Stuttering THE CROWN, TC-100

The Crown TC-100 is stuttering from 0-35 km/h every time when I am over about 30 A battery current.
Then works normal from 35 to about 80 km/h at halve or at full throttle (about 60 A battery current)
With little throttle (under 30 A battery current) it is working normal from 0-35 km/h

My system 20s 4p (20Ah hobbyking), without BMS

I have tried 4 different controllers (greentime 15, 18, 24Fet, and infineon 18 fet), sensorless and sensor with right combinations. Voltage sag of my Lipos is about 3V. LVC of my controllers is 65 V so this is not the reason.
All the controllers work normal on my other bike with Cromotor!
I have tried different throttles, external battery pack, checked and changed all wires and connections, measured phase wires of the Crown (all the same resistance), fuse, without CA3, even with a second TC-100!!!...
Always the same problem!

I think the TC-100 is the problem!!? Never had such a problem with my Cromotor!
Any good suggestions would be appreciated! Thanks

I got my TC65 yesterday. Installed and tested it today. I have a Lyen 18 FET controller with the R115 mod to handle 24S. It does not work. I have the same problems as you described

I also tried to run a 20S setup. Still the same problem. My LiPo's where charged to around 4.0V so 24S gave me 96V and 20S gave me 80V. I will try a setup at 72V or lower tomorrow to see if/where the jerking/vibrations will disappear.

Did you find a solution?

-- Daniel

No, I havent found a solution yet.

Maybe the original expensive crystalyte controller works or an expensive sinus wave controller...

I know nothing about these motors but I remember reading somewhere when these things were just coming out this exact problem. They seemed to have come up with it being too high of a voltage. It doesn't make sense to me, but the only time they had proper results without incident were when they used 16s and under. Not sure if this is in fact true or helps in any way since I never bought one over this stipulation...never mind the crazy price tag

hmmm
not sure, I've been using the TC100 for a few months and haven't had that problem except maybe some oscillation that I think comes from the way the CAv3 handles the current rise in reduced power mode
but with no power scaling and running 18s (75V hoc) I have a smooth throttle response
currently switching over to miniE and I guess I'll see how it does with that

Bitcom, what kind of current are you pushing, and what is your total load and wheel size?

Alexander,
What is the stuttering like? Is it similar to the stuttering when using the Greentimes in sensorless mode and you push heavy current during acceleration? Since it's common it sounds like a design issue, either some kind of resonance when higher current is pushed through the stator, or a timing issue. Could higher flux in the stator interfere with the operation of some types of hall sensors? I've only experienced stuttering in 2 conditions. One was with intermittent shorts in the hall wires from heat damage to the insulation, which increased as the motor heated up, but the stuttering only happened during the first half of acceleration, more pronounced at higher current, and never at cruise. A new motor shouldn't have that. The other was trying to use sensorless operation pushing 12kw, which only worked well enough for getting home in the event of a hall failure or hall wire damage.

Another thought I just had, is what would be the effects of using an epoxy that is electrically conductive to glue down the hall leads? Could that cause some small current to get to the hall wires and affect their operation. What about simply a small error in hall placement?....Does it run the same rpm and no load current in forward and reverse?

Sorry about more questions than answers.

bitcom77 said:

hmmm
not sure, I've been using the TC100 for a few months and haven't had that problem except maybe some oscillation that I think comes from the way the CAv3 handles the current rise in reduced power mode
but with no power scaling and running 18s (75V hoc) I have a smooth throttle response
currently switching over to miniE and I guess I'll see how it does with that

Click to expand...

Which controller are you using? As I wrote I've tried 24S and 20S. I'll try 18S and lower as soon as I gets home. Maybe the problem disappears @75V and below.

John in CR said:

Since it's common it sounds like a design issue, either some kind of resonance when higher current is pushed through the stator, or a timing issue.

Click to expand...

It must be a timing issue. I have the problem even at low current. Using a CA setup with current throttle. Have tried really gentle acceleration as well as setting current limit to 10A. Still oscillating/stuttering/jerking behavior.

An other question is the number of motor poles. Crystalyte claims 46. If i set 46 in my CA the speed shown in the CA is half the actual speed I'm doing. Might the problem have to do with actual motor poles an hall sensors?

-- Daniel

John in CR said:

Bitcom, what kind of current are you pushing, and what is your total load and wheel size?

What is the stuttering like? Is it similar to the stuttering when using the Greentimes in sensorless mode and you push heavy current during acceleration? Since it's common it sounds like a design issue, either some kind of resonance when higher current is pushed through the stator, or a timing issue. Could higher flux in the stator interfere with the operation of some types of hall sensors? I've only experienced stuttering in 2 conditions. One was with intermittent shorts in the hall wires from heat damage to the insulation, which increased as the motor heated up, but the stuttering only happened during the first half of acceleration, more pronounced at higher current, and never at cruise. A new motor shouldn't have that. The other was trying to use sensorless operation pushing 12kw, which only worked well enough for getting home in the event of a hall failure or hall wire damage.

Another thought I just had, is what would be the effects of using an epoxy that is electrically conductive to glue down the hall leads? Could that cause some small current to get to the hall wires and affect their operation. What about simply a small error in hall placement?....Does it run the same rpm and no load current in forward and reverse?

Sorry about more questions than answers.

Click to expand...

no problem
my build thread is here:
http://endless-sphere.com/forums/viewtopic.php?f=6&t=58757

I'm running 26" with the Grinfineon 40A (or at least was, going to hopefully have the miniE ready to roll soon, so we'll see)
I have it setup as current throttle and no issues there even at high motor temps

when I use the reduced power setting in the CAv3 to 33% ( so 13.2A ) accelerating hard (or less hard hehe) etc still works without issue
What does happen however is that if I rotate throttle to something less than WOT and hold it there I get a short period of acceleration with some sort of oscillation
this does disappear as soon as I move the throttle or at WOT or at higher power settings

this leads me to believe that the pid loop tune in the CA is not quite right for low current settings
so at the moment I don't blame the motor
but as I said I'll try with the miniE and see what happens, just a bit of wiring to sort out still

dande74 said:

John in CR said:

Since it's common it sounds like a design issue, either some kind of resonance when higher current is pushed through the stator, or a timing issue.

Click to expand...

It must be a timing issue. I have the problem even at low current. Using a CA setup with current throttle. Have tried really gentle acceleration as well as setting current limit to 10A. Still oscillating/stuttering/jerking behavior.

An other question is the number of motor poles. Crystalyte claims 46. If i set 46 in my CA the speed shown in the CA is half the actual speed I'm doing. Might the problem have to do with actual motor poles an hall sensors?

-- Daniel

Click to expand...

Daniel,
You would need to take the CA out of the equation to make sure it isn't causing the behavior you have. BTW, the pole count is the number of magnets/2, so 23 if the magnet count is 46.

John in CR said:

dande74 said:

John in CR said:

Since it's common it sounds like a design issue, either some kind of resonance when higher current is pushed through the stator, or a timing issue.

Click to expand...

It must be a timing issue. I have the problem even at low current. Using a CA setup with current throttle. Have tried really gentle acceleration as well as setting current limit to 10A. Still oscillating/stuttering/jerking behavior.

An other question is the number of motor poles. Crystalyte claims 46. If i set 46 in my CA the speed shown in the CA is half the actual speed I'm doing. Might the problem have to do with actual motor poles an hall sensors?

-- Daniel

Click to expand...

Daniel,
You would need to take the CA out of the equation to make sure it isn't causing the behavior you have. BTW, the pole count is the number of magnets/2, so 23 if the magnet count is 46.

Click to expand...

I'll disconnect my CA and do a test. Back to speed control. Just to make sure the CA isn't the issue. Do you use the default PID values?

Did a test with 15S (CA still connected). Still the same behavior...

Problem solved!! ...or at least behavior acceptable.

First I tried without my CA and the oscillating totally disappeared. But it's not possible to drive without current throttle when having 7kW. So I connected my CA and changed the IntAGain parameter. Default it is 300. I had it changed to 400 due to as fast throttle response as possible with my old motor. A 9C 7x9. To get my new TC65 to run smooth I had to set IntAGain to 50. The problem with such low value is that the throttle response will be a bit to slow. At least for me. At a value of 75 there are almost no oscillating when battery are in the end of the discharge cycle. But when fully charged there still are some oscillating. I'm no running at a value of 64. Which seems to be a OK compromise. I really would like a more rapid throttle but I guess a have to live with what I have now. I wonder if CA v3 has a better PI-regulator? I have CA v2 with firmware 2.23.

-- Daniel

Hello Daniel, hello guys.

I recently bought a TC 100 for modifying and project purposes. The "stuttering" is ways more present with non-sinus-based (cheap) controllers, even in freespinning usage, but I could only test it up to ~50V until now (due to haveing mostly RC controllers available right now).

It is no self-oscillation in my case (at least not from the rotor itself, I'll check it for counterbalancing issues later on, after the modifications are done). It also is less present with a smooth working sinus controller - even without hall sensor usage (sensorless).

As I am modifying it anyways, I will post you some images, which could help solve some questions (they are out of an actual production unit).

It has 46 magnets, which are 23 of each pole (as John stated). Each second one is flipped pole (I verified this to exclude faults in manufacturing process about the stuttering cause).

The magnets are roughly 40x14x3 mm in size and glued at the outer ring of the rotor with very low space between (almost as one ring, but with straight magnets, not curved ones).

Unfortunately my model had also some piece of one magnet missing / holes in them. This can indeed cause a slight imperfect turning, but shouldn't be the exclusive cause for this case.



It has 51 stator fields (divided through three phases results in 17 phase fields/heads (I am not a native english speaking person, sorry)).



Here you can see the position and cut-outs for the temperature and hall sensors:



Don't wonder about the missing axle on my images. This is part of my modifications (I would not suggest anyone to try and remove it as it is pressure- and glue locked in place).

The bearings are two pieces of NSK Japan (or well-made fakes of those) 6006Z, one in each side carrier plate of the outer ring. Measurements are 30x55x13 mm. They run without problems, but are not stainless. The stator is more or less 39 mm wide, with 1 mm added through GFK side plates.

Hope it helps. More questions? Ask me..

Best regards and keep up the good work
Amon

Amon said:

Hello Daniel, hello guys.

Hope it helps. More questions? Ask me..

Best regards and keep up the good work
Amon

Click to expand...

Thanks for all info.

Just to clarify. Are you using a Cycle Analyst? I only notice stuttering when using Cycle Analyst in Current Throttle mode.

Hello Daniel,

no problem!

I didn't use a cycle analyst, as it was a temporary (test) solution (low power eBike controller) and different voltage-limited RC-controllers to test it. Unfortunately I don't possess my final controller yet.

The "stuttering" in my case was not intense enough to let me worry, still I would love to fix the reason(s). Your solution regarding the cycle analyst settings is very helpful, should the "stuttering" with cycle analyst raise, thanks for this finding!

Best regards
Amon

The mighty Crown is a 40mm X ≈ 200mm stator. I find that, and the famous X53xx being only a 30mm stator, to be hilarious.

BTW, what's with the grinding marks on the magnets. Almost all of the protective coating is gone, and that looks like it was rubbing on the stator while in operation. Is part of the lamination edge similarly scraped, but that side not shown in your pics? A stator not perfectly centered in the gap would certainly be detrimental to performance, and scraped lam edges can lead to corrosion and shorts between lams which would increase eddy currents.

If the stator isn't scraped up, that would mean they ground those magnets down at the factory.

Hello John,

yes, with the side plates (those are from GFK as it seems) it is 40 mm wide. The steel sheets itself seems to be more like 39 mm~ (as exact as I can measure it without disassembling it fully), but that doesn't change much at all. I just mentioned the measurements, because it is of my (and probably others, too) interest to know the technical facts of it. The diameter is around 205 mm, that's true.

About the scratched magnets: Those are hard coated by chrome in galvanic baths (as it seems), but they got a slight cover by something which could be epoxy or some type of glue (non magnetic). I would guess it is epoxy, which is also used to glue them in place. This coating remains are not resistant/hard, I can rub it off with my fingers, it has more of a powdery consistence.

The stator didn't rub on it while turning (at least not in my place, I checked this very well, but I bought it slightly used). Axle was straight. The only point where it actually touched/scratched the "powdery coating" was, when I removed the stator, which has (for sure, being made from steel laminatings) a very strong magnetic force holding it in place in the outer ring (rotor).

But those were not scratches in the direction of turning, more like "sliding" scratches sideways. I also used a piece of cotton to prevent too deep scratches in this case.

After I parted those two pieces I looked around of them very carefully and saw that the outer edge of the stator indeed has some rust prints (It seems like they are finger prints from the assembly - if one touches those typically phosphorized-passivated (as those in "the Crown" are not painted&baked together and they don't possess any physically parted insulation sheet between) steel pieces, they tend to build rust up from the remaining skin acids, which is left.

This is also one of the reasons why I wanted to exclude eddy currents and weakness in the field through insulating them myself from each other sheet. This will be done after some more modifications by inserting very, very thin (I'll have to see if 0,02 mm is available) cut insulating foil pieces between each steel sheet on every "stator-head" and afterwards draining the whole outer stator (turning it) in a thin layer of electrical non-capacitive insulating and low-viscosity epoxide (and let it drip off sideways to prevent building up drops on the stator / rotor field areas).

The stator itself seemed somewhat scratched as well (on the "heads", where the coils are wound around), but those also don't look like a result from the turning direction of the rotor (thry are more like diagonally and in multiple directions). There was also no magnetic dust or pieces of the broken magnet corner anywhere, which speaks against rubbing.

I guess this is some type of "non-critical asian manufacturing flaws", which won't influence the usage as such, at least until something happens which might ruin the whole thing (like you said, electrically connected steel laminate sheets raise not only losses but heat as well, not to mention the rust). But for me this is acceptable, as it is easily changeable and preventable (with some work).

Best regards
Amon

Edit: Oh, forgot one of your questions, I'm sorry. No, the stator looks the whole way around the same, no excessive rubbing seems to have occured. The scratches are all multi-directional (always only small pieces of the stator heads lamination edge towards the magnets), like the stack was rubbed sideways and diagonally over a hard surface before assembling.

If you got it used then I guess all bets are off as to how the magnets got like that, but it does seem all the way down to the magnet itself. Even though it looks like from the stator rubbing, it obviously wasn't with your stator. FWIW I've tried to come up with a way to protect both the magnet coating and the stator edge when pulling it apart and putting it back in. Some of the newer motors that have a thin protective coating on the stator that I hate to scratch up. Wax paper worked to some extent during removal, but it would just rip and tear and bunch up during assembly. I just spray some lubricant/rust protection on now.

I laugh about Xlytes, because back in 2008 I was getting quite well built hubbies with a 40x203 stators and 0.5mm lams for $125 including controller, DC/DC controller, moto rim and spokes, but almost no one was interested. Xlyte was quoting me $220 for a 30x203 raw motor at that time. Their prices are far higher now, but only a bit more stator steel for fit a bit more copper, and that benefit to me is offset by spokes that are harder to get serviced. For a motor quoting 100kph in it's name I find it absurd that higher quality thin laminating steel along with a shell designed for much better heat dissipation aren't used to help justify the premium prices.

Hello John,

oh, about the usedness - the motor was not opened prior to my work. Those screws were like drowned in screw lock and not unscrewed before. Almost impossible to get loose. I used extra hardened bits as the first ones didn't move it the slighest bit and I didn't want to heat the whole motor shell to 100-150°C until it gets soft.

Such an aluminium chassis really draws any heat applied to the (stainless) screws. They are M5x12 countersunk head ones with 3mm allen key, while most M5 screws have at least a 4mm allen key, which allows more torque before one of the parts gets round..

The magnets in the Crown don't look like they are coated by any fluid substance, and if they are, it is VERY thinly applied. It looks more like the epoxy used to glue them in place got through the room between them on positioning them in place and was rubbed away before it hardened. The magnets themself are not badly scratched or filed, aside the damages (small hole dents and broken corner).

Might use some PTFE foil around the stator for sliding it in, this is pretty robust and doesn't allow much of abrasive force to damage it as it glides very well, even under pressure.

I don't like the scratching as well, it annoys me if parts are potentially worsened (stator sheets connect due to insulation damage).

$125, wow. That was nice, even for asian standards. I think that the Crown motors are overpriced as well, but I wanted to look into it and modify it nonetheless, so.. had to pay the price. The chassis is pretty massive and it indeed spreads the heat very quickly.

Aside from this you see alot of asian quality on it, unfortunately. The axle flakes the chrome off like no tomorrow. It is also the same "I want to shear off!"-Design seen in many other axles: 10x14 mm flattened sides to lead the countertorque over. I'll let a good friend lathe a self designed titanium axle with some benefits regarding the cable tunnel and axle broadness as well as torque arm mounting. Lets see how this works..

So for the price.. it could be ways better. But compared to some other asian motors it is a step up in quality, at least from what I can compare.

About those spokes: I hope that my wheel builder has no problem in modifying some quality spokes (DT Swiss/Sapim) to the fitting lengths, we'll see. I like the chassis design alot, but it'll get a new coating.

Best regards
Amon