Regarding the KT display potential to read motor temp, agree the displays are able to display it, but I've never seen anything anywhere on where or how that temp is input. There are no specific/dedicated wires for that purpose anywhere on the controller. The white wire coming from the motor is from the motor's speed sensor....
What can a Grin controller + cycle analyst do that KT controllers + KT display can't do?
- Thread starter Comrade
- Start date
ZeroEm
1 MW
Controllers now can be configured by your phone (wireless) so a display is not needed. I find most display's limited, they do the basic functions. CA gives me a lot of options on how i'm using power. Have two controls connected to it. Can change what they do and have it at my finger tips as I ride. On the fly adjustments. Most do not need or want a CA. I like to keep a check on how fast i'm depleting my battery and how much range i'm doing.
One screen that displays my cadence as I need to work on that as it's to low. Have my trike where unless you pedal at 0.5 kph it has no power. Stops people pushing the throttle and trike taking their feet out from under them.
One screen that displays my cadence as I need to work on that as it's to low. Have my trike where unless you pedal at 0.5 kph it has no power. Stops people pushing the throttle and trike taking their feet out from under them.
Yup, not for everyone; but on the other hand, I didn't know I needed one until a got one. My original purpose was to add PAS to my setup, when upgrading my controller and batteries. Now, that's one of the lower value features compared to how I use the CA. Even though the controller is programmable, I leave the settings wide open/unlimited, and do everything with the CA. The only controller features I use are the cruise control, 3 position switch and regen.ZeroEm said:
I use the quick access settings for my battery, to easily switch between 52V and 72V, and all the other settings adjust; and use the quick access mode settings for normal, off road, and unlimited setups. I also didn't expect that the bike wouldn't be rideable at 72V for normal use, but luckily the CA throttle ramping took care of that, so I only uses throttle bypass mode in the unlimited setup.
If I want to move to a torque based PAS, the CA supports that; which is a rare feature. The KT open source firmware supports it, but I don't know many other off the shelf controller that do. But, the features the CA adds to cadence PAS, like ramping up the power based on higher cadence, really work well for me and the hills I ride on.
I went from not knowing if I needed one to accepting that it's a sunk cost or given for any other ebike project going forward. I don't really use the display much while riding; I know when I'm going fast or slow, or using more or less power without looking at a display, but I do use it to monitor my battery level. I like looking at the cell level voltage, and run from 4.1V down to 3.6V, so I like that the CA does the math for me on that.
Comrade said:
You're referring to the signal from the controller to the display. I get that piece. That's easy. What I don't get, and challenge you to find, is regarding how to supply the controller with input/a signal to send to the display.
Put another way-
Lets say you take the time and trouble to install a temp sensor in your motor. Kindly show me where you hook that temp sensor output to a KT controller so it will send the signal to allow the display to show the motor's temp.
Hint: this is not about the display, or the "multiplexed" wires between the controller and the display.....
AHicks said:
The software and descriptions claim to support it, but there's no hardware to support it, so no way to substantiate those claims. If you look at the wiring for the KT controllers, the 6 pin Molex only has 5 wires, so no sixth wire to monitor temps.
With the KT method of doing things (serialized data wires), there would be no extra/dedicated wire necessary.
It is a hardware issue, as there is no place to connect the data wire coming from the temp sensor. That said, would love to be shown that's not the case. I've installed a bunch of these and have never seen any way to hook up a temp sensor....
It is a hardware issue, as there is no place to connect the data wire coming from the temp sensor. That said, would love to be shown that's not the case. I've installed a bunch of these and have never seen any way to hook up a temp sensor....
AHicks said:
I'm referring to the signal from the motor to the controller.
AHicks said:
The speed sensor wire (white) is what sends the analog temperature to the controller. The controller talks to the display via 2 digital data wires.
E-HP said:
My KT controller has 6 wires for the hall sensors. Positive (red) to power all halls, black is ground, then 3 wires (blue, green, yellow) from the 3 hall sensors that read the position of the motor, and 1 wire (white) that reads speed from the speed hall sensor.
It is on the white wire that in theory temperature gets "multiplexed", if the motor has an internal temperature sensor tied in with the speed hall sensor.
AHicks said:
I have a bench signal generator and I should be able to easily simulate this "multiplexed" signal to test it out. I'll return to my workbench in early December, so we'll see what happens.
The display certainly is able to display a motor temperature.
TMP = ambient
MTP = motor (no thermistor)
ZeroEm said:
*Some* controllers, more typically the medium-to-high-end ebike/scooter types.
Most controllers cannot. (based on a look around google for controllers in general)
Perhaps in the future more will be, but the "average" controller doesn't even have a way to use a display for it (and has no programmability). (More of those do now than a few years ago, but it's still limited to some (common) series of them.)
Comrade said:
As others seem to find this information of no interest, to maintain peace, I'll defer further discussion awaiting your testing done at a future date. Very interest in your finding. Have an open mind on the topic. Would love to be proven wrong, and will be very interested in how/what you used to do it. -Al
ZeroEm
1 MW
Did not try to suggest that non programable controllers would have wireless. If it came across that way then apologize for the confusion.
We are talking about high end features that are not provided with most common controllers.
We are talking about high end features that are not provided with most common controllers.
No luck with the signal generator. Altering the amplitude of the signal from the hall sensor did not change the reported motor temperature. 
The hardware is all there, but the firmware in the controller is not interested in multiplexed speed+temp signal. I've been reading more about it, and it seems people have altered the data stream between the controller and the display to inject temperature. So it's the fault of the firmware in the controller alone.
The controller reads about 0.32v low across the the full range (so all functions related to voltage should be off a little), and speedo tire circumference presets are for thinner tires in general, around 1.75" for 26".
The hardware is all there, but the firmware in the controller is not interested in multiplexed speed+temp signal. I've been reading more about it, and it seems people have altered the data stream between the controller and the display to inject temperature. So it's the fault of the firmware in the controller alone.
The controller reads about 0.32v low across the the full range (so all functions related to voltage should be off a little), and speedo tire circumference presets are for thinner tires in general, around 1.75" for 26".
Comrade said:No luck with the signal generator. Altering the amplitude of the signal from the hall sensor did not change the reported motor temperature.
The hardware is all there, but the firmware in the controller is not interested in multiplexed speed+temp signal. I've been reading more about it, and it seems people have altered the data stream between the controller and the display to inject temperature. So it's the fault of the firmware in the controller alone.
The controller reads about 0.32v low across the the full range (so all functions related to voltage should be off a little), and speedo tire circumference presets are for thinner tires in general, around 1.75" for 26".
So to sum that up, it sounds like you are now agreeing with my earlier statement saying there is currently no way to input the motor temp that will result in the display showing it?
AHicks said:
Partially agree. You claimed earlier that there is no hardware to make this work, no wires. But all hardware is present. And half the firmware is present also. The KT display firmware has support for motor temps. The KT protocol has bits for motor temps. It's just the firmware in the controller that does not appear to support it.
If KT made their firmware open source (or decided to implement motor temp, and released the file for flashing) then it would all work.
Comrade said:
Without splitting hairs, it's pretty clear from where I'm sitting that the KT motor temp function is incapable of working. When/if you make it work, I'd be interested in how you did it IF what needs to be done could be duplicated by an average DIY'er. At 70, I have no motivation to learn enough about it to have any useful input in the matter.
mintycrayon
1 mW
The 6 pin Molex INPUT connector on the KT only has 5 wires; no sixth wire to monitor temps. The multiplexing scheme can be done with a simple transistor on the controller so nobody ought to be disputing the feasibility - only that it doesn't matter if there is no data source.Comrade said:
mintycrayon said:
So am I understanding correctly that you are saying if there WAS a data source (temp sending unit) YOU could make it work? That being the case, please share how that might be done.
I'm asking, as I believe there IS a 6th wire. It's white, and generally used for a speed sensor internal to the motor, or external. Without a signal supplied to that wire, you will not have a speed shown on the display.
mintycrayon
1 mW
Without a motor thermocouple supplied to the KT controller there is no possibility of showing motor temperature.AHicks said:
If the KT controller sampled the thermocouple from the motor the KT controller could multiplex as described below. An oscilloscope would show a slightly noisy square wave where the "noise" represents the multiplexed analog thermocouple data. Kinda neat backwards compatible enhancement. If it were me I'd put in a CANBUS and create havok. :wink:AHicks said:
Comrade said:
mintycrayon said:
Have you actually seen what a properly multiplexed signal from a motor that was designed to send multiplexed signals looks like?
From my understanding an NTC thermistor (not a thermocouple) would pull down the rail for the hall sensor, and the top would represent the temperature. I tried simulating that with a signal generator, but the controller didn't care.
So either I'm (1) not generating a properly multiplexed signal, or (2) controller firmware does not support it yet. If someone has a Grin motor with an embedded hall+temp sensor, and can look at the signal on an o-scope, that would help.
The CA (Cycle Analyst) doesn't multiplex any signals. It has a completely separate wire set for the temperature sensor and the speed sensor, and completely separate inputs on it's internal MCU as well.mintycrayon said:
LInk to the v3 wiring, as an example:
https://ebikes.ca/product-info/grin-products/cycle-analyst-3.html#CablesandConnectorPinouts
The phaserunner (and presumably the baserunner) also use separate wires
https://ebikes.ca/product-info/grin-products/phaserunner.html
noted specifically in teh "L10 Particulars" pinout diagram and the "MT Particulars -- Speed/Thermistor Pass Thru - JST-SM" sections .
Other wiring harnesses listed below that section note variances with different adapter cabling, such as one that doesn't even connect the speedo signal to the PR, etc.
