GNG 450 brushless Mid Drive - Primary Drive Modifications

I'm definitely "IN" with the adjustable sheets. However, I have a gut feeling that part of the reason the GNG primary has that "deep plunge" idler, is for teeth engagement. I seem to remember someone installed a smaller belt (same width) on the stock kit. This resulted in stripped teeth, thought to be caused by less engagement (or lack of an idler?). Anyone remember this, gotta link?

Anyhow, I see these sheets as the best "first step" because it allows us to be rid of that crazy stock idler, the likes of which I have never seen before. If I end up with a chain drive later, the adjustable sheets should make adaptation that much easier, anyway.

Edit: Oops! Just realized that the sharp edged (thanks, LightningRods) drive pulley could be the "teeth stripper" of the smaller belt conversion, hmm...
 
The order for the first beta production run of adjustable sheets is at the CNC shop. I stuck my neck out and ordered 10 sets for the Forum to get a better price. Thanks to your support I sold all of those and had to order more. In addition to the sharing of valuable knowledge and experience this is another great thing about E-S. We can go in together and have custom parts made at prices that start to become reasonable. I'll be contacting everyone who placed an order to wrap up billing and shipping. Everything will go out this week.

If you've found this post late, as often happens, and want a set of adjustable plates for your Gen 1 GNG send me a PM and I'll see what I can do. The initial run is sold out as of 1/14/13.
 
Cyclebutt said:
I'm definitely "IN" with the adjustable sheets. However, I have a gut feeling that part of the reason the GNG primary has that "deep plunge" idler, is for teeth engagement. I seem to remember someone installed a smaller belt (same width) on the stock kit. This resulted in stripped teeth, thought to be caused by less engagement (or lack of an idler?). Anyone remember this, gotta link?

Anyhow, I see these sheets as the best "first step" because it allows us to be rid of that crazy stock idler, the likes of which I have never seen before. If I end up with a chain drive later, the adjustable sheets should make adaptation that much easier, anyway.

Edit: Oops! Just realized that the sharp edged (thanks, LightningRods) drive pulley could be the "teeth stripper" of the smaller belt conversion, hmm...
Click to expand...
IMO the sharp teeth on the small drive gear are a result of using a round nosed cutting bit on the cylinder of steel that the entire rotor spindle is machined from. They make longitudinal cuts and don't bother to radius the top of the cog (which engages the base of the belt cog). Easy, lazy, cheap production the way that much of the kit was made. At this point all we can do is keep going down the list of half-assery, correcting until we get acceptable results. I come from a background in hot rodding cars so I am used to making silk purses out of sow's ears. :D
 
LightningRods said:
crossbreak said:
I proposed to build such sheets earlier, but the main factor that keeps me from buying is the tiny belt sprocket. A chain solves all issues IMO in a much better way.
Click to expand...

I'm going to start out trying to keep the belt drive. I'll admit in advance that I may not be able to. The stock pulley is too small in diameter and possibly too narrow at 15mm. I think it may work fine at 1 kW which is where I'm going to start out. There is another issue with the stock drive pulley, and I don't know that I've ever seen this mentioned on E-S. The way that it is machined creates a sharp edge on the cog profile. If you compare the GNG sprocket to a more typical profile you can see that the way it was machined creates a sharp cutting edge (red arrow).

Sprocket_Profiles.jpg


I think this may be a big factor in premature belt wear on the standard drive.

So to sum up the primary belt drive on the stock GNG:

No provision for proper tensioning
Provided tensioner bends belt backwards damaging carbon support layer
Drive sprocket too small a diameter
Drive sprocket has atypical profile with sharp edges
Driven sprocket has excessive runout causing belt tension to fluctuate
Mild steel sheets may be bent in transit causing improper alignment of belt
Entire belt system may be too narrow (15mm vs. 20mm)

If I continue to have belt problems after correcting for belt tension I'm going to start solving for the other issues. If it still doesn't work I'll admit that carbon belts suck and joing the rest of the gang with a chain primary.
Click to expand...

just a thought, but what are the chances of useing a belt tensioner from a car timeing belt for the primary drive?? ok the bearing might have to be removed from the centre then a spacer made up ,, or the centre to be modded some how, but they are made so that they do not strip the belts ,, and the belts last for 10s of thoughsands of miles under a lot more stress than one of these kits could very give out ..

there is also the timing wheels on a car that could possibly be used as well for the larger wheel

here is a link to different types ,, its a thought to ponder on http://www.ebay.co.uk/sch/Vehicle-Parts-Accessories-/131090/i.html?_from=R40&_nkw=timing+wheel

file.php
 

Attachments

  • 2009-04-30_181043_Suzuki_Timing.jpg
    2009-04-30_181043_Suzuki_Timing.jpg
    39.7 KB · Views: 2,288
Needmorespeed said:
justforfun said:
We also have got rid of the belt, in the UK mud this time of year was killing belts on average every 8 miles, if it gets wet and slips it snaps the belt.
We have changed to BS 04b chain and sprockets 12t on motor & 57t on the jackshaft. With this setup there is no need for a tensioner or different mounting plates :D but we did use a half link.

Link for chain and sprockets http://www.technobotsonline.com
Click to expand...
Here's some pics to go with your post!
You'll have to excuse the muddy bike! It's been stupidly muddy recently :wink:
Now the snows arrived it might clean it up a bit :D
It's a great little mod, hope it holds up.
Both running over 60v nominal ,also running 48 tooth sprocket to good effect.
Click to expand...
 

Attachments

  • image.jpg
    image.jpg
    37.6 KB · Views: 9,210
  • image.jpg
    image.jpg
    33.1 KB · Views: 9,210
  • image.jpg
    image.jpg
    28.3 KB · Views: 9,210
So it appears the BS-04 chain is 6mm pitch which would be very close to the #25 chain I use which would be roughly 6.35mm , nice to know these sizes are close and seem to be holding up well on both sides of the pond.
 
I received an e-mail from an engineer at Gates Corp. that I have been communicating with about our belt problems. I sent him pictures of the small 14T drive pulley and also attached my diagram showing how the pulley teeth have a flat top with sharp edges. This was his response:

"The sprocket with the sharp edges is a definite no-no. It’s very likely a big part of the cause of how your belts look after the short run time that you mention."

He asked for some more information which I sent on to him. I'll share return information as I get it from Gates.

You trail riders have some very valid points in favor of chain drive. I've ridden both dirt and street motorcycles for years and they're very different. I'm building a street fighter out of a hardtail MB so I want clean and quiet. For off road you just want the thing to perform and get you home.

I expect that the stock drive pulley is junk and needs to be machined off no matter what. The only option is to keep a backpack full of drive belts with you. I'm going to talk to Christer about extracting the rotor spindle so that it can be properly turned down to 10 or 11 mm and a keyway machined in. The file technique is okay but it limits the choices on pulleys and sprockets.

Sprocket_Profiles.jpg
 
ATTACH]
Here's another photo of the stock belt configuration after 170 miles. Only around 30 of these miles were trail. Can you see the edge fraying?

I'm going with adjustable sheets for now.

1) During installation I will rotate the motor power lead to a more "protected" clock position.
2) Run shorter belt with sheets in most closed position. Anyone have that part number? I'll look for it. test it..
a) In case belt's too short for easy installation, go back to original belt. test it...
3) Find a screw type idler for this configuration. It is my belief that a spring loaded idler is not appropriate for this type of belt, also not designed for serpentine use, other members have confirmed this. A spring type idler allows jerking motion which (IMHO) leads to premature belt failure. A screw type idler would allow enough slack in the belt line for easy belt change in the field, would allow perfect "FIXED" tension adjustment as well. test it...

I'm fairly confident these mods will at least quadruple the belt life. If it'll go three thousand miles, I'm satisfied. If not, I'll go with the chain mod. Adjustable sheets will still allow easier adjustment for chain drive choices. Besides, if I have to turn down (machine off) the drive pulley, a drive "sprocket" will replace it, I'll be finished with the attempted salvaging of the stock primary drive.

This bike is being built as a commuter with occasional off-road capability. Sort of a KLR (Kawasaki motorcyce) of bicycles.

So this is my "IF-THEN" plan. Anyone remember the part number for the shorter belt and any suggestions for a screw-type adjustable idler that will slap right on, without (or very little) modification?
 

Attachments

  • 100_2034 (2).jpg
    100_2034 (2).jpg
    90.8 KB · Views: 2,867
Cyclebutt said:
ATTACH]
Here's another photo of the stock belt configuration after 170 miles. Only around 30 of these miles were trail. Can you see the edge fraying?

I'm going with adjustable sheets for now.

1) During installation I will rotate the motor power lead to a more "protected" clock position.
2) Run shorter belt with sheets in most closed position. Anyone have that part number? I'll look for it. test it..
a) In case belt's too short for easy installation, go back to original belt. test it...


I know that the people who tried the short belt said it was an extreemly tight fit and at least one was damaged trying to install it, you may be able to deepen the slots in the adjustable plates to allow ease of installation. I know the part # is in this thread somewhere but couldnt find it right away. I also think the hard edges of the pully teeth need to at least be smoothed over, I am trying to think of a way to make a tool to do this.
Click to expand...
 
Denisesewa said:
Cyclebutt said:
ATTACH]
Here's another photo of the stock belt configuration after 170 miles. Only around 30 of these miles were trail. Can you see the edge fraying?

I'm going with adjustable sheets for now.

1) During installation I will rotate the motor power lead to a more "protected" clock position.
2) Run shorter belt with sheets in most closed position. Anyone have that part number? I'll look for it. test it..
a) In case belt's too short for easy installation, go back to original belt. test it...


I know that the people who tried the short belt said it was an extreemly tight fit and at least one was damaged trying to install it, you may be able to deepen the slots in the adjustable plates to allow ease of installation. I know the part # is in this thread somewhere but couldnt find it right away. I also think the hard edges of the pully teeth need to at least be smoothed over, I am trying to think of a way to make a tool to do this.
Click to expand...
Click to expand...

Thanks, Denisesewa. I just couldn't find the details on the short belt results. I'll skip that part now, for sure. Any suggestions for a screw (bolt) adjustable idler for belt drive? I'm positive that a constant tension idler would deliver better belt performance/life than the spring loaded type. By the way, hows the chain drive doing? You and a couple of other member have great looking, working chain drives. Bike on!

My new sheet should arrive soon. Thanks, LightningRods!
 
So far the chain drive is holding up very well even with the cheap chinese chain, I found a soure for Diamond chain that I think I'll replace it with.
I'm thinking there would be no need for an idler if using the adjustable sheets, I suppose if you really want one you could do something like I did for the chaindrive system and use a turnbuckle instead of the spring.

003-13.jpg
 
BTW, the big reason for going to chain for me is that unless the two belt pullys are perfectly aligned and absolutely perpedicular to each other ( this includes any idlers) the belt will "walk" to one side or the other and wear along its edges, there also has to be zero runout to use a static tension system or the belt will experience various tensions as it spins, that along with the poor tooth profile of the stock pullies leads me to believe its a ton more work to get it right than the chain mod as the chain will torerate some runout.
 
Denisesewa said:
BTW, the big reason for going to chain for me is that unless the two belt pullys are perfectly aligned and absolutely perpedicular to each other ( this includes any idlers) the belt will "walk" to one side or the other and wear along its edges, there also has to be zero runout to use a static tension system or the belt will experience various tensions as it spins, that along with the poor tooth profile of the stock pullies leads me to believe its a ton more work to get it right than the chain mod as the chain will torerate some runout.
Click to expand...

Good point. My hope is that the adjustable sheets (probably in transit now) will solve the existing alignment problem. My belt walks toward the bike frame (earlier photo), as is. The pulley profile is another matter. My hope there, is that removing the crazy plunge idler, will seriously help this primary drive. I'd like to salvage the belt system with these new sheets. Converting to chain looks like the best answer, though more costly and probably more challenging to the average consumer.

On the other hand, Denisesewa, your step-by-step instructions may coax some to give it a go! And personally, If those pulleys still significantly shorten belt life (after I go through my steps) you already know what my plan is. But I still believe the original design worked better than what eventually came out of the factory. Bike on!
 
Stock Primary Drive belt is a Gates PowerGrip HTD 475-5m-15. This is a 95 tooth belt, 475mm long, 15mm wide, with a 5mm pitch.
Short belt, thought to be used in original GNG test configuration, is a Gates HTD 450-5m-15. As GNG abandoned this short belt and reports from other members using it were negative, I've decided to skip testing it, it's been done already.

It may be that GNG had stripping problems with the short belt, went to the next length, and added the plunger style idler for increased tooth engagement, retaining the original sheets. In any case, with the adjustable sheets, I'll be able to test with and without an idler. Sheets are on the way. I believe Amazon and eBay, are good sources for the belt.
 
OK, so I am doing the second belt to chain drive mod with lightingrods adjustable plates, although there would be no real differance using the stock plates other than having to use a halflink, I will show the proccess step by step over the next couple days.
First I dissasembled the drive from the motor, this gives better space to work.

001-1.jpg


Then using channel lock pliers I bent the outer washer untill it comes off.

002-2.jpg


003-2.jpg


Using a new flat bastard file I set the motor speed to a relativly slow rpm and began to file against the rotation of the motor, keep the edge of the file parallel to the bearing surface and check your work often especialy after removing the belt teeth, your eye is a remarkably acurate tool and you will see any taper developing, adjust the file accordingly, after 11 minutes I am down to 12 mm and the motor has become warm so I am letting it cool while I write this. beyound 12mm the material will file off very fast so check your work often and take your time.
you can hear the rpm by clicking on this picture of my first conversion.


004-2.jpg


More to come.
 
After 5 more minutes of filing with the flat bastard file I am down to just over 11mm, at this point I switch to a mill bastard file and carefully check the shaft for taper with calipers, I see I have a slight taper with the shaft being bigger near the motor so I use a felt tip marker to blacken the entire shaft while its spinning this way I can see with each file stroke were I am taking metal off,


005-1.jpg


I keep doing this checking shaft diameter and taper every 10-15 seconds untill I am down to 10mm with no taper

009.jpg


I end up with a hand tight press fit with no wobble or runout and spent another 30 minutes doing it, total working time spent about 45 minutes.

006-1.jpg


Once everything is back together I will find the location for the sprocket on the shaft so I have perfect chain alignment, the sprocket will be tapped for set screws and the shaft will have flats filed in the appropriate places to recieve the setscrews.
Next step is the large pully modification.
 
OK, now for the big pulley. I reassembled the drive using Lightingrods adjustable plates.
First I drew a line thru the exsisting holes in the #25 65 tooth sprocket from TNNSCOOTERS, I then carefully measured to get a 10 cm bolt circle diameter, once satisfied I centerpunched the crosses in preperation for drilling.
011-1.jpg


012.jpg


now, using a 4 inch angle grinder with a thin cutoff wheel I cut the outer flange of the pulley off and ground and filed down any high spots checking to make sure the sprocket lays flat.

013.jpg


014.jpg


015.jpg


I made a temorary centering bushing from a pvc sink nut which I turned down the outer diameter to fit tight in the sprocket hole, I had to add a few wraps of electrical tape to the pulley flange to get a good fit.

016.jpg


017.jpg


018.jpg


019.jpg


I then secured a "ziptie" across the motor and used the tip as an indicator to check for vertical and lateral runout of the sprocket , keep moving things untill its as right as it can be making sure the sprocket is flat against the pulley.


022.jpg


I try to line up the original holes in the sprocket with the dimples in the pulley and once everything is correct I use a hotmelt glue gun to secure the two together, apply the glue everywhere there is access, once the glue cools I carefully remove the unit from the shaft and use a couple small "C-clamps" to further secure it while drilling the 8mm holes to bolt the two together.



025.jpg


More later
 
I have now drilled the new holes for the 8mm bolts, bolted the sprocket to the pulley and then removed the temporary centering bushing and the glue.

026.jpg


027.jpg


As can be seen, the chain rides very close to the pulley teeth, just as a precaution and in case debris, twigs, telephone poles ect. might get caught in between the chain and the pulley teeth I file the pulley teeth down a bit right next to the sprocket using the edge of the file.

031.jpg


029.jpg


And there it is, Thanks to lightingrods adjustable sheets we have perfect chain tension without the need for a tensioner, thats a big savings in amp draw and added performance at the cost of about 2 1/2 hours work and $30 worth of parts, ( plus the cost of a new file but that will get used for other things).

036.jpg


037.jpg


B'da , B'da , B'da , that all folks! :D
 
spinningmagnets said:
AWESOME pics! Thanks for posting them in the style of a tutorial. Now you're ready for 72V X 30A = 2100W?...
Click to expand...
I'm quite happy with the first bike at 54 volts and 38 amps peak , all I have to do to that one is install the adjustable sheets and get rid of the power robbing tensioner.
I hope this will help others not be afraid to tackle the conversion.
 
You must log in or register to reply here.

Similar threads

U
Lightning Rods Big block belt mid drive / No longer for sale
Replies
1
Views
597
Universalus
U
S
    • Like
Astro Flight motor kit - NEW PRICING
Replies
2
Views
556
Silver Fox
S
H
Drivetrain Help for an upcoming ToSeven DM-01L Build
Replies
13
Views
925
Heloc
H
A
Help Reprogramming Bafang mid drive BBS02B.
Replies
1
Views
296
docw009
docw009
R
    • Like
A critique of mid-drive riding recommendations (and hub drive criticisms)
Replies
15
Views
5,147
Bullfrog
B
Back
Top