Sensor or Sensorless Controller...?

[dogman dan]

It should.

RE the applying throttle while rolling back. What I read somewhere, 4 years ago, was something about possible blowing the controller. Here's my experience with the Aotema sensorless 36v 20 amp controllers. From a dead stop, they will usually start forward. Once in a while, they will try to start backwards, roll one magnet, then swap directions and then start forward. Just a teeny tiny roll forward while the throttle is turned on, and they will ALWAYS start smooth in the forward direction.

If you are parked on a hill perhaps, and you start to roll back as you put on the throttle, the motor may chatter a lot, getting confused. It tries to roll forward, but you and your weight keep rolling backward. So it keeps repeating that which way do I go cycle.

I assume this sends big amp spikes through the controller. I never blew my controller as I was warned could happen. But I would think it's a harsh test for the components the controller is made from.

 

[motomech]

My [limited] experience with sensorless controllers is that the smaller/lower power the geared motor, the better they work.
....
Oh, there was one difference. When pushing the bike backwards without the sensors, the motors would cog. You can feel the push and pull of the magnets as they pass over the poles. Kind of like bumps from a square-toothed gear wheel.

But the hub motor still freewheel (no cogging) in the forward direction, right?

Oh yes, it free wheels forward the same.
All this talk about running backwards reminds me of a 1972 Yamaha MX100 i had. It was old skool two-stroke, piston port and breaker-point ignition. Without the reed valve induction, the intake timing was symeterical, it didn't care which way it was rotating.
But at the time, I didn't know a reed valve from a muffler bearing, I was just trying to climb the big "Sand Hill| near our house in Idaho. And as usual, I didn't make it and before I turn around[it was steep], the bike started to roll backwards with the clutch out[it had stalled]. A couple of engine bumps, and it imediately restarted going backwaRDs. I pulled in the clutch, but still fell over.

 

[SamTexas]

All this talk about running backwards reminds me of a 1972 Yamaha MX100 i had. It was old skool two-stroke, piston port and breaker-point ignition. Without the reed valve induction, the intake timing was symeterical, it didn't care which way it was rotating.
But at the time, I didn't know a reed valve from a muffler bearing, I was just trying to climb the big "Sand Hill| near our house in Idaho. And as usual, I didn't make it and before I turn around[it was steep], the bike started to roll backwards with the clutch out[it had stalled]. A couple of engine bumps, and it imediately restarted going backwaRDs. I pulled in the clutch, but still fell over.

Learning something new today. Thanks for sharing. I had a 50cc 2 stroke Suzuki back in the early 70s. No idea if it had a reed valve or not.

 

[amberwolf]

I have an early 70s Suzuki 2stroke (whcih needs a cylinder job to run)...nice to be forewarned about the possible problem.

I dunno much about it's engine at all, but it's this one:

 

[zooz]

I would go with the complete crystalyte system but one thing needs to be fixed with the hs3540, i think something of a rubberpart near the wires going into the wheel. after some times the rubber seal reveals a string of metal, solved with changing it to a 100% rubber one. seems to be easy fix.

48v, and lmn battery will manage 25A as what i am using now, about 1300W going uphill. Sensorless will need that u push the bike while seated for less than 1 sec and 1 feet and you'll be on your way accelerating. maximum speed of 45km/h in a straight line, 30km/h uphill if you are 120lbs. 10ah battery i have i usually use for 18miles , about 1200$

My next bike system will be about 72Volt at 45a sensorless controller and hs3540 rear motor and 18s1p batteries 5ah for 10mile ride. about 900$

I have a similar setup, I have got fully crystalyte HS3540 with 48V 40A controller. I am wondering what controller are you going to use for your 72V 45A setup? Thanks

 

[zooz]

I would go with the complete crystalyte system but one thing needs to be fixed with the hs3540, i think something of a rubberpart near the wires going into the wheel. after some times the rubber seal reveals a string of metal, solved with changing it to a 100% rubber one. seems to be easy fix.

48v, and lmn battery will manage 25A as what i am using now, about 1300W going uphill. Sensorless will need that u push the bike while seated for less than 1 sec and 1 feet and you'll be on your way accelerating. maximum speed of 45km/h in a straight line, 30km/h uphill if you are 120lbs. 10ah battery i have i usually use for 18miles , about 1200$

My next bike system will be about 72Volt at 45a sensorless controller and hs3540 rear motor and 18s1p batteries 5ah for 10mile ride. about 900$

I have a similar setup, I have got fully crystalyte HS3540 with 48V 40A controller. I am wondering what controller are you going to use for your 72V 45A setup? Thanks

 

[snellemin]

Lyen 12fet sensorless controller is pretty badass.

 

[zooz]

Lyen 12fet sensorless controller is pretty badass.

Does it have a cycle analyst direct plugin and regen support? Thanks

 

[snellemin]

Mine doesn't have regen, but Lyen can add it for you along with the CA connection.

 

[silverheels]

As far as 'getting off the line' is concerned, I would say dogman hit the nail on the head. These controllers always end up driving the motors 'forwards' unless there is some resistance (perhaps slight incline) in which case they 'cog' backwords, try to drive forwards and stall out.

On the flat I do not notice the difference between Sensored and Sensorless (though I combine pedal pressure with throttle application to help with efficiency). On steeper hills I consciously get the bike rolling, not just pressure the pedal forward, but start the wheels turning, as I add throttle. It would seem that the controller will not apply much power until a couple of 'segments' have rotated past the commutator, after that then full current is let loose and the wheel drives forward. Given just a slight push forward and they do hill starts just fine. My seat of the pants guess would be less than 1/8 of a turn of the wheel...

Ed modified these controllers at purchase so that they can run either upto 18s (72v nominal) or at 24s (88v nominal) through the use of an external switch. Ed also added extra long phase leads and PP45 connectors per my request. I'm not sure if Lyen would advise it, but I have one that occasionally runs at 6kw off the line!
They seem hard to beat. Great price, good quality and Lyen's legendary service. And I love not having any more hall sensor issues (and use the hall wires for temp sensors).
From my perspective (commuting and touring), Sensorless controllers are the way to go.

Note: My experience is limited to Lyen Mk2 Sensorless 12Fet controllers, I've not tried any others. YMMV!


Here's the current (no pun intended) Stable (all running Lyen Mk2 Sensorless 12FET controllers):

CLWB Recumbent (BikeE AT) - 30a at 18s - un modified shunt - to a 9c 8x8 20" rear.
Wedgie (non recumbent for all y'all un-enlightened e-cyclists ) - 45a at 18s - slightly modded shunt - to 9c 10x6 26" rear.
LWB Recumbent (Ryan Vanguard - known hereabouts as 'The Flying Belly' - something to do with the shape of it's owner!) - @60a at 24s - 'way' modded shunt - HT3525 26" rear 6200W .

 

[ohzee]

So I am using a Lyen Sensorless 12 fet mk2 controller that's stock from Lyen , but when I asked him the shunt value
I am told 1.333 or 2.0. So far I have been using this with my lifepo4 batteries so I currently don't have a way to
measure how much I put back into them when I charge.

Can anyone help with the shunt value ?

Worse case I have something to measure at that voltage just need it to show up.

Thanks in advance.

 

[ohzee]

Well did a ride yesterday with my new battery and with my CA @1.333 I did 30 miles and used 32ah out of a 24s20ah lifepo4.
So going to try the shunt value at 2.0 and see how that works.

Anyway glad to see I have more battery then I had thought.

 

[zooz]

Well did a ride yesterday with my new battery and with my CA @1.333 I did 30 miles and used 32ah out of a 24s20ah lifepo4.
So going to try the shunt value at 2.0 and see how that works.

Anyway glad to see I have more battery then I had thought.

I have got the same controller, but mine is "overclocked", Lyen upgraded, so it has different FET afaik. Lyen told me that RShunt value is 1.3333 and that seems to match what I use and what I put back while charging.

So I guess it makes sense to try rshunt 2.0.

 

[ohzee]

Thanks zooz for the reply that helps.

How much more do you get out of it being over clocked ?

If I knew that was an option at purchase id have asked for it myself.

 

[kudos]

I've have a DD Crystalyte setup HS3540 40A Sensorless run at 12S LiPo - and - I have a Geared sensored Mac 10T 30A setup too.

Personally I find this whole starting thing a non-issue. Yes if you run sensorless and you twist the throttle from a dead stop, you might sometimes get a slight backwards rotation before the controller locks on then propels you forward, but honestly how difficult is it to give the bike a miniscule forward movement before turning the throttle? Totally natural and easy.

I find sensored geared can be a little jerky on takeoff anyway, unless you have some good control of the throttle. The Crystalyte seems to give a nice smooth pull no matter what.

The other issue to consider is that the Crystalyte looks more professional than most of the other Chinese kit and comes plug and play, no soldering, no finding obscure connectors, no multi-coloured wires everywhere like a childs' toybox. Don't get me wrong they both have pros and cons, DD xlyte is heavier and more expensive, possibly less efficient but I'm glad I got that kit first.

Kudos

 

[zooz]

Thanks zooz for the reply that helps.

How much more do you get out of it being over clocked ?

If I knew that was an option at purchase id have asked for it myself.

As far as I remember it was not an option, he just sent me one. For me there is no difference, I thought I was going to run my motor at higher amperage, but I am running it on 35-45A peak and it's more than enough.