• Howdy! we're looking for donations to finish custom knowledgebase software for this forum. Please see our Funding drive thread

100 Volt controllers

johnnyz383

100 W
Joined
Jul 28, 2013
Messages
212
Location
London
Ok....does anyone know where i can buy a SENSORLESS controller capable of handling 100 volts and 65 amps. I have a Lyen controller that needs repair but I have not heard back from him to confirm a repair service. I have a Crystalyte HS3540. I know Kelly makes a robust controller but only for sensored motors. I wont put a $1000 controller in my trike and Sabvoton also doesnt make a sensorless controller it would seem.

Anybody have any ideas??


Thanks

John
 
If you mean 100V max, either of these should work. I've been running a 72V 15 fet controller for almost 3 years on 24s rc lipo charged to 100.8V without a problem If you want to go much over 100V max, look for a controller with 4115 fets and appropriate caps.
http://www.aliexpress.com/store/group/24-mosfet-type/313864_259098441.html
 
johnnyz383 said:
MadRhino said:
Sensorless and 100v don't do so well together. I suggest hall sensors.

Explain please....


John
at higher power levels it becomes harder for the controller to read the motor position.
At 100v, it's really easy to hit that level.
it can be done, but that's also why it's hard to find controllers.
 
MrDude_1 said:
johnnyz383 said:
MadRhino said:
Sensorless and 100v don't do so well together. I suggest hall sensors.

Explain please....


John
at higher power levels it becomes harder for the controller to read the motor position.
At 100v, it's really easy to hit that level.
it can be done, but that's also why it's hard to find controllers.

ok..that would explain why when i bumped up the power from 72 to 92 volts, the "stuttering " at low speed was considerably worse...problem is that I coated the inside of the motor with tranny fluid as I wanted to try an oil cooled set up but when I couldnt stop the thing from leaking, I drilled holes in the side covers and drained as much of the oil as i could but the windings are covered in oild...not sure how you would wire the thing up for hall sensors now....


John
 
LockH said:
ES thread "High Voltage versus High Current":
http://endless-sphere.com/forums/viewtopic.php?f=30&t=43974


AFAIK, "safe" voltage generally taken as only about 50 volts (in building codes).

Current=torque
Volts=rotational speed

So the idea that increased voltage with a sensorless controller is a bad idea is correct?....but why?...I still havent heard an explanation.I heard that increasing the voltage makes it harder for the controller to initiate and coordinate the phases but WHY?


John
 
johnnyz383 said:
Current=torque
Volts=rotational speed

LOVE torque. Speed? Not so much.
http://www.carbibles.com/EV_bible.html

"Electric motors are all torque"
evtorquegraph.jpg


... up to a point.
 
Sensorless is making a lot of cogging on a start. Ever noticed how they want to start backward sometimes ? Well, they do fine once the motor has started to turn, but the current surge cogging in a bad start will fry the controller at high voltage. So if you are using a sensorless controller at 100v, make sure to start the bike rolling before giving throttle, and preferably use a slow winding motor.

The 4110 mosfets are reliable with very high current at 100v and even a tad above, in a sensored controller. Sensorless, you can short them easily.
 
MadRhino said:
Sensorless is making a lot of cogging on a start. Ever noticed how they want to start backward sometimes ? Well, they do fine once the motor has started to turn, but the current surge cogging in a bad start will fry the controller at high voltage. So if you are using a sensorless controller at 100v, make sure to start the bike rolling before giving throttle, and preferably use a slow winding motor.

The 4110 mosfets are reliable with very high current at 100v and even a tad above, in a sensored controller. Sensorless, you can short them easily.

Well I think im convinced. Im having the motor wired with Hall sensors and going this route. Its a wonder most reliable and very good controllers like kelly dont have sensorless controllers and especially not much more than 72 volts. Live and learn....

Hey Lock....once I get this mess sorted out better watch out if you guys are having the Uphill race in Toronto again this summer...ill be there with the Arctrike, better and badder than ever.. ;)

forendlessarc.jpg
 
Next TO race (same location) next Sun Sep 27 (sorta traditional full moon that night). Ya huge cheater. (Well, we pretty much ALL "cheat" w/power assist. Hehe.) This year, just go ahead and take the darned trophy. Plan to add a base w/prior years winner plaques. Back to thread title, `far as I can figure it's "ALL about the amps". (But know near zero about electronics. Did at least pass "advanced" physics - and chemistry - classes in last year high school.)

(PS: Still in trad. ES Off Topic Mode, 39 is count today for ebikers threatening to descend on Toronto City Hall plaza for "Ebike Day" next Sunday, May 24.)
 
LockH said:
Next TO race (same location) next Sun Sep 27 (sorta traditional full moon that night). Ya huge cheater. (Well, we pretty much ALL "cheat" w/power assist. Hehe.) This year, just go ahead and take the darned trophy. Plan to add a base w/prior years winner plaques. Back to thread title, `far as I can figure it's "ALL about the amps". (But know near zero about electronics. Did at least pass "advanced" physics - and chemistry - classes in last year high school.)

(PS: Still in trad. ES Off Topic Mode, 39 is count today for ebikers threatening to descend on Toronto City Hall plaza for "Ebike Day" next Sunday, May 24.)


What...me cheat??...I have pedals up front and I will be pedalling.... ;)
 
This just confirms my idea of staying at 74V, even at that my motor on the simulator is like 60km/h. I think I will be scared at 50!
 
johnnyz383 said:
problem is that I coated the inside of the motor with tranny fluid as I wanted to try an oil cooled set up but when I couldnt stop the thing from leaking, I drilled holes in the side covers and drained as much of the oil as i could but the windings are covered in oild...not sure how you would wire the thing up for hall sensors now....


John

Not a problem. Just use brake cleaner on the area you're going to epoxy/silicone the hall sensors
 
Hall sensors are really essential for what you've got and what you would like it to do. If yours is one of the earlier HS3540's that didn't come with hall sensors then you might have a *really* fun time getting them in there as I don't think there's provision for them on the stator.

You're going to get a much better outcome just replacing the motor. The MXUS 3000 watt hub is vastly superior in fit and finish, quality of the axle and will be more efficient and more powerful despite weighing a very similar amount to the HS3540. They are also quite cheap. You can get them in a number of different winding options, allowing for your desired wheel speed at any voltage level. 100v is really just a bad voltage level to be at. Controllers with 4110 fets are right at the maximum rating and will probably be unreliable. Some have had good luck, others have exploded countless FETs. 100v is also not high enough to justify the jump to 4115 based controllers. Your efficiency takes a hit from the higher resistance of the high voltage FETs and frankly they're still not very reliable.

Do yourself a favour and buy a Sabvoton and a MXUS 3k hub. No more issue with heat, more power than you will know what to do with and you won't need to open the motor or do anything tricky. Drop your maximum voltage and run a '72v' model OR increase it and run a 96v model. Personally I would recommend lower voltage with more current.

If you're serious about potentially building a number of these trikes then it would also make sense to design around the MXUS motor as it's more readily available and substantially cheaper than the much worse Crystalyte options.
 
Ohbse said:
Hall sensors are really essential for what you've got and what you would like it to do. If yours is one of the earlier HS3540's that didn't come with hall sensors then you might have a *really* fun time getting them in there as I don't think there's provision for them on the stator.

You're going to get a much better outcome just replacing the motor. The MXUS 3000 watt hub is vastly superior in fit and finish, quality of the axle and will be more efficient and more powerful despite weighing a very similar amount to the HS3540. They are also quite cheap. You can get them in a number of different winding options, allowing for your desired wheel speed at any voltage level. 100v is really just a bad voltage level to be at. Controllers with 4110 fets are right at the maximum rating and will probably be unreliable. Some have had good luck, others have exploded countless FETs. 100v is also not high enough to justify the jump to 4115 based controllers. Your efficiency takes a hit from the higher resistance of the high voltage FETs and frankly they're still not very reliable.

Do yourself a favour and buy a Sabvoton and a MXUS 3k hub. No more issue with heat, more power than you will know what to do with and you won't need to open the motor or do anything tricky. Drop your maximum voltage and run a '72v' model OR increase it and run a 96v model. Personally I would recommend lower voltage with more current.

If you're serious about potentially building a number of these trikes then it would also make sense to design around the MXUS motor as it's more readily available and substantially cheaper than the much worse Crystalyte options.

Im looking into these motors. You mentioned that they are cheap. Where did you find them>...any local distributors or just from china?


John
 
Back
Top