's got a lot to do with it too.

The S12S has current control, so you can temporarily remove your external current control and test it with the PAS. I don't think that the S12S current control applies to the throttle, which will give full current. Using the current control with the PAS should enable you to see how the motor runs with different currents by using the PAS levels on the display. If you only select the higher levels at high speed, you might be able to run with higher current than if you used fixed current and throttle control. The PAS control with these controllers is really good. Much better than using a throttle.

It's unusual that's for sure. It's unlikely to be purely the motors fault, if at all. The bike don't sound like it's built yet, so I'm going to back pedal rapidly. I don't wish to comment on a half built bike. The controller and display are one. Yes, you can fudge the wires to make the controller run, but nobody is selling a plug to do that for you, it's just an emergency procedure. Finish your build then try it.

@d8veh: current is limited by the battery internal BMS. One comment i have is that it might not be current control that limits the speed as i assume that in no-load the current remains low.
I charged the battery to 58V yesterday and will try the speed again, if no-load still remains at 260RPM when scaled to 36V i think i'll open the motor and check gearing.

Due to the limited info on the BMSbattery site i did not order speed sensor and LCD3 to go with the controller so had to jump the pins as the wiring diagram described. No mention there that it is just a halfway workaround, it was only at ES i recently found this info (possibly from d8veh, thanks a lot for all the good data !)

It is so stupid to sell components without informing what parts are needed to get functionality. The order for display and sensor was placed some days ago, now i only have to wait for the stuff to be shipped.

I'll update when the parts have arrived.

By the way: the supplied PAS could not be mounted on my bike.
PAS magnet ring hole is too large so it rotates around the axle and sensor cannot be mounted either (on the non drive side there's no flange on my bottom bracket and on the drive side there's no room for it.)
Are there other PAS rings that are different in diameter and geometry?

Without a pedal sensor, you're wasting many of the good features of the S12S. There's lots of things you can do to fit the PAS, but I don't understand the bit about the hole being too big for your spindle. Do you have a standard square taper BB? There's a new type of PAS now that clips into the BB on the left side, though I haven't tried one yet:

https://bmsbattery.com/search?controller=search&orderby=position&orderway=desc&search_query=pas&submit_search=

I can't believe that your BMS is limiting your current to 15A. I've never know one as low as that. Even in the days of crappy 1C batteries, the BMSs were normally set to 20A.

FYI, the Ku65/63 controllers default to level 1 when you bridge the red and blue wires without a control panel. It's possible that the S12S also does something like that. You need the LCD to see the settings.

I have this battery:
https://bmsbattery.com/ebike-battery/496-48v9ah-lifepo4-shrink-tube-ebike-battery-pack-battery.html?search_query=9ah&results=4

It says "1pcs 15A continuous discharge current BMS". I guess that means it's limiting to 15A? Or could it average amps somehow? That's more advanced than i would expect!

I do have a standard square taper but my axle is really short --> the taper is still slim where the magnet ring sits & there's no room for ring OR sensor. If i get my motor running correctly i'll mount a road crankset, i can solve it then.

Checked the no load speed at 58V, it's 406 rpm.
Turns into 253 rpm when scaled to 36V, pretty close to the previous 259 I measured.
Doesn't prove anything if the level is limited by some percentage through the controller.
Found this thread here at ES which stated the selectable levels in the LCD menu to 50/33/20/10/0
I should have 328*58/36=528 rpm so I have 77% of my rpms.

Doesn't really fit into the levels as the closest is 50%. I will open the motor when I have some time and count the gears.
Tonight I'll try mounting a speed sensor from an old bike computer to see if I can get rid of the 10 min shutoffs.

According to D8veh, all 36V Cutes use the same compound gear sets. The speed difference comes from different windings.

http://endless-sphere.com/forums/viewtopic.php?f=2&t=52838&p=786376&hilit=+cute#p786376

I read the post and I believe d8veh is saying the gear package space is the same although the ratios are different. A single stage gearing is not the same as a dual stage.

Info on bmsbattery for the gears:
A set of Q100 Gears. It does not fit Q100H and Q100CST. It is only suitable for Q100 motor.

@d8veh: can you clarify?

I think I smoked the controller yesterday, after connecting the speed sensor I had no function. Turns out the jumper leads on the LCD connector had gotten loose which might have shorted 36v to Rx or Tx which I guess normally is 5v. No throttle response but moving the PAS gets the wheel moving.

Sooo extremely annoying! Especially with the long lead time for a replacement.
Could it be some other cause? Should I open the casing to check? If tx or Rx indeed was shorted then I guess the Mcu is affected and that would maybe not be repairable.. I have only a rough idea of how the electronics work so any input is really appreciated!

I'm not sure about the gears on the Q100h because I haven't opened one, but they're definitely different between the Q100C and Q100. I posted a thread about the difference between those two somewhere. I never really looked at the difference between a 201 rpm and 328 rpm Q100. If the gears were more than 50% different, it would jump out. I think that they must have the same gears, but different windings. Only the Q100H has a 260 rpm version. They might therefore have different gears.

Your battery is 15A continuous, which means about 30A max, so there's effectively no external limit on your controller, which could mean that you've been giving your motor 23A at 48v, which would normally smoke it, but maybe without your LCD there is a limit of some sort -thankfully!

If the motor gives power with the PAS, I don't think you have damaged your CPU. More likely a problem with the throttle or speed sensor.

I had a longer bottom bracket lying around, could mount the PAS with that. Did a trial run and it turns out that d8veh is right, output when using the PAS is a LOT higher than with throttle when the display is not connected. It was too powerful for my purpose, turning the bike into a scooter basically.

The LCD3 and speed sensor is on the way from bmsbattery so soon I can really test the bike and mileage.

One thing: PAS cable is really short so I need to extend it somehow. Does anyone have a favorite cable and connector type? Or would you solder and shrink tape the joint in the extension?

Cheers!

I just cut the cable and solder in an extension piece.

Don't forget that 23A from your controller is too much for that motor. Don't use it until you get the LCD. The controller should be set to current control by default, so you will be OK using levels 1 to 3 for startup, level 4 when you're up to speed. You can limit the overall current by shaving or crimping the shunt to increase its resistance. If you do that, you need some way to measure the current externally, to see what you get. A cheap wattmeter can do it, or you can use a voltmeter to measure the drop down one of the battery wires.

The s12s I just set up had throttle speed limited to 15 somethings from factory. Lower than the main speed limit which is a separate setting. It can also be set to not work till after pedaling commences. Though I don't remember the factory setting.

You need the LCD before you do anything.

As I pointed out early in this thread, the S12S without the display defaults to the #2 speed setting.

When in "Throttle Mode", full throttle acceleration is the same regardless of the speed setting. It just doesn't feel as powerful in the lower settings because the speed limiter kicks in early.

I used the S12S in throttle mode on a CST for 100's of miles with no damage to the motor(it never even got warm). The controller is so smooth and "soft", that there is no juttering or noise from the motor at start-up. It could be, that long hills on a warm day at #5 speed setting could cause a problem, but I never felt any need to go higher than the #3 setting.

I did have a problem when I attemped to shave the shunt, as I believe some particals of metal got on the PCB and shorted it out.
For anyone who wants to try and do this, I would recommend removing the internals from the housing and cleaning afterward with compressed air or a spray electrical cleaner.
Frankly, I think it's best to leave it alone.

One reminder to those using PAS, an ebrake circuit is mandatory!

The latest s12s has a power trim setting that can reduce the 23a down to about 18. I forget exactly. Below 18 the s06s is probably the right controller. So hopefully nobody needs to thin the shunt anymore.

Just a pity the latest s12s is about as big as a house brick. Though that means you can chop the ends down to size and use the added width to hide connections inside it. If your after a really good finish.