GNG, 1000W 48V BB-drive, $400

[crossbreak]

This roller pressing down against the belt at just the spot where the teeth would lift up and begin skipping might be a "doable" improvement that would raise the power limit on a primary belt.

For what? Bike parts are already made to throw them into trash. Why even speed up this? Why not make something that lasts? I have never seen any engineer try something that horrible for efficiency means. Belt pressure has to be adjusted. What you try here is only cosmetics on the symptoms, no solution.

Matt has gotten good success by widening the belt and pulleys.

That's a solution.

Denise made the best one I saw on this drive here.

 

[Denisesewa]

This roller pressing down against the belt at just the spot where the teeth would lift up and begin skipping might be a "doable" improvement that would raise the power limit on a primary belt.

For what? Bike parts are already made to throw them into trash. Why even speed up this? Why not make something that lasts? I have never seen any engineer try something that horrible for efficiency means. Belt pressure has to be adjusted. What you try here is only cosmetics on the symptoms, no solution.

Matt has gotten good success by widening the belt and pulleys.

That's a solution.

I would have to agree, bandaids just wont provide the solution for poor engineering.

 

[LightningRods]

I'm going to make a good effort to keep belt primary drive. OEMs and first rate fabricators are making it work. I'm going to start with 3/4" (19mm) longer sheets that have 3/4" of adjustment in them. The cutting paths are drawn. I'll take it to the CNC laser cutter tomorrow. If the stock belt still fails or has premature wear I'll machine the drive pulley off and go to 20mm wide, 8mm pitch pulleys and belt. I've been e-mailing an engineer at Gates and he's offered to help trouble shoot if we keep having problems.

Once I have my own prototype sheets sorted anyone on this forum that wants a set can get them at cost (sheet metal plus CNC) and shipping.

 

[Denisesewa]

I have been working on the resistor mod with the stock controller and the stock throttle I got with the second kit, this throttle is way worse than the first one, so I ended up having to put three resistors in series for the low speed(red wire) @ 88 ohm and two in series for the high speed (Black wire) @ 66 ohm however in order to make the throttle work at all I have to touch the red wire to the green wire which spins the motor momentarily and then it works for less than a minute, it does have the desired effect of taking out alot of the dead space but its not usable.
So, is it that I cant put resistors in series to get to the desired Ohms? are resistors directional? the local shop has a very limited assortment of resistors and a single 33 ohm had no effect that I could tell, also I put them in at the throttle side of the plug if that matters, kinda confused.
I also tried the magura throttle with the stock controller and it works only up to hitting the high rpm then quits untill I release it and apply throttle again, it also has huge dead spots on both ends.

 

[christerljung]

This resistormod is a little off-the-rules in the way that the hall-sensor inside the throttle doesnt get the correct voltage.
To get most out of it i just try two 100ohms variable resistiors at pos and negative wire.

The magura is an old-school reistorthrottle from what ive´ read. It has 0-5 kohm.
This means you should be able to fiddle around alot more with the reistorvalues, beacause there is no hall-sensor getting confused by the manipulated 5v.
But you probably gonna need larger resistorvalues. (variables at 2,2 kohm).

 

[LightningRods]

Arrgh. That sounds terrible, Denisesewa. I am NOT looking forward to re-engineering the electronics on this kit after I finish re-engineering the mechanicals. I'm a lot less competent in that venue.

I found a three chainring front freewheel while I was looking for options in this area. I thought I'd heard people wishing that they could use their front derailleur with the freewheel. It appears you can. Plus the driven sprocket is a 48 instead of the stock GNG 44 which will give you about 9% more secondary reduction.

http://www.eclipsebikes.com/chainwheel-cranks-p-1008.html?osCsid=dc2189032d15f9e4933ac2534efbef54

 

[Rassy]

Denise, look at the bottom of page 60 of this thread. Teklektik gave the best explanation of what is being done. I couldn't have made mine work with off the shelf resistors. There are very definite points where the throttle quits working entirely. When I had no resistor on the positive wire the upper limit on the negative wire was less than 30 ohms. When resistance was applied to the positive wire an equivalent additional amount (approximately) of resistance could be applied to the negative wire. I think I ended up with about 30 ohms on the positive wire and 50 ohms on the negative wire.

 

[LightningRods]

I think I ended up with about 30 ohms on the positive wire and 50 ohms on the negative wire.

Stock throttle and controller, Rassy?

 

[gwhy!1]

Most ( but not ) all controllers that I have played with have a maximum throttle input voltage, and above this voltage it will just cut out, this is a safety feature. Should the signal wire and +5v become shorted this will normally put the controller into full throttle, but if the controller has this safety feature then controller will shut the throttle off if the input is higher than around 4.2v, this is more than likely why the magura cuts outs out full throttle. you will need to put around a starting point of around 500ohms in the + supply to the magura ( it may need a bit higher or a bit lower ), this will not allow the magura to put out so much voltage at full throttle so will not cut the controller out.


Edit:

I have been working on the resistor mod with the stock controller and the stock throttle I got with the second kit, this throttle is way worse than the first one, so I ended up having to put three resistors in series for the low speed(red wire) @ 88 ohm and two in series for the high speed (Black wire) @ 66 ohm however in order to make the throttle work at all I have to touch the red wire to the green wire which spins the motor momentarily and then it works for less than a minute, it does have the desired effect of taking out alot of the dead space but its not usable.
So, is it that I cant put resistors in series to get to the desired Ohms? are resistors directional? the local shop has a very limited assortment of resistors and a single 33 ohm had no effect that I could tell, also I put them in at the throttle side of the plug if that matters, kinda confused.
I also tried the magura throttle with the stock controller and it works only up to hitting the high rpm then quits untill I release it and apply throttle again, it also has huge dead spots on both ends.

it sounds like you will need a much lower resistance on the hall throttle than what you already have. The Values of resistors that you are using have taken the hall sensor below its operating voltage/current range.

These are exactly the reasons I now make my own throttles

 

[Denisesewa]

Thanks to everyone posting on this, now I understand a bit more ( and had a few beers to alleviate a day of frustration) so will buy some pots and try all this again.
I bought the magura for the 12 fet lyon controller so hope I can make it work on that bike. What I have learned is that the best thing about electronics is the cool little wisp of smoke when you really screw up and that LCD screens put out some really fun psychedelic displays when you get pissed off enough to punch your computer :lol:
Ahh, stress relief!!



Give me a hammer,saw and cresent wrench and I'll build you a car, but you have to wire it.

 

[Denisesewa]

These are exactly the reasons I now make my own throttles

There may be some plagiarism happening real soon, watch your six!!!, I'm not above a little R+R ( ripoff and replicate)

 

[gwhy!1]

These are exactly the reasons I now make my own throttles

There may be some plagiarism happening real soon, watch your six!!!, I'm not above a little R+R ( ripoff and replicate)

I bought the magura for the 12 fet lyon controller so hope I can make it work on that bike. What I have learned is that the best thing about electronics is the cool little wisp of smoke when you really screw up and that LCD screens put out some really fun psychedelic displays when you get pissed off enough to punch your computer :lol:
Ahh, stress relief!!



Give me a hammer,saw and cresent wrench and I'll build you a car, but you have to wire it.

The mag throttle will work ok with the 12fet controller so stick with it... but will need a bit of fudging to get it right.

I always thought that a hammer,saw and wrench was the basic tool kit for electronics

Feel free to R+R

 

[LightningRods]

Give me a hammer,saw and cresent wrench and I'll build you a car, but you have to wire it.

My thoughts exactly!

 

[LightningRods]

What do the Sparkys think about this option for throttle tuning?

Wire a Potentiometer as a Variable Resistor
http://www.instructables.com/id/Wire-a-Potentiometer-as-a-Variable-Resistor/

If I read this correctly one pot can be wired one way to adjust the low range (i.e. left lead and center lead) and then the other wired the opposite (right lead and center lead) to adjust the high range. Sounds ideal.

 

[mr.electric]

I battled with jumpy or twitchy throttle until I installed a Fechter throttle control circuit. It worked amazingly well. Unfortunately it only works for old brushed motor circuits. At least the schematic I have only works on brushed set ups. It was such a useful circuit for improving drivability that Bramo bought the design from Fechter and added it to all their bikes. I have only built a handful of circuits in my life but I was able to build this small simple circuit. It installs out of harms way inside the controller. Now that I think of it the Cycle Analyst introduced a variation of this circuit in the newer units. The original circuit was more simple and has better driving characteristics than the Cycle Analyst variants that I have tried. It had a knob that adjusted the throttle curve and had one great feature, full throttle is always full throttle it only controls the lower portion of the throttle curve. I'll hunt for the schematic here.

 

[Rassy]

Rassy wrote:

I think I ended up with about 30 ohms on the positive wire and 50 ohms on the negative wire.

LightningRods asked:

Stock throttle and controller, Rassy?

All stock Gen 1 on 48V. I did have another thought based on what others reported, that the QC may not be too great on these throttles and controllers so what works for one may not work the same for the next, etc.

EDIT:
LightningRods wondered:

Wire a Potentiometer as a Variable Resistor

http://www.instructables.com/id/Wire-a-Potentiometer-as-a-Variable-Resistor/

I was wondering the same thing. The resistors I got from dnmun vary from 0 to 5k. Made fine tuning quite difficult. I looked around and saw some 0 to 100 ohm potentiometers that would probably work pretty slick.

 

[crossbreak]

As you guys all bother about throttle issues, I have decided to start an Arduino project on this. I use the Arduino Nano V3 to map a new throttle behavior. Still have some issues since I lack of knowledge about this topic. The Arduino will measure phase current and it can compute motor speed from this. Yet I have only connected a hall throttle, remapped it and connect it to a 6-Fet. This works fine on the bench.

It's really straight forward since it only took an hour to get so far. I dont know much about programming yet. Will start a new topic about this soon and post a link here. Hope I get as much resonance from you as for my last project when converting the Hubmotor.

 

[mr.electric]

This is a brushless variant of what I used with great success.

 

[Sirbevalot]

Just had my first ride!!

My local bike shop man fitted it for me for $60. He also adjusted my brakes and gears etc.

Just what i expected. It is a tottally different experience. went for a 7 km trip and my 12s 10 ah pack was reading around 3.9 volts per cell when i got back home. I went up some hilly bits so i reckon on flat i could go about 15 km draining the lipos to 3.3 volts each.

What is everyone alse getting (range wise) on 12s 10 ah ????

Just need to tidy up the wiring and devise some sort of battery enclosure. Still thinking of doing a back pack set up.

Having trouble putting pics on...... as soon as i figure it out i will post them.

 

[Sirbevalot]

This is my first build. SPECIALISED Hardrock. Now 8 speed.

GNG 450 watt kit.

4 x Zippy 6s 5 AH to make 12s 10 ah pack. (currently duct taped on!!)

Here are some quick (bad) pics.

 

[amberwolf]

wow those are huge files. why not use a resizer before uploading, and put them as jpg or similar instead of zip, so people can view them in the post?

 

[Sirbevalot]

wow those are huge files. why not use a resizer before uploading, and put them as jpg or similar instead of zip, so people can view them in the post?

I'll have a play around. First time i've ever uploaded a pic

 

[christerljung]

This thread has nice pics and explanations on the resistormod for the magurathrottle.
http://endless-sphere.com/forums/viewtopic.php?f=2&t=26283
It is praised for look and feel.
However wouldn´t a variable resistorthrottle be more sensitive for water and corrosion?
Hallsensor must be bullet-proof and should be working even if youre able to ride underwater?

 

[gwhy!1]

This thread has nice pics and explanations on the resistormod for the magurathrottle.
http://endless-sphere.com/forums/viewtopic.php?f=2&t=26283
It is praised for look and feel.
However wouldn´t a variable resistorthrottle be more sensitive for water and corrosion?
Hallsensor must be bullet-proof and should be working even if your able to ride underwater?

The maguras are a very well made bit of kit and it do feel nice but still lack in the range of twist, for 95% of people it will be a massive improvement on a cheapo hall throttle. I found that the pot in the magura became noisy after around 2 hours of trials use ( so bad that it became unusable for me ). I would put trials use as one of the top things that will wear out and highlight any problems with a throttle. Normal bike riding involves smooth transitions from one end of the throttle to the other , trials use is continually adjusting the throttle position. The pots that are in the maguras claim they are water proof so water/dirt should not effect them but I would rather have a hall throttle any day of the week for reliability and control ( as long as its setup well )

 

[Denisesewa]

Well , I went and purchased two 1k 15 turn cermet potentiometers and installed them ,

I figured the 15 turm would allow very fine adjustments, this does work to a point but the throttle still has huge dead spots both top and botttom, I dont know why anyone would say this is a good mod, in my opinion its a waste of time, there is no way to get full actuation from the throttle at least with these stock GNG components.
For all the effort put into these bike builds I cant believe this issue hasnt been resolved or that manufacturers would put components together in a package that are so incompatible and inconsistant.
The throttle that came with the second kit appears to be the same except for the grip but it behaves way worse than the first one even with the same controller so it seems to me that either the hall sensor is built with huge tolerances or the magnets are inconsistant, I am determined to find reasonable solution for the stock components or a inexpencive replacement which will improve things enough to make these kits safe and fun to use.
For this second build I did not want to spend the money for another Lyen controller and magura throttle , I am hoping to get the magura to work well with the Lyen on the first bike but I'm not going to get into that untill I get something working on the second bike.
I think I will also attempt to build a Gwhy setup once I scrounge up a decent motorcycle throttle but I think there will still be issues trying to match throttle output to a specific controllers specs so "one size" propably wont fit all.
The saga continues.
Edit - this resistor mod may indeed work well on certain types of throttles, I now know the GNG kits ship with at least 2 different types.