LightningRods mid drive kit
- Thread starter spinningmagnets
- Start date
teslanv
1 MW
LR- what is your suggestion for optimal battery voltage. I am hoping either a 16S or 20S would work well, as I have grown fond of the HK Turnigy 4S hardcase packs.
Wishes
1 kW
teslanv said:
Both voltages will work fine. I use HK hardpacks for my RC applications and know them well. If getting the max power is not your objective, I would stick with 16s simply for battery placement convenience. It will be easier packing and placing multiples of four 4s packs versus five 4s packs. And the second reason I would recommend 16s, again only if your objective is not achieving maximum power, it will be easier on your drive train. Although LR's parts hold up under that kind of power. The rest of your bicycle drivetrain will take a beating.
With 16s at the optimal 40 amps for this motor, you will still get lots of power and spare your drivetrain a little. Go to 20s ONLY if you want to achieve the maximum power. But be prepared to replace the rest of your drivetrain. I speak from experience with running 18s for a while and now 24s. You eat chains, chain rings, rear cassettes, I bent 2 gng jackshafts (i'm awaiting my LR hardened jackshaft), I went through 1 gng freewheel and also a second ACS freewheel, I'm on my third freewheel on the jackshaft. (Which will be replaced by LR"s #219 chain secondary upgrade mod because that third freewheel is not going to last.) And today i discovered that the cheap Chinese shaft of my original gng motor is now also bent. I got a significant wobble in my small pulley now. Not something you will have a problem with if you buy your motor off LR, he upgrades those shafts with quality ones and not the cheap soft Chinese shafts. All those parts in 1 season. And i ride it on the road mostly. The power is nice, overvolting is fun, I'm definitely an addict. But I do it knowingly that it comes at a price.
Unless you are using all of LR's upgrades, at those power levels, most of the stock gng parts will not last.
Unless you plan on going to a single speed setup, where you can use BMX quality chainrings, chain and rear sprocket, a typical bike drivetrain starts to take a beating at around 2500 watts.
I love the concept of having the use of my gears through the motor, but I do it knowing well that it comes at the price of the rest of my drivetrain. The price for overvolting.
Wishes
windtrader
1 kW
Fantastic post, wishes. I am designing my build now and just making the decision to go with a left side power drive exactly for the experience you state. I want a RELIABLE and DURABLE bike, not one wearing out and breaking down. Plug and go as they say. Also want to upgrade power when I figure out more power is more fun! After a LOT of research and diving into the long time and lots of experience in the gas motor bike world, it is clear driving the bike on its own drive line is the only way to go if you are going to power it up.
Initially I was considering a Nexus and running the LR via the right side drive. More than once I was advised that would result in a bucket of torn metal shards. Then a derailleur config. But that destroys the look of the board track racer I am building as well as now learning the practical design limits to true bicycle components. With power to the left side it does not matter what I run on the right side, even a single coaster brake hub or an eight speed Nexus would hold up equally well. That said, maybe a SA 3 speed drum brake rear would take the power. I have seen a gas motor bike send power through one, no idea if it is holding up.
Just getting back to the ebike side of the design now and feel my current requirements don't fit the full LR kit. But there are a number of parts I still plan to order even if I left drive the bike. Hope to finalize the ebike design in the next day or so and get my order in!
I asked before but did not see a response - Does anyone have some power graphs of the LR motor running at 48 and maybe 60ish?
Thanks
Wishes
1 kW
windtrader said:
I'm sure there was someone on the forum that did voltage tests and had posted a graph. But I just searched, quite a bit for it and cannot find it. It is probably buried somewhere deep in the main gng thread. Maybe someone else still has a reference to that and can post it.
Wishes
LightningRods
1 MW
turbo1889 said:
I totally agree. From the response that I've gotten from around the world there are still a lot of people who know that the US can do it when we want to. We've gotten a reputation sort of like that talented but under-motivated kid in school. "He has plenty of ability, he just doesn't apply himself". The US needs some good competition to remind us that our grandparents got ahead through hard work.
Another advantage the US has is that China doesn't want that work, at least not now. They want to send you a shipping container of cheap parts. I've had multiple suppliers tell me that an order of 100 sprockets or whatever is too small for them to bother with. In one case I was able to get 100 parts by doubling their per unit price on 500.
My biggest challenge is trying to upgrade these parts for the GNG and still keep them compatible with the original kit. And bring it all in at a price point that makes some kind of sense in the frame of reference of an entire kit that cost $400 delivered. In addition to cheap materials there is bad design everywhere in the GNG. I'm still routing that out, still upgrading parts, still looking for better suppliers that can deliver better parts at a better price.
People like Wishes, and Krister, and Bee are the torture testers who will sort out the weak parts. I listen to all of you and when something breaks or wears out prematurely I'm looking for ways to make it better. Wishes has been running 4 kW through my original Gen. 1.0 sheets and the open bearing block jackshaft! He's ordered my newest upper sheets and sealed jackshaft. We'll see how it stands up to his extreme use. If he breaks it, which he may, I'll upgrade it.
I'm not ignoring you on the power graphs, Windtrader. I don't have that data. Sorry.
I have a simulator program in Excel that is capable of producing graphs of any motor if I have enough data about it and the controller to plug in as the variables but it is not by any means 100% accurate and is only a theoretical simulation based off of known "rule of thumb" motor performance equations.
If you are interested in such graphs I can make them up for the GNG motor based on the specs. provided for it in the main GNG thread at various voltages and controller amps but there are only theoretical and I have found that in almost every case theory and actual practical although they may come close never do match exactly and sometimes the differences can be substantial enough to be misleading.
But if you all want some graphs I can provide some theoretical graphs based on the input specs. Both mid-drive multi-gear and left drive single reduction. But they will only be theoretical and in my experience with this particular motor the practical results tend to be less impressive then the theoretical calculated graphs based on its specs at least in the one build I have so far using this GNG motor.
If you are interested in such graphs I can make them up for the GNG motor based on the specs. provided for it in the main GNG thread at various voltages and controller amps but there are only theoretical and I have found that in almost every case theory and actual practical although they may come close never do match exactly and sometimes the differences can be substantial enough to be misleading.
But if you all want some graphs I can provide some theoretical graphs based on the input specs. Both mid-drive multi-gear and left drive single reduction. But they will only be theoretical and in my experience with this particular motor the practical results tend to be less impressive then the theoretical calculated graphs based on its specs at least in the one build I have so far using this GNG motor.
Wishes
1 kW
turbo1889 said:
Even theoretical graphs would be useful. Please post if you can. And thank you!
Wishes
windtrader
1 kW
The primary reason that I and I suspect others interested in this configuration is better understanding the capabilities of the motor and drive system to validate their component selection, batteries and controller too, will meet the desired performance goals in terms of raw power and distance between charges.
It seems like I came across some motor graphs but have no idea where and can not locate them now. It just seemed a pedestrian request as this motor is powering many bikes. Thanks Don
It seems like I came across some motor graphs but have no idea where and can not locate them now. It just seemed a pedestrian request as this motor is powering many bikes. Thanks Don
crossbreak
1 MW
bad thing is that most graphs end at very low power... the most reason why i never indented to invest in this (GNG) drive... i guess it will get lossy well below <2kW even at 48V. Nothing real special...you'd need 72V but can't pedal any more so..wtf? i think that my bafangs really can do better i have no proof..we really need dyno tests, at best on the same dyno with the same controller.
What we really need are torque mode throttle controllers with units like this.. everything else really is a waste of time.. lightningrods, please offer us such a unit with torque mode controller! it is really worth it i believe
What we really need are torque mode throttle controllers with units like this.. everything else really is a waste of time.. lightningrods, please offer us such a unit with torque mode controller! it is really worth it i believe
LightningRods
1 MW
Amplitude modulated controllers are a good idea. For the time being I am offering Infineon controllers that interface with the Cycle Analyst 3 which has several different tactics to moderate the 'kick in the back' that makes e-bikes difficult to control at take off and punishes drivelines. Here is the literature from Grin Technologies:
"Here are some of the more popular features you can include by using the V3 CA as your central ebike/EV console:
Temperature Sensor: A temperatures sensor input allows the V3 Cycle Analyst to scale back power as a motor heats up and prevent overheating damage.
PAS Sensor: The V3 Cycle Analyst can take the pedal pulse inputs of almost any ebike Pedal Rotation Sensor to display your pedal cadence. It can use this to provide automatic power whenever you pedal (PAS mode) .
Torque Sensor: There is an analog input to sense a pedal torque sensor (like the THUN), display your human power input, and give proportional no-throttle pedalec control.
Ebrakes: A digital input for ebrake sensors to plug in directly to the CA3, without needing to run another set of cables all the way down the bike to the controller.
Throttle Input Mapping: The CA3 can map your input throttle range to an output range suitable for the controller in order to eliminate dead-band zones at the beginning and end of the throttle motion.
Throttle Modes: Setup your throttle in Pass-Thru mode for conventional voltage control, or have your throttle directly regulate the battery current or motor power instead.
Throttle Ramps : Reduce the sensitivity and off-the-line kick of high power systems by setting ramp limits on the CA’s output throttle signal.
On the fly limit adjustments: Use a potentiometer or multi-position switch to instantly adjust any of your current/power/speed or pedal assist limits.
Mode Presets: Configure up to three distinct power mode presets each with their own limit settings and PAS/Throttle behaviors, and then easily switch between them with a double button press.
Battery SOC and LVC: Accurate State Of Charge (SOC) estimator based on your cell chemistry and voltage. Configure up to two batteries each with their own details and Low Voltage Cutoffs.
Plus many advanced capabilities to discover, and of course all the functionality that you have in the V2 CA like volts, speed, amps, amp-hours, %regen, watt-hours/km etc. An official user manual has not yet been prepared, but almost all of the information is exceedingly well documented in the unofficial users guide written by Endless-Sphere forum member Teklektik."
I've been informed by other technicians that this is not a pure solution, that it's a workaround. Honestly, I don't care. If it works it will get e-bikes on the road and get this hobby, this industry moving forward. Better solutions will come along eventually. We can all be out riding our e-bikes in the meantime.
"Here are some of the more popular features you can include by using the V3 CA as your central ebike/EV console:
Temperature Sensor: A temperatures sensor input allows the V3 Cycle Analyst to scale back power as a motor heats up and prevent overheating damage.
PAS Sensor: The V3 Cycle Analyst can take the pedal pulse inputs of almost any ebike Pedal Rotation Sensor to display your pedal cadence. It can use this to provide automatic power whenever you pedal (PAS mode) .
Torque Sensor: There is an analog input to sense a pedal torque sensor (like the THUN), display your human power input, and give proportional no-throttle pedalec control.
Ebrakes: A digital input for ebrake sensors to plug in directly to the CA3, without needing to run another set of cables all the way down the bike to the controller.
Throttle Input Mapping: The CA3 can map your input throttle range to an output range suitable for the controller in order to eliminate dead-band zones at the beginning and end of the throttle motion.
Throttle Modes: Setup your throttle in Pass-Thru mode for conventional voltage control, or have your throttle directly regulate the battery current or motor power instead.
Throttle Ramps : Reduce the sensitivity and off-the-line kick of high power systems by setting ramp limits on the CA’s output throttle signal.
On the fly limit adjustments: Use a potentiometer or multi-position switch to instantly adjust any of your current/power/speed or pedal assist limits.
Mode Presets: Configure up to three distinct power mode presets each with their own limit settings and PAS/Throttle behaviors, and then easily switch between them with a double button press.
Battery SOC and LVC: Accurate State Of Charge (SOC) estimator based on your cell chemistry and voltage. Configure up to two batteries each with their own details and Low Voltage Cutoffs.
Plus many advanced capabilities to discover, and of course all the functionality that you have in the V2 CA like volts, speed, amps, amp-hours, %regen, watt-hours/km etc. An official user manual has not yet been prepared, but almost all of the information is exceedingly well documented in the unofficial users guide written by Endless-Sphere forum member Teklektik."
I've been informed by other technicians that this is not a pure solution, that it's a workaround. Honestly, I don't care. If it works it will get e-bikes on the road and get this hobby, this industry moving forward. Better solutions will come along eventually. We can all be out riding our e-bikes in the meantime.
I can't wait to get my hands on the kit and try it out. Mike (LightningRods) has come up with upgrades for each part of the GNG kit, one by one. So far, each upgrade wasn't just better than the GNG part, it was great...with nothing left to improve.
I would enjoy trying out a plug-and-play torque-mode throttle + controller, there was a lot of discussion about those a year ago, and some work is being done on that (I don't understand it, but people I respect on electronics like it, so I want to try it when they are available). I am guessing Mike really has his hands full right now, and I have to take my hat off to him for fronting the cash to get 100 of these motors direct from the factory. I would wait at least six months before expecting something like that.
I was impressed with the performance that Christerljung was getting at 100V, but I can't imagine fitting 24S on my frame (I agree that 100V would be lossy, no matter how much fun it might be), the reports that really got me excited was Skyungjae. I have known him long enough that I trust his opinion on an off-road mid-drive. If a better motor shows up, I am certain Mike will try it out and make plates for it, but right now...I am just chomping at the bit to get this one on my bike!
I would enjoy trying out a plug-and-play torque-mode throttle + controller, there was a lot of discussion about those a year ago, and some work is being done on that (I don't understand it, but people I respect on electronics like it, so I want to try it when they are available). I am guessing Mike really has his hands full right now, and I have to take my hat off to him for fronting the cash to get 100 of these motors direct from the factory. I would wait at least six months before expecting something like that.
I was impressed with the performance that Christerljung was getting at 100V, but I can't imagine fitting 24S on my frame (I agree that 100V would be lossy, no matter how much fun it might be), the reports that really got me excited was Skyungjae. I have known him long enough that I trust his opinion on an off-road mid-drive. If a better motor shows up, I am certain Mike will try it out and make plates for it, but right now...I am just chomping at the bit to get this one on my bike!
Wishes
1 kW
spinningmagnets said:
I've read through the thread on the current throttle versus the voltage throttle that we are all used to using. It works with the new CA version 3. But I do not think anyone has got it working perfectly yet. I'm sure if someone has, they will correct me if i'm wrong. But the last I read on it, they had a hard time getting those oscillations down to respectable levels. It does not work plug and play out of the box. Not yet anyways. I am also considering trying it.
Wishes
LightningRods
1 MW
This evening we assembled our new 20 ton hydraulic press and started pressing the spindles out of the motors that I have in the shop.
It's far better than trying to drive the spindles out with a mallet. The spring action of the magnets absorb hammer blows. But they are helpless against a big hydraulic press.
Here is an assortment of different sizes of 'GNG' type motor spindles. On the left are big block spindles with cooling fans, in the center are 500 watt spindles with cooling fans, and on the right are the standard case width 500 watt spindles with no cooling fan. I found out the hard way that 500 watt motors that have the cooling fan also have wider cases. The armature and stator is no wider and the motor makes no more power, it just has a bigger case and the fan.
The big block armature is a beast! It looks like it's twice as wide as the small block armature. The next time I go over to the shop I'll take some measurements.
Tomorrow we'll fire up the furnace and anneal these motor shafts. We need to get this process down. 100 motors coming in about a week. All of them have this helical gear that has to be machined off for our purposes. We're going to be busy!
It's far better than trying to drive the spindles out with a mallet. The spring action of the magnets absorb hammer blows. But they are helpless against a big hydraulic press.
Here is an assortment of different sizes of 'GNG' type motor spindles. On the left are big block spindles with cooling fans, in the center are 500 watt spindles with cooling fans, and on the right are the standard case width 500 watt spindles with no cooling fan. I found out the hard way that 500 watt motors that have the cooling fan also have wider cases. The armature and stator is no wider and the motor makes no more power, it just has a bigger case and the fan.
The big block armature is a beast! It looks like it's twice as wide as the small block armature. The next time I go over to the shop I'll take some measurements.
Tomorrow we'll fire up the furnace and anneal these motor shafts. We need to get this process down. 100 motors coming in about a week. All of them have this helical gear that has to be machined off for our purposes. We're going to be busy!
crossbreak
1 MW
Nope. As the CA V3 uses closed loop control, it will always be laggy, no matter how good you tune.
Great press! i like it!
Wishes
1 kW
crossbreak said:
Nope what? You can use current throttle with the older versions of the CA?
Wishes
crossbreak
1 MW
no, V3 only..and only with lag
Wishes
1 kW
crossbreak said:
That is what I thought I understood. Only works with V3 and nobody has got the lag down to respectable levels. Its unfortunate because a current throttle would have been nice.
Wishes
--freeride--
100 W
- Joined
- Jul 5, 2011
- Messages
- 237
There are allready S09 sinus wave (bms-battery) controllers out there that have 9fet and are able to do current controll. We need to talk to Lyen.
panurge
10 kW
LightningRods said:
Hi LR
Iso hubs of the same width have the same offset between the flanges so doesnt matter which kind of hub you will use if not for the strength/reliabilty, that depends much on the bearings, though...
Used with my astro wihtout issues a DMR Revolver, a WTB laser disc, and a sunringlé 150mm.
The most sensible variables of the LH transmission indeed are: the offset of the spider adapter, the width of the inner part of the caliper body (Narrower = easier) the tire size, and the chain-line of the motor output. Also, Find out a very useful frame for this purpose, where the stays are not interfering with the motor line, it is not easy even among the rigid frames, figure out, about FS ones....
With a fatbike (I have one in my hands right now) I found that I could be able to mount the Astro Drive made by Matt (Recumpence) and keep the primary big pulley between the pedal chainring and the frame tubings (100mm BB shell)
I only need to make a longer jackshaft and its shell for this mod....
crossbreak
1 MW
this is a middrive thread, no controller thread.. anyway, the S-Series from BMS battery has no torque throttle, just torque steps if you use PAS. If you connect a thumb throttle, they behave like any other cheap controller.
Only torque throttle controllers I know: Kelly, Bafang, Sabvoton, and some DIY (Lebowski...)
Edit: That's wrong, Kelly is also no real torque throttle
Only torque throttle controllers I know: Kelly, Bafang, Sabvoton, and some DIY (Lebowski...)
Edit: That's wrong, Kelly is also no real torque throttle
LightningRods
1 MW
panurge said:
Are you planning a chain run that goes both above and below the BB shell on your fat bike? It seems like the only way to clear the rear chain stays when starting the chain run from in front of the BB.
Do you have photos of your LH drive rear hub, Jules? I'd like to see how it went together. Thanks.
LightningRods
1 MW
crossbreak said:
What about the Kelly controllers? Do they work as you suggest or is there a problem? They have a range of e-bike controllers....
http://kellycontroller.com/ebike-brushless-controller-keb24v-72v-c-24.html
panurge
10 kW
LightningRods said:
Yes, that would be the plan for the fat, but I'm thinking about to find a proper hubmotor for that kind of rear dropout width (170mm). I think scooter's hubs should fit....would like a long and narrow 750w geared hubmotor for this bike, But I don't know if that kind of motor is available....
Another place for the motor output in a rigid frame, is at middle of the seat tube level, but depends on its length and the chainstay...and mostly on the rear sprocket size; it has the disadvantage that if the chain fails, its power line point out just at your ass/legs
I realized personally, that pedaling out of the saddle with that kind of setup is really dangerous 
I was lucky that the motor was not powered, that time, but the same freewheel in action, tried to eat my trousers and even more.... 26" fat-wheels on a 29" fat-frame should leave the room to mount the drive behind the set-tube that should be a convenient place....
My rear wheel before and after
LightningRods
1 MW
panurge said:
That's really sanitary! And you managed to keep the disc brake. You're using the same mounting holes on the hub for both sprocket and rotor with spacers in between? I know that regular old center pull brakes are probably adequate for a rear brake but I much prefer a disc brake if possible.
The hub drive/fat bike setup that sounded kind of cool to me was a dual hub that gives you all wheel drive. That sort of suits the fat bike go anywhere utility.
