Switch from internal to external controller - hub motor

[rick_p]

By the way, I popped the cover off another one of the TransX wheels that are available and it looked as cooked as the one I converted. There are others available but I only opened one. My guess is, overheating was an inherent problem and they are probably all toasted, which is a shame because most of these wheels have less than 8k miles on them according to the odometers built into the original controllers.

 

[amberwolf]

I suspect that without any smell, and the "brushed on" look it has to the edges, that it is just a coating like you said originally. If it was burned windings, there wouldnt' be some nice new shiny spots *under* the darker ones. .

 

[amberwolf]

@amberwolf as promised, pictures of the Stromer wheel. <snip> Suddenly the idea of converting it to an external controller got a lot more appealing,


View attachment 376612

That connector indicates it is almost certainly not an internal controller model, but instead has 5 hall wires, 3 phase, and two other signal wires (probably temperature, possibly a cadence or torque sensor).

 

[rick_p]

I suspect that without any smell, and the "brushed on" look it has to the edges, that it is just a coating like you said originally. If it was burned windings, there wouldnt' be some nice new shiny spots *under* the darker ones. .

I forgot to check for any burnt smells, I'll reopened it and check for that, and also look for any other indicators that might prove it's not from heat and more likely to be a coating of some kind.

 

[rick_p]

That connector indicates it is almost certainly not an internal controller model, but instead has 5 hall wires, 3 phase, and two other signal wires (probably temperature, possibly a cadence or torque sensor).

I guess the reason I assumed it had an internal controller was a combination of the connector style, which looked proprietary to me, and the number of pins (10) which happens to be the same as the TransX motor I just converted. But when I saw your reply I researched the connector and found that it's not proprietary, and goes by the models/names L1019, M10, and G10.

This is actually exciting news, I would love to be able to use this wheel without having to make a conversion, but I am curious what it was about the connector that tipped you off to determine that it's almost certainly not an internal controller model, is it the orientation of the pins or what?

 

[rick_p]

I went on a longer ride today on the converted wheel and even though it runs pretty well, I noticed a couple of things that don't seem 100%. It's hard to describe, but when using PAS, the transitions to power on and power off are not smooth, it's like there is a short hesitation, with maybe even a little thud on either transition. Another thing I noticed is, there's a vibration that happens when I'm using throttle mode rather than PAS, and you can really feel it when you're decelerating slowly with power still on. Would any of this be a sign of the halls not yet being configured correctly?

 

[amberwolf]

I am curious what it was about the connector that tipped you off to determine that it's almost certainly not an internal controller model, is it the orientation of the pins or what?

Well, first, it's the same as the connector used on my non-internal-controller Ultramotors off the A2B Metros and the Phaserunner v6 I use on one of them.

Next, there are three fat pins, which is the same as you'd need for phases. You don't need three fat pins if there's an internal controlelr, just two for battery + and -.

I don't remember seeing a pic of the transX connector, but I would guess it only has two fat pins and the rest small.

 

[rick_p]

You don't need three fat pins if there's an internal controlelr, just two for battery + and -.
I don't remember seeing a pic of the transX connector, but I would guess it only has two fat pins and the rest small.

You are correct.

 

[rick_p]

I went on a longer ride today on the converted wheel and even though it runs pretty well, I noticed a couple of things that don't seem 100%. It's hard to describe, but when using PAS, the transitions to power on and power off are not smooth, it's like there is a short hesitation, with maybe even a little thud on either transition. Another thing I noticed is, there's a vibration that happens when I'm using throttle mode rather than PAS, and you can really feel it when you're decelerating slowly with power still on. Would any of this be a sign of the halls not yet being configured correctly?

As a follow up to my own post, I rode another bike today with one of these wheels that still uses the internal controller. I noticed that it also has a slight but noticeable resistance and "thud" feeling when you stop pedaling and the power is cut off to the motor. It doesn't do it every time, but it's there. However, it does not do it when pedaling resumes and power is applied, that is a very smooth transition, and there is never any motor vibration. So, I'm still left wondering if my converted motor is running as expected, meaning those behaviors are normal when used with a different controller that may not have algorithms to deal with those transitions, or if I need to try different hall wire configurations?

 

[rick_p]

Well, first, it's the same as the connector used on my non-internal-controller Ultramotors off the A2B Metros and the Phaserunner v6 I use on one of them.

Do you happen to remember which pins are which colors? It might not be a guarantee for me, but it’s a good place to start.

 

[amberwolf]

Depends on who made the cable, etc. The ones on the A2B motors have various wires of colors that aren't typical, so yours could too.

There's no guarantee that they use the same wiring pattern, etiher, regardless of color, so you need to verify which pin is which signal, power, or ground before trying to operate it. The three fat pins are the phases, but the others could be anything.

(if i were making a pinout for a connector like that, the two separated small pins would be 5v and ground, and the smile of 5 pins would be the 3 hall signals, and whatever other sensors are needed in there. But that's not how Grin Tech uses them, and I don't know if anyone else does either. )

 

[amberwolf]

As a follow up to my own post, I rode another bike today with one of these wheels that still uses the internal controller. I noticed that it also has a slight but noticeable resistance and "thud" feeling when you stop pedaling and the power is cut off to the motor. It doesn't do it every time, but it's there. However, it does not do it when pedaling resumes and power is applied, that is a very smooth transition, and there is never any motor vibration.

Then that's a controller thing.

Each controller is different, between design choices in hardware and software (especially SW), you can have subtle or drastic operational differences.

 

[rick_p]

Depends on who made the cable, etc. The ones on the A2B motors have various wires of colors that aren't typical, so yours could too.

There's no guarantee that they use the same wiring pattern, etiher, regardless of color, so you need to verify which pin is which signal, power, or ground before trying to operate it. The three fat pins are the phases, but the others could be anything.

(if i were making a pinout for a connector like that, the two separated small pins would be 5v and ground, and the smile of 5 pins would be the 3 hall signals, and whatever other sensors are needed in there. But that's not how Grin Tech uses them, and I don't know if anyone else does either. )

I was afraid that might be the case, which is a bit of a setback because I don’t have the giant three-prong puller required to open the hub, which would be my first preference. I would have to either have to remove some of the sheathing or play the guessing game, and would probably opt for the guessing game so as not to damage the cable and wires. I’ll do more searching, maybe someone has been down this road already

 

[rick_p]

Then that's a controller thing.

Each controller is different, between design choices in hardware and software (especially SW), you can have subtle or drastic operational differences.

I'm glad to hear you say that, it's kind of what I was thinking might be the case. I'm going to ride it to work tomorrow, about 8-10 miles one way. On a related note, I rode the bike to work with the wheel that still has the controller inside it today, and when I got home after about an hour ride in mid-80s heat, I put my hand on the hub, which is something I never do, and it was pretty warm, I could keep my hand on it, but barely. I'm not sure if that is normal under conditions, but thought it was worth mentioning in case it's not, because it might be an indication that these motors are prone to heat buildup.

 

[amberwolf]

Any motor with a controller in it has "twice as much" heat buildup, because the controller makes waste heat, and the motor makes waste heat, so having them both with the same radiative surface means it can only shed heat "half as fast" as if they were separate.

Actual proportions depend on too many things to list, but if you have them separate, you have much more ability to rid the system of heat.

If the system is *designed* around the limitations of heat radiation the internal controller causes, then it's not an issue, since you're unlikely to be hotrodding such a systme--they're usually locked down so you can't do anything with them the designers didn't foresee / intend.

The only catch would be if it's not protected against usages outside the design constraints, and then used that way--like if it wasn't meant for hills, or repeated hard acceleration, etc., and then is used for those things. But at minimum it should have thermal protection to rollback power or shutdown or both.


But, if it works, and the systems aren't reported to have a lot of controller failures, then the heat is unlikely to be an issue other than the faster aging of all the electronics in therre from that (especially capacitors). If there *are* a lot of controller failures (prbably only reported as wheel replacements by most users), then the system doesn't "work" for it's designed usage and would have to be scaled back in power to handle it.


Now, the ones that are *really* problematic are the ones with battery and controller in the motor. (copenhagen wheel, for instance), because the battery is forced to endure the heat created by both of them, as well as whatever it generates itself. Those can be forced to severe power restrictions for this reason.

 

[rick_p]

Anything I report from this point forward is likely to be specific to this particular wheel, meaning anyone converting another brand/version than mine may have different results, which may be better or worse than mine.

I rode the wheel to work today, which is 8 miles one-way, and the overall performance was good. However, the heat buildup was higher, I was able to hold my hand on the hub but it was tough to keep it there, it was pretty hot.

This is significant to report on because a comment it the last post by @amberwolf ...

The only catch would be if it's not protected against usages outside the design constraints, and then used that way--like if it wasn't meant for hills, or repeated hard acceleration, etc., and then is used for those things. But at minimum it should have thermal protection to rollback power or shutdown or both.

To this point, I was probably using it outside of design constraints, but there was no thermal protection to roll back power or shut it down because I did not connect those two white wires to anything, which was mentioned in comments more than once that I probably should.

In addition to the heat build up, battery efficiency seemed slightly reduced compared to the geared hub motor I normally have on this bike. There could be a number of factors here, ranging from previous heat damage to the windings, maybe I was pushing it harder than I normally do, or maybe the lack of thermal protection to roll back power.

The only other thing worth reporting is the noticeable transition from no power (coasting) to power (PAS kicks in). During that transition you can noticeably feel it drag before the power kicks in, and the faster you're going, the more noticeable the transition is. This same controller does not do this with the geared hub that is normally on this bike.

 

[amberwolf]

This is significant to report on because a comment it the last post by @amberwolf ...

To this point, I was probably using it outside of design constraints, but there was no thermal protection to roll back power or shut it down because I did not connect those two white wires to anything, which was mentioned in comments more than once that I probably should.

My comment was in reply to your post about *internal controllers*, and wasn't intended to apply to the external controller setup you're using on it.



EC stuff would also need thermal monitoring and management (separate, each, for motor and controller) if you want to push it harder than it's really meant for (or if you don't know what it's meant for, etc).

 

[amberwolf]

The only other thing worth reporting is the noticeable transition from no power (coasting) to power (PAS kicks in). During that transition you can noticeably feel it drag before the power kicks in, and the faster you're going, the more noticeable the transition is. This same controller does not do this with the geared hub that is normally on this bike.

I presume the geared hub has an internal freewheel / clutch (like almost all of them). This means there wont' be any motor drag (that you will get from any DD hub, or clutchless geared hub).

Some controllers (like Phaserunner, baserunner, etc) have an option to cause a tiny amount of power to always be provided to overcome that drag.

 

[rick_p]

My comment was in reply to your post about *internal controllers*, and wasn't intended to apply to the external controller setup you're using on it.

This was understood, I was merely pointing out that my conversion removed this important role the internal controller provided, and that people making this type of conversion need to be aware this.

EC stuff would also need thermal monitoring and management (separate, each, for motor and controller) if you want to push it harder than it's really meant for (or if you don't know what it's meant for, etc).

Yes, this was my point exactly. I did extend the white wires to the outside, and may pursue monitoring the heat, but I will need to learn more about how that gets done.

 

[rick_p]

I presume the geared hub has an internal freewheel / clutch (like almost all of them). This means there wont' be any motor drag (that you will get from any DD hub, or clutchless geared hub).

Some controllers (like Phaserunner, baserunner, etc) have an option to cause a tiny amount of power to always be provided to overcome that drag.

This is excellent news, I was concerned that this behavior might be abnormal and destructive in some way. Note that I am not all that concerned with this particular motor/wheel, but I want to get it right for the next one, which I might care more about.

 

[amberwolf]

Yes, this was my point exactly. I did extend the white wires to the outside, and may pursue monitoring the heat, but I will need to learn more about how that gets done.

Depends on what you want to happen.

If you don't want anytthing to happen automatically, you just connect them to a temperature display that's conveineitn for you to watch while riding. Then you decice what haepppens and when.

if you want the system to prevent overheating, you connect them to something that can do this. Some vcontrollercs can do so, and some of those have user settable limits, some don't. Some shut off over the limit, some rollback output in various ways. The cycle analyst bv3 cand monitor teperature and decrease thorttle output, etc. You could code an MCU adruino etc to read the temp and output whateer control sitgnal you prevfer to do whateverf yuou want to happen.

 

[furcifer]

I suspect that without any smell, and the "brushed on" look it has to the edges, that it is just a coating like you said originally. If it was burned windings, there wouldnt' be some nice new shiny spots *under* the darker ones. .

Funny, I was under the impression "ozone" was the smell of burning enamel, but then they say you can smell it during lightning or under high voltage wires at times. But there's "an electrical smell" that's for sure.

I've never seen one that dark before. Good to know not everything is always what it appears.

 

[amberwolf]

Funny, I was under the impression "ozone" was the smell of burning enamel, but then they say you can smell it during lightning or under high voltage wires at times. But there's "an electrical smell" that's for sure.

Ozone is just oxygen in a different molecular form than we process for normal breathing. We can't "smell" that oxygen, but we can smell the ozone version because it's different.

YOu smell ozone from arc discharges large and small because the energy is high enough to cause it to form from the regular oxygen in the air that energy passes thru.

The smell of burning enamel is not ozone or the ozone smell; unless you get arcing from melted gaps in the wiring or arc shorts between wires whose insulation has failed, there's probably no ozone present from overheating windings. Enamel burning (or just overheating) releases various carbon compounds, as that's what its' made of (at least some of the fumes are probably toxic, as ozone itself can be).

 

[rick_p]

Depends on what you want to happen.

If you don't want anything to happen automatically, you just connect them to a temperature display that's convenient for you to watch while riding. Then you decide what happens and when.

This (if anything) is the way I would want to go, and I may seek guidance when I'm ready to do that, but for now I'm just going to take it easy on motor by not pushing as hard as I have been during the testing phases, and make stops to check the heat build up by touch because I know what "normal heat" feels like from checking a motor that has built-in protection under the same conditions.

 

[rick_p]

I have talked about a Strommer Syno motor in this thread, but I'm going to move that to a new thread so as not to make this thread confusing for future users, and also because that thread may get long on it's own.