Custom order of Turnigy 80-100 feasible?

[joec]

Hi All, I've been lurking for a while. I see that the Turnigy 80-100's are still backordered at Hobby King, but are available elsewhere. Does anybody here know how one might go about requesting a custom order of a quantity of these? I have an application that could really use a re-terminated version of the motor that's star instead of delta, and with a sensor bracket pre-installed. I could probably put in an order for 20 or 30 of these by myself, and would be able to pay a higher rate for the first two prototypes. This is a wheelchair thing, in case anybody's curious, and not backed by any company or investors (yet).

 

[joec]

Alternatively, if anybody has a good idea for proportional forward/reverse control that can give ~200 pounds of thrust (per motor) and a top speed of ~7mph and a resistance below about 0.06 ohms, I'm all ears! The wheelchair parts can get the job done for around $1700 for a drivetrain, batteries not included.

 

[E-racer]

The user RemoteContact has ordered a custom batch before. He had his anodized black. He wanted $200 a motor or $2k for the 10 units he had left. Either he's greedy or they dont cut very good deals for small quantities.

 

[joec]

$200 might still be doable if they have sensors and the right winding. Even if it took me $150 to get a wheel and sprockets done properly, breaking even on the motor would be OK if it let me use the cheaper E-bike controllers instead of the wheelchair or Roboteq controllers.

 

[acuteaero]

You're not going to get proportional forward/back control with any ebike controllers, regardless of what motor you use. The only brushless 3 phase controller I am aware of that I believe (have not tested it myself) might do proportional BLDC control is the Roboteq BLDC controller. The low cost and high power density of these brushless outrunners is very attractive but setting them up for mechanical and electrical reliability in traction applications is not straightforward. I also bet you could get the chinese factory to wind/terminate differently, but I doubt they'd be willing to develop and sell you a sensor setup at any reasonable cost or change the shaft/bearing setup of the motor much (which is important if you expect to side-load the shaft with a belt or chain drive). The only brushless wheelchair setup I'm aware of is Invacare GB style- which AFAIK still makes up a pretty small part of their product line. I'm guessing there's a reason that brushless has achieved so little traction in the wheelchair market!

By the way- where are the sources of this motor that you refer to? HK is out, Leaders is out, they've even disappeared off ebay- I only see them on Alibaba Express for $150ea.

Good luck with your project! Please update us as it progresses!

 

[E-racer]

I don't think this is exactly what you want. Kelly controllers give you the option of running forward and reverse on one stick they call it "Stick Shift Throttle Firmware"

http://kellycontroller.com/stick-shift-throttle-firmware-p-627.html

 

[joec]

E-racer, that's precisely what I was going to get from Kelly. Acuteaero- can you explain if there's a reason why this doesn't work for a sensored 80-100 for proportional control?

I was also doubtful that they'd do the sensor setup for me. I guess I'll have to get clever about having that as a reliable add-on without throwing the cost through the roof.

Are the motors really incapable of side loading with a chain? I swear I'd seen people here doing exactly that. I'll go back to my homework reading some build threads

The Invacare GB line isn't very popular due to the expense. I'm sure you've all heard about "the economy" and cuts to government spending. Well, the government insurance that private insurance bases their policies on has been steadily reducing the price they'll pay for a powerchair, functionality be damned. The drive has been for cheaper and cheaper chairs, and so the GB line didn't expand. It's more efficient and durable than the geared brushed motors, but if you've seen the threads where they've been disassembled, they are actually filled with the kinds of things that make a motor expensive. Lots of rare earth magnets, big windings, big bearings, machined parts- it's kind of exquisite, but the price isn't likely to fall.

That sucks that Leaders is sold out! I looked there maybe two weeks ago when I saw them mentioned here, and then again a bit more recently than that and they were in stock. I had assumed that the Turnigy brand 80-100 was back in production but being sold outside of Hobby King for some reason.

Another thought I have is to take a hub motor and fit the absolute smallest tire possible on it to trade torque for speed. The game speed limit is 6.2 mph, and it's all about acceleration instead of raw speed. I haven't found an affordable way to do that with a hub motor yet- not one with the kind of thrust I'm seeking.

 

[E-racer]

I have been running a side loaded (chain drive) 80-100 for 2 years successfully. If your worried about it upgrade the bearings. I wouldn't worry to much though... I've moved on and built another bike with a side loaded shaft as well. Epoxy the sensors into the stator teeth. Its quick easy and robust. Or run burties optical board. I've done both and both work well.

 

[acuteaero]

The motor is not made by HK- everywhere else they're labeled EMP and the group that sells them on Alibaba is usually called Million Power or something- I suspect the prices are cheapest at HK because they buy enough to muscle the factory into selling them cheaper. It's usually called C80100 KV130 or KV180.

I agree that the GB drive is awesome looking and a great idea, and probably works great. My point is that I think it takes that kind of expense to make BLDC drive work well for a wheelchair. Wheelchair motor control is (in my observation) one of the trickiest applications around- the torque and speed has to be perfectly smooth and even between the two motors, the controller has to filter user commands to allow any change in drive state- full forward to full stop, to full reverse- smoothly, predictibly, safely. I don't know if the kelly firmware will intelligently decelerate the motor and reverse direction when the stick is quickly moved from full forward to full reverse. I would bet that even if it does OK it will not be tunable to the extent it needs to be to make it feel completely confidence inspiring on a wheelchair. As well, if the controller has forward/reverse capability but is not designed to smoothly transition between the two it may destroy the controller and/or create big torque spikes on transitioning between states that will be hard on the driveline.

What I was referring to when I said proportional control was proportional torque control- there was a guy around a little while ago who wanted to use a BLDC hub motor for a self balancing scooter, where instantaneous proportional torque control in both the forward and reverse direction is crucial. http://endless-sphere.com/forums/viewtopic.php?f=14&t=32803 Again, I don't know about the exact behavior of the kelly controller, but I know that the infineon ebike controller will not be good for that application.

The 80100 motor has two tiny 12mm ID 21mm OD 5mm thk bearings on the end-bell side and one of the same bearing on the face mount side. I am not a real mechanical engineer and don't know how to run the calculations for the bearing but my inclination would be to not expect much life out of the single bearing if you put a sprocket on the face mount side of the shaft. The side load will be a factor of the rest tension of the chain/belt plus the torque that the motor is generating. On the 80100 powered bike I'm rebuilding now I'm replacing that front bearing with a much larger bearing mounted outboard of the motor case.

I bet a pair of 10-turn or 12-turn 9C hub motors from Methods in small rims (16?) on low voltage, high current would put out as much torque as you could possibly put down through bicycle tires and get you to 6.5mph plenty quick... Again, the controller is the missing link- but at least the high-turn count 9c will be about as easy as possible on any controller (unlike the 80100)

 

[joec]

Good to hear. The goal will be to build a test mule to see if the E-bike controllers even "feel" right for the job; that's to say that they can do smooth undelayed control of the motors, and the chair drives as desired. Right now I'm asking around on here and scouting out the possibilities to see if just maybe I'd be able to do it economically for a dozen chairs. If I can't hit an affordable price with these, it's not going to do anybody much good.

If I have to spend a little more up front or do some fiddly time intensive stuff that can be designed out with a little investment, that's OK if the path is there. If I have to break a few motors to figure out what's truly acceptable, that's also OK.

 

[joec]

Thanks for the detailed reply, acuteaero. That's a great thread- I will have to go through it in detail.

The speed of response and ability to quickly go from forward to reverse will be crucial. Matching the left and right motor speed may be less critical than you think- the human brain can do some cool stuff with hand/eye coordination if the system is consistent, low latency, and gives a wide enough control band that matches something that your hand can do. Some of these guys practice a few hours per day at the sport.

I actually am a mechanical engineer, so I'll have a look at the math!

 

[E-racer]

just a thought. 2 80-100's is like 10x more power than you nee to go 6.5 mph

 

[joec]

just a thought. 2 80-100's is like 10x more power than you nee to go 6.5 mph

That's absolutely the point. This is what I use now: http://www.robotmarketplace.com/products/NPC-T74.html

We feed that 120 amps at 24 volts, and it's about the best I've done so far. When you can't go faster than 6 mph, and everyone can go 6 mph, the guy who gets to 6 first is the guy who gets to the ball first. Traction is a big deal, which is why I want a motor with a low resistance- so it can keep applying maximum torque up to close to the top speed, rather than being voltage limited and reducing acceleration above a couple of miles per hour.

 

[E-racer]

Mwahahaha well if 0-6 mph is the goal 80-100's wont break a sweat. Putting the power to the ground will be the challenge. I have kellys on both my builds. The scooter has a fixed gear (geared for 40 ish mph) it has no trouble moving around slowly <5mph. I'm quite impressed with the low speed control ability that the sensored controllers offer. I would gear your project for 15ish mph. You can limit the motor speed in the controller to 6 mph. This will allow you to turn it up when its not bound by rules :-D

A friend of mine built a Goped with the torkinator motor (much like that robot market place motor) I assure you the 80-100 would eat its lunch in performance and reliability all day long.

 

[joec]

That's what I wanted to hear! I was seriously looking at the Torkinator as a brushed/geared way of doing this, and it looked like an upgrade to what I have, but maybe not an enormous upgrade for the price and trouble. Taking a step down on motor and controller price for this is enormously attractive.

I'd probably gear for close to 7 or 8 mph. That NPC-T74 is what a lot of other people are using, and they're getting some 120 pounds of thrust out of it with their 10" tires. If I can't get at *least* 120 pounds of thrust per side, this is going to be a lousy "upgrade".

 

[E-racer]

That's what I wanted to hear! I was seriously looking at the Torkinator as a brushed/geared way of doing this, and it looked like an upgrade to what I have, but maybe not an enormous upgrade for the price and trouble. Taking a step down on motor and controller price for this is enormously attractive.

I'd probably gear for close to 7 or 8 mph. That NPC-T74 is what a lot of other people are using, and they're getting some 120 pounds of thrust out of it with their 10" tires. If I can't get at *least* 120 pounds of thrust per side, this is going to be a lousy "upgrade".

From our experience the torkinator at 24v is pretty reliable >2kw. When you run it at 48v 100amp it will destroy itself. The 80-100 on the other hand has been running 48v for over 2 years with my 100 amp cont 200 amp burst kelly with zero reliability issues.

The scooter has a 180kv 6t delta wind and it screams! I built an 88kv version for my new bike I just finished. It works really well and definitely delivers the power but I am about to tear it down and rewind 7t delta. This motor just doent sound or feel itself at 88kv. I recommend you keep the motor rpm's high and gear appropriately.

 

[acuteaero]

Beware! This application (wheelchair- sounds like wheelchair basketball or something like it) is very different from an ebike or scooter application- those kW numbers are practically not applicable.

If you run "hot" low turn-count, high kV motors and a high current limit, and then use some sort of governor in the motor controller to limit top speed it will indeed pull crazy amps and deliver tons of torque- but the motor operation will ALWAYS be very inefficient (at full throttle). Spend a little time playing with the ebikes.ca simulator- imagine running a Clyte HS motor at high power, but limited to only ever run under 15kph. It'll accelerate just as well as a slower wind motor- but it will ALWAYS be under 30% efficient (while at full throttle, which it will get a lot of in the proposed usage). In addition, the PWM duty cycle in the motor controller will always be small in order to limit the current as a result the phase currents will be high, generating heat in the controller, wiring and motor. At this point the challenge becomes how to make sure you won't burn up these components- it's a lot harder to overheat a big T74 motor in a few minutes of thrashing than a 80100 with much less thermal mass. It may take the T74 a lot longer to cool again (indicating higher continuous power capability of the 80100)- but for the few minutes you're actually running the waste heat in the bigger motor will take it to a lesser temperature than the smaller 80100- that difference could be the difference between motor life and motor death.

According to a Kelly page that came up on a search their KBL series uses synchronous rectification which is a good sign that it may be able to manage that "proportional torque control" that I think you need for this application. Depending on how adjustable it is you may be able to tune it up and get it working OK. For this application particularly I would get one several sizes bigger than you think you need, given the large average phase current as described above.

(edited slightly for clarity about efficiency, throttle level)

 

[joec]

Thanks for the heads up- I'm definitely in favor of a mechanical gearing only slightly above the desired speed. I think that will keep efficiency as high as possible. I already assume that the motors will be hot enough to warrant screens around them, and think it's likely that fans will be helpful. The wheelchair motors are built in such a way that under normal use the external temperature never gets hot enough to cause injuries- it's a medical device thing.

I think there will just have to be some trial and error to see if the way I intend to use the motors will overheat them or not.

 

[acuteaero]

Traction is a big deal, which is why I want a motor with a low resistance- so it can keep applying maximum torque up to close to the top speed, rather than being voltage limited and reducing acceleration above a couple of miles per hour.

I'm definitely in favor of a mechanical gearing only slightly above the desired speed. I think that will keep efficiency as high as possible.

I think these statements are contrary to each other- I'm not an expert. This pushes the edges of my understanding. But I do know that as the motor speed approaches the no load speed, back emf will approach battery voltage, current will decrease, torque will decrease. I'm not sure if you can have other characteristics (resistance?) of the motor windings change the rate at which torque decreases as speed approaches no-load speed.

Forced air cooling has massive potential with these motors.

You can certainly get a T-74 hot enough to burn you- run it at low efficiency for long enough it'll heat soak and get way hot. The heat just has to travel a long way out of the rotor (armature), through the magnets and into the can or through the shaft into the end-plates.

 

[Green Machine]

What do you guys think of the noise level of the 80-100? How does it compare to the noise of an astro 3210?

 

[Bluefang]

Its noisier. $100 motor vrs $700 motor.

Outrunners cause more air movement so it has extra noise from that, but been a cheap motor 99% of them are out of balance and run shit bearings and weak shafts. Even with all that they are a awesome motor and with afew mods the C80-100 motors are awesome.

 

[AussieJester]

What do you guys think of the noise level of the 80-100? How does it compare to the noise of an astro 3210?

Appologies ERiC totally forgot to answer your question in recent pm regarding noise level of the 80-100...this video i took
while back IMHO gives you a pretty good indication, this is running a CC HV160 ESC ..when i ran sensored with 12fet Infienon it was noticeablaby
quieter but also noticebly slower teehehehe...

[youtube]KxNta7yGY2A[/youtube]

KiM

EDiT: p.s this is a stock Turnigy 80-100 130kv mounted on a Thud 2 speed dog box...

 

[Philistine]

Thanks for that Kim, so it is louder than a dentists drill, but quieter than a 747 coming in to land. Got it.

 

[Miles]

:lol:

 

[John in CR]

I think you need to be geared for 10mph or so, and then have top speed limited to 6 in some other manner. That's the only way to avoid the acceleration from tapering. To make up for the thrust reduction of the steeper gearing, you simply throw more power at it.

I just did a search and found a number of brushless wheelchair motors both geared and direct drive for pretty reasonable prices, so I'd suggest starting there. Then use the same methods us crazy ebikers use to increase performance, since that's all you're really after. ie Increase voltage, increase current, ventilate the motors including some fans in there, and if the controllers get hot ventilate those too.

If it was a flat out race, sure start from scratch, but for playing ball motor control is so critical that reinventing the wheel only makes sense if you already have real experience in the design of those systems.

Whatever route you go, it's an interesting project, so please keep us updated as you progress.

John