Q100 motor – parameter P1 explained – speed alignment

Slightlypedantic · Oct 18, 2023

DuncanDK said:
I changed the parameter P1 value from 223 (original value) to 131. I live in Europe, so therefore I want the motor to stop/taper off at 25 km/h, and I typed that in as max speed.

Having typed in max speed = 25 km/h and wheel size = 20, my motor tapered off at 23 km/h (actual speed), measured of GPS-speed-tracker on my smart phone, while the LCD showed a speed of 27 (wrong speed).

When I increased the P1 to 141, the motor would stop/taper off at exactly 25 km/h (correct speed) and the display would show +28 km/h (wrong speed). To begin with, this difference puzzled me.

It turns out, that the system operates with three speeds.

- Actual speed you ride with
- Max speed registration, when the system should taper off power
- The speed displayed in the S-LCD-Diplay

P1 is used to align the actual speed with the max speed setting of 25 km/h, and not the speed that is displayed in the LCD-display when riding.

Very interesting and useful post. Although the pragmatic approach to setting P1 seems to work well, I have been puzzling over the need to adjust P1 away from the calculated value by trial and error.

So, the system has two ways of calculating speed:

For the display speed
The display uses the wheel speed sensor to count rpm and applies the wheel size to calculate speed. Being independent of P1 settings there is no scope for casual users to inadvertently mess P1 up, and hence the speed displayed, and then struggle to correct it. At the basic user level the wheel size setting provides a reasonably accurate speed display that is independent of everything else. (The level of accuracy depends on the actual tyre circumference.)

For system control/tapering off (P1)
The system counts Hall sensor pulses, i.e. number of magnets passing. The P1 setting (magnets x reduction ratio) allows the system to calculate wheel rpm. Not speed. To calculate speed the system needs to know the wheel size. Presumably it must use the same wheel size setting as the display?

This is where I get confused
The OP reported different speeds for the display and for the system control. However, if the system control uses the same wheel size setting as the display speed, why would the answer be any different? The Hall sensor wheel rpm value should be exactly the same as the wheel speed sensor value, provided that the number of magnets and reduction ratio used to calculate P1 are correct.

Perhaps the speeds are in fact the same but the last bit of tailing off is hard to detect, leading to an apparent difference?

(The variation in tyre circumference with different widths/profiles will cause errors in speed calculation but this would affect the display speed and system control speed equally and does not seem to account for the variation reported by the OP.)

For my Yose rear hub motor (oem version of Aikema 95RX):

I am going to upgrade to a KT controller shortly so the P1 setting is of interest.

I know there are 20 magnets because I have opened the motor up and counted them.

Yose say the reduction ratio is 4.4 and in order to verify this I counted the gear teeth:

Inner gear (motor) - 18T, helically cut
Idler gear assembly (larger diameter) - 40T, helically cut
Idler gear assembly (smaller diameter ) - 18T, straight cut
Motor casing - 66T, straight cut
(To change the gearing, it is only necessary to change the two helically cut gear sizes.)

Reduction motor/idler = 40/18 = 2.22
Reduction idler/casing = 66/18 = 3.67

Overall reduction = 2.22 x 3.67 = 8.15 (> P1 =163) ...but Yose say 4.4 (> P1 = 88). I'll know which is nearer once I have installed and set up the KT controller.

Presumably the idler gear assembly rotates inside the casing, at wheel speed or differently? Could this rotation account for the much lower Yose reduction ratio figure? Are my calculations correct or, if not, does anyone know how to calculate the reduction ratio correctly?

Yose also say the motor is wound for 190-200 rpm but this is hard to verify without first verifying the reduction ratio. Does anyone know whether a reduction ratio of 4.4 is plausible given the Yose claimed torque of 58Nm, or would a low reduction ratio defeat the object of a relatively high torque motor?

I know I can optimise P1 using the trial and error method often recommended but find it hard to understand why this approach should be necessary instead of just setting the "correct" value. If P1 needs significant adjustment, could this be because motor suppliers have provided incorrect reduction gearing data? With differently geared versions of the same motor being available together with bespoke OEM versions, this may possible? Magnet number errors seem less likely but may also be possible.

Thoughts, anyone?

saneagle · Oct 18, 2023

Slightlypedantic said:
Very interesting and useful post. Although the pragmatic approach to setting P1 seems to work well, I have been puzzling over the need to adjust P1 away from the calculated value by trial and error.

So, the system has two ways of calculating speed:

For the display speed
The display uses the wheel speed sensor to count rpm and applies the wheel size to calculate speed. Being independent of P1 settings there is no scope for casual users to inadvertently mess P1 up, and hence the speed displayed, and then struggle to correct it. At the basic user level the wheel size setting provides a reasonably accurate speed display that is independent of everything else. (The level of accuracy depends on the actual tyre circumference.)

For system control/tapering off (P1)
The system counts Hall sensor pulses, i.e. number of magnets passing. The P1 setting (magnets x reduction ratio) allows the system to calculate wheel rpm. Not speed. To calculate speed the system needs to know the wheel size. Presumably it must use the same wheel size setting as the display?

This is where I get confused
The OP reported different speeds for the display and for the system control. However, if the system control uses the same wheel size setting as the display speed, why would the answer be any different? The Hall sensor wheel rpm value should be exactly the same as the wheel speed sensor value, provided that the number of magnets and reduction ratio used to calculate P1 are correct.

Perhaps the speeds are in fact the same but the last bit of tailing off is hard to detect, leading to an apparent difference?

(The variation in tyre circumference with different widths/profiles will cause errors in speed calculation but this would affect the display speed and system control speed equally and does not seem to account for the variation reported by the OP.)

For my Yose rear hub motor (oem version of Aikema 95RX):

I am going to upgrade to a KT controller shortly so the P1 setting is of interest.

I know there are 20 magnets because I have opened the motor up and counted them.

Yose say the reduction ratio is 4.4 and in order to verify this I counted the gear teeth:

Inner gear (motor) - 18T, helically cut
Idler gear assembly (larger diameter) - 40T, helically cut
Idler gear assembly (smaller diameter ) - 18T, straight cut
Motor casing - 66T, straight cut
(To change the gearing, it is only necessary to change the two helically cut gear sizes.)

Reduction motor/idler = 40/18 = 2.22
Reduction idler/casing = 66/18 = 3.67

Overall reduction = 2.22 x 3.67 = 8.15 (> P1 =163) ...but Yose say 4.4 (> P1 = 88). I'll know which is nearer once I have installed and set up the KT controller.

Presumably the idler gear assembly rotates inside the casing, at wheel speed or differently? Could this rotation account for the much lower Yose reduction ratio figure? Are my calculations correct or, if not, does anyone know how to calculate the reduction ratio correctly?

Yose also say the motor is wound for 190-200 rpm but this is hard to verify without first verifying the reduction ratio. Does anyone know whether a reduction ratio of 4.4 is plausible given the Yose claimed torque of 58Nm or would this defeat the object of a relatively high torque motor?

I know I can optimise P1 using the trial and error method often recommended but find it hard to understand why this approach should be necessary instead of just setting the "correct" value. If P1 needs significant adjustment, could this be because motor suppliers have provided incorrect reduction gearing data? With differently geared versions of the same motor being available together with bespoke OEM versions, this may possible? Magnet number errors seem less likely but may also be possible.

Thoughts, anyone?

The controller has to make some assumption to calculate the bike's speed from the motor RPM. If you want an exact cut off speed, say 15.5 mph, you need to give it exact parameters to match your bike. It's quick to do it by trial and error, so I don't see any problem.

Slightlypedantic · Oct 18, 2023

PS Apologies for resurrecting an old thread, which I tripped over this morning.

Thanks saneagle. I would like to verify my reduction ratio so I can get somewhere near the right P1 value and then adjust in small increments rather than have to hunt over a wide range. But from what you say that may not make a lot of difference anyway. Am I being slightly pedantic?!

I'll try Yose's and my values first, hopefully one of them is near he mark.

jimriley · Oct 18, 2023

Just to add a point, there's an error in the lcd3 display manual, it says the range of P1 is 1 - 255 . It isn't, it's 20-255. With a direct drive motor p1 should be equal to the number of magnets, 18 in a crystalyte 408. However, as I found trying to set up my DD motor, it will not save a setting below 20.
I got around the issue by fitting a wheel speed sensor and magnet, p2 then set to 1. It works fine.

WheezyRider · Oct 30, 2023

Slightlypedantic said:
Very interesting and useful post. Although the pragmatic approach to setting P1 seems to work well, I have been puzzling over the need to adjust P1 away from the calculated value by trial and error.

So, the system has two ways of calculating speed:

For the display speed
The display uses the wheel speed sensor to count rpm and applies the wheel size to calculate speed. Being independent of P1 settings there is no scope for casual users to inadvertently mess P1 up, and hence the speed displayed, and then struggle to correct it. At the basic user level the wheel size setting provides a reasonably accurate speed display that is independent of everything else. (The level of accuracy depends on the actual tyre circumference.)

For system control/tapering off (P1)
The system counts Hall sensor pulses, i.e. number of magnets passing. The P1 setting (magnets x reduction ratio) allows the system to calculate wheel rpm. Not speed. To calculate speed the system needs to know the wheel size. Presumably it must use the same wheel size setting as the display?

This is where I get confused
The OP reported different speeds for the display and for the system control. However, if the system control uses the same wheel size setting as the display speed, why would the answer be any different? The Hall sensor wheel rpm value should be exactly the same as the wheel speed sensor value, provided that the number of magnets and reduction ratio used to calculate P1 are correct.

Perhaps the speeds are in fact the same but the last bit of tailing off is hard to detect, leading to an apparent difference?

(The variation in tyre circumference with different widths/profiles will cause errors in speed calculation but this would affect the display speed and system control speed equally and does not seem to account for the variation reported by the OP.)

For my Yose rear hub motor (oem version of Aikema 95RX):

I am going to upgrade to a KT controller shortly so the P1 setting is of interest.

I know there are 20 magnets because I have opened the motor up and counted them.

Yose say the reduction ratio is 4.4 and in order to verify this I counted the gear teeth:

Inner gear (motor) - 18T, helically cut
Idler gear assembly (larger diameter) - 40T, helically cut
Idler gear assembly (smaller diameter ) - 18T, straight cut
Motor casing - 66T, straight cut
(To change the gearing, it is only necessary to change the two helically cut gear sizes.)

Reduction motor/idler = 40/18 = 2.22
Reduction idler/casing = 66/18 = 3.67

Overall reduction = 2.22 x 3.67 = 8.15 (> P1 =163) ...but Yose say 4.4 (> P1 = 88). I'll know which is nearer once I have installed and set up the KT controller.

Presumably the idler gear assembly rotates inside the casing, at wheel speed or differently? Could this rotation account for the much lower Yose reduction ratio figure? Are my calculations correct or, if not, does anyone know how to calculate the reduction ratio correctly?

Yose also say the motor is wound for 190-200 rpm but this is hard to verify without first verifying the reduction ratio. Does anyone know whether a reduction ratio of 4.4 is plausible given the Yose claimed torque of 58Nm, or would a low reduction ratio defeat the object of a relatively high torque motor?

I know I can optimise P1 using the trial and error method often recommended but find it hard to understand why this approach should be necessary instead of just setting the "correct" value. If P1 needs significant adjustment, could this be because motor suppliers have provided incorrect reduction gearing data? With differently geared versions of the same motor being available together with bespoke OEM versions, this may possible? Magnet number errors seem less likely but may also be possible.

Thoughts, anyone?

Thanks for doing the donkey work of cracking open your motor and counting teeth. I have been trying to do the same with a Yose 250W motor, KT controller and an LCD3 display.

The default P1 value I've found to be 90 and this seems to work ok in terms of power cut out at a given set speed. I'm using a sine wave controller and it is very quiet and smooth, however, the motor seems a bit sluggish, considering the torque it is supposedly able to develop, so I was interested in optimising P1. On another thread, values such as 130, 168 and 211 have been suggested. I have tried these values, the motor seems to have a little more grunt, but it completely ignores the cut out speed setting and is a bit noisier. I'll give 88 and 163 a go and see how I get on.

Nealh · Oct 30, 2023

It has been said that the akm/yose hubs internal gearing is different from the usual akm hubs with a higher gearing . Typically dependant on model and rpm the akm's use to be around 8:1 - 13:1 for reduction ratio where as most simple clones of basic Bafang hub motors use 4:1 ot 4:4:1 so 80 - 88 may be the correct range to look at for the latter ones.

Slightlypedantic · Nov 4, 2023

WheezyRider said:
Thanks for doing the donkey work of cracking open your motor and counting teeth. I have been trying to do the same with a Yose 250W motor, KT controller and an LCD3 display.

Here's how I got on:

I set the max speed to 25kph.

I tried P1=88 and the bike cut out at around 9 mph.

I tried P1=163 and it seemed to cut out just right at about 15 -16 mph, with strong torque at lower speeds. Later, with the wheel off the ground, I found the maximum no-load speed was 15.8 mph. Perfect.

(Note: The max speed setting is merely a reference speed and does not affect controller behaviour. The controller calculates actual speed based on P1 and wheel size, and compares the result with the reference max speed. If actual tyre circumference happens to vary significantly from the display/controller's assumed value, then a small adjustment to P1 would compensate for this.)

My calculated gear reduction ratio 8.15:1 must be correct. Yose quote 58Nm torque against Aikema's 50 Nm. This seems to confirm that the Yose version of the AKM 250W 95RX is made to their own spec - presumably lower speed and thus more suitable for use in UK/Europe.

WheezyRider · Nov 4, 2023

Slightlypedantic said:
Here's how I got on:

I set the max speed to 25kph.

I tried P1=88 and the bike cut out at around 9 mph.

I tried P1=163 and it seemed to cut out just right at about 15 -16 mph, with strong torque at lower speeds. Later, with the wheel off the ground, I found the maximum no-load speed was 15.8 mph. Perfect.

(Note: The max speed setting is merely a reference speed and does not affect controller behaviour. The controller calculates actual speed based on P1 and wheel size, and compares the result with the reference max speed. If actual tyre circumference happens to vary significantly from the display/controller's assumed value, then a small adjustment to P1 would compensate for this.)

My calculated gear reduction ratio 8.15:1 must be correct. Yose quote 58Nm torque against Aikema's 50 Nm. This seems to confirm that the Yose version of the AKM 250W 95RX is made to their own spec - presumably lower speed and thus more suitable for use in UK/Europe.

Nice! I'll give 163 a try.

I don't know how they can claim 58 Nm torque, sounds way too high.

saneagle · Nov 4, 2023

Slightlypedantic said:
Here's how I got on:

I set the max speed to 25kph.

I tried P1=88 and the bike cut out at around 9 mph.

I tried P1=163 and it seemed to cut out just right at about 15 -16 mph, with strong torque at lower speeds. Later, with the wheel off the ground, I found the maximum no-load speed was 15.8 mph. Perfect.

(Note: The max speed setting is merely a reference speed and does not affect controller behaviour. The controller calculates actual speed based on P1 and wheel size, and compares the result with the reference max speed. If actual tyre circumference happens to vary significantly from the display/controller's assumed value, then a small adjustment to P1 would compensate for this.)

My calculated gear reduction ratio 8.15:1 must be correct. Yose quote 58Nm torque against Aikema's 50 Nm. This seems to confirm that the Yose version of the AKM 250W 95RX is made to their own spec - presumably lower speed and thus more suitable for use in UK/Europe.

The reference speed is the maximum speed you set. The controller counts the pulses from the magnets. It uses P1 to translate that into RPM. It then uses the wheel size to get km/h, and it compares that with your set max speed for the cut-off. The speed sensor is only used for speed display on the LCD.

Slightlypedantic · Nov 4, 2023

WheezyRider said:
I don't know how they can claim 58 Nm torque, sounds way too high.

The torque is fairly strong. Definitely not a weedy motor. The bike maintains 25kph in level 2 in neutral conditions (flat, no wind). I've yet to try it up really steep hills but have noticed the motor feels happier if you keep the speed up by increasing the power level. I've only needed level 3 briefly so far. On the flat, in strong headwinds (15-20mph), and with a bit of pedal input I can maintain 13-14mph in level 2.

It's not possible to say whether the "maximum" torque is or isn't 58Nm. The actual torque would depend on the current supplied (my controller limit = 17A) and road speed. Presumably manufacturer's figures are obtained in controlled laboratory conditions.

Aikema claim 50Nm for the 95RX on their web site, but it's entirely possible that Yose have had the motor manufactured to a lower speed, and hence higher torque (58Nm), for UK/Europe's 25kph limit. Yose seemed to imply this (in Chinese English...) when I asked the question. The max no-load speed I measured at 18.9mph tends to confirm this, leaving a bit of torque headroom around 15mph so it's got some response if the speed drops on a hill. It's certainly very rideable.

saneagle said:
The controller counts the pulses from the magnets. It uses P1 to translate that into RPM. It then uses the wheel size to get km/h, and it compares that with your set max speed for the cut-off.

For my Yose 250W rear hub motor (OEM version of Aikema 95RX) Yose quoted a reduction ratio of 4.4:1, which suggested P1=88. This figure was clearly incorrect as it resulted in motor cut-off at about 9mph, showing that P1=88 caused the motor rpm and hence actual road speed to be grossly over-estimated.

Just trying to clear up this loose end fwiw. On reading other threads it seems that a reduction ratio 4.4:1 might be more typical of a Bafang motor? In which case, Yose may have given me data for an earlier 250W motor sourced from Bafang?

WheezyRider · Nov 4, 2023

Slightlypedantic said:
The torque is fairly strong. Definitely not a weedy motor. The bike maintains 25kph in level 2 in neutral conditions (flat, no wind). I've yet to try it up really steep hills but have noticed the motor feels happier if you keep the speed up by increasing the power level. I've only needed level 3 briefly so far. On the flat, in strong headwinds (15-20mph), and with a bit of pedal input I can maintain 13-14mph in level 2.

It's not possible to say whether the "maximum" torque is or isn't 58Nm. The actual torque would depend on the current supplied (my controller limit = 17A) and road speed. Presumably manufacturer's figures are obtained in controlled laboratory conditions.

Aikema claim 50Nm for the 95RX on their web site, but it's entirely possible that Yose have had the motor manufactured to a lower speed, and hence higher torque (58Nm), for UK/Europe's 25kph limit. Yose seemed to imply this (in Chinese English...) when I asked the question. The max no-load speed I measured at 18.9mph tends to confirm this, leaving a bit of torque headroom around 15mph so it's got some response if the speed drops on a hill. It's certainly very rideable.

For my Yose 250W rear hub motor (OEM version of Aikema 95RX) Yose quoted a reduction ratio of 4.4:1, which suggested P1=88. This figure was clearly incorrect as it resulted in motor cut-off at about 9mph, showing that P1=88 caused the motor rpm and hence actual road speed to be grossly over-estimated.

Just trying to clear up this loose end fwiw. On reading other threads it seems that a reduction ratio 4.4:1 might be more typical of a Bafang motor? In which case, Yose may have given me data for an earlier 250W motor sourced from Bafang?

Tried out 163 for P1 today. Didn't help, as with other settings I've tried, it just ignored the set max speed and ran as fast as it could go. P1 of 88 sort of worked, it cut out according to the set speed, but it was not as smooth in terms of moderating power around 25 kph to maintain a steady speed. So I'm not sure if there is another setting we have different to explain this, or whether the hardware is not the same.

I need to try a better battery on this bike, there is definitely something wrong with the one it has. A few days after charging, the voltage drops from 42 V to just over 40. This shouldn't happen.

However, in terms of torque, the Yose/AKM motor is ok, but nowhere near as good as a Bafang SWX02 in my experience. I've done hill climbing experiments with both (using a Lishui controller) and the Bafang will get you up a steep hill at a higher speed and use less energy doing so. If I had a choice between a Bafang and an AKM, I think I'd go for the Bafang.

Slightlypedantic · Nov 4, 2023

I've no idea what the problem might be, but I would check the following (with apologies for stating the obvious) just in case:
1. Is the motor cable plug fully inserted, right up to the line?
2. Is the wheel size set correctly?
3. Is C2 set correctly? My motor runs fine on C2=0, all other settings bar one (C2=2?) made it run roughly or very roughly. Remember to long press the power button to save and exit setting mode.
4. Send a video to Yose so they can see/hear the symptoms. They may send a replacement motor, then you could do a substitution test and see if the problem is with the motor or controller.
5. Don't assume your motor is identical to mine. Take the insides out and count the gear teeth and magnets so you can calculate the reduction ratio and know for a fact what P1 should be - number of magnets x reduction ratio. Taking it apart is easy, undo the collar on the non drive side, remove six Torx screws on the drive side, hold the freehub and lift it out. I held the shaft in a carpenter's vice with hardwood cheeks. To see/count the magnets, undo the drive side collar and six screws, and the end plate and freehub come off as one.

One other thought - when I changed P1 to a high value to see what the max no-load speed was, the motor still cut out at 15.8 mph (wheel off the ground). When I spun the cranks really fast the motor hesitated at 15.8mph and then ran up to 18.9mph.

I hope this helps but, if not, you need help from someone who knows a lot more than I do, which isn't hard!

saneagle · Nov 4, 2023

Slightlypedantic said:
The torque is fairly strong. Definitely not a weedy motor. The bike maintains 25kph in level 2 in neutral conditions (flat, no wind). I've yet to try it up really steep hills but have noticed the motor feels happier if you keep the speed up by increasing the power level. I've only needed level 3 briefly so far. On the flat, in strong headwinds (15-20mph), and with a bit of pedal input I can maintain 13-14mph in level 2.

It's not possible to say whether the "maximum" torque is or isn't 58Nm. The actual torque would depend on the current supplied (my controller limit = 17A) and road speed. Presumably manufacturer's figures are obtained in controlled laboratory conditions.

Aikema claim 50Nm for the 95RX on their web site, but it's entirely possible that Yose have had the motor manufactured to a lower speed, and hence higher torque (58Nm), for UK/Europe's 25kph limit. Yose seemed to imply this (in Chinese English...) when I asked the question. The max no-load speed I measured at 18.9mph tends to confirm this, leaving a bit of torque headroom around 15mph so it's got some response if the speed drops on a hill. It's certainly very rideable.

For my Yose 250W rear hub motor (OEM version of Aikema 95RX) Yose quoted a reduction ratio of 4.4:1, which suggested P1=88. This figure was clearly incorrect as it resulted in motor cut-off at about 9mph, showing that P1=88 caused the motor rpm and hence actual road speed to be grossly over-estimated.

Just trying to clear up this loose end fwiw. On reading other threads it seems that a reduction ratio 4.4:1 might be more typical of a Bafang motor? In which case, Yose may have given me data for an earlier 250W motor sourced from Bafang?

You're right about the reduction ratio. It's over 8:1. You had it right when you counted the teeth.

Search

Q100 motor – parameter P1 explained – speed alignment

DuncanDK

Pedelecer

Slightlypedantic

Pedelecer

saneagle

Esteemed Pedelecer

Slightlypedantic

Pedelecer

jimriley

Esteemed Pedelecer

WheezyRider

Esteemed Pedelecer

Nealh

Esteemed Pedelecer

Slightlypedantic

Pedelecer

WheezyRider

Esteemed Pedelecer

saneagle

Esteemed Pedelecer

Slightlypedantic

Pedelecer

WheezyRider

Esteemed Pedelecer

Slightlypedantic

Pedelecer

saneagle

Esteemed Pedelecer

Advertisers

First Timers' Guide

UK Electric Bike Law Guide

Cycle to Work Scheme

How to Choose an Electric Bike