• Don't miss out on all the fun! Register on our forums to post and have added features! Membership levels include a FREE membership tier.

Next build GEN 3 snowbike, home build project

swedenturbo

Well-known member
Lifetime Membership
Dec 4, 2009
330
397
63
Sweden
We have trough the years, designed built a few different snowbike kits.
A really fun hobby project. Most of all, they are awsome to ride!

We have no plans for serial production. These are only one off prototype builds.

For each year, we learn something new. This is an everlasting product development and test new solutions. It also brings riding your own snowbike to a new level as you always try to improve handling and performance.
Here is a link to our previous project build thread: (sorry that thread is messed up with missing photos, Look at later posts as pics are reposted)
http://snowest.com/forum/showthread.php?t=359272&referrerid=82502


031e354fdbf4acedddf6837e9e35b884.jpg

f0c35855a5d185cf8236684849f493c6.jpg

Our GEN 2 snowbike has come to the end of its development and we are excited to begin our next build.
In some aspects, we're trying to think outside the box with some new features.
We will keep our dedicated Gen2 turbo snowbikes. This build will be on our normaly wheel equipped "summer bikes", 2015 KTM 500.


We haven't come that far on the project yeat. No machining of actual parts but more focus is on design and make accurate drawings.
We hope to have it ready for testing in early/mid March.


10600533_10153450149977672_3247258166763468439_n_zpsupduzt7m.jpg

Starting the build thread here and post you some photos and thougts and ideas as we go.
Hope this will inspire more people to build their own custom snow bike
face-icon-small-wink.gif




Finally start start of the build!
The plan for the GEN 3 is a skid suspension similar to our previous build, the GEN 2.6. It will also have a sprung strut similar to the TSS.


It will not have a welded tubular frame but instead an assembly out of cut aluminum parts that are bolted together. The jackshaft will be moved forward slightly to provide a more compact design and less weight. The track will be a 121" Yeti Maxtrack.


The engine will this time be naturally aspirated but with a Thumper Racing 590 big bore. Why no turbo?
We are really pleased with the performance of the turbo but it's a lot of work to start building a turbo kit all over again. Maybe later, but we are eager to get this kit on snow this winter. BigBore is an easy install even if it doesn't even come close to the performance of a turbo.


As previous builds, we begin with development of the suspension geometry. An analog mockup has proven to be a necessity at such advanced simulations such as these
face-icon-small-wink.gif
.
We have seen that its crutial to get all of the attachment points in the right position to get maximum travel and keeping track tension.
Just by moving one point a fraction of an inch will completley ruin the suspention travel or mess up the track tension.


8368_10153454414657672_4137738287808654500_n_zpsgwovqret.jpg

From the mockup, we have come to the conclusion that we will run the same shock length front and rear 360 mm. The travel of the rear suspension is about 250 mm. Suspension strut will give an angle change on the tunnel section with a maximum of approx. 12.5 degrees, which provides an additional travel of approximately 150 mm.

Previous build had to have a sliding rear attachment point on the rail in order to keep the right track tension. By going with a shorter track, we have seen that we hopefully can go for a direct attachment with no sliding part. That simplifies the design and assembly.

The next step is to start making some drawings and produce a few more testmockups, mainly frame and linkage to make sure that everything will fit as planned before we take the next step.


We set the bike on makeshift supports to get the correct ride height. In that way we can take accurate measurements and angles.
1001078_10153454422577672_2104443674878303327_n_zps4ipkwv9g.jpg

Some simple sketches to ponder on before we get on ....
face-icon-small-wink.gif

10590551_10153454425727672_6037839613264298374_n_zps7dtf4hw8.jpg

One feature we will test is a suspension strut similar to the TSS. We are not sure that a pivoting kit is good in all conditions. In powder we think that a non pivoting will perform better. On a bumpy trail, additional travel is preferred.
The idea is to use a shock from a mtb bike with the function to lock it in two steps. In this way, we believe we have the best of both worlds: the soft and pliable sprung strut on trail and the best powder snow properties with rigid strut. The locking of the shock will be done by a wire operated lever on the handlebars so you can do it while riding.
You might think that this shock will be way too weak for this aplication but our idea is to use a lever. The lever ratio will limit travel but also reduce load on the shock. By being able to adjust the shocks attachment on the lever we can roughly adjust the hardness and then refine it with air pressure. The lever will have adjustable rubber endpoints in order not to put too much positive or negative load on the shock.
We are really excited to try this!
Daumlmpare1_zpsoh1qyrpi.jpg



Parts starts to arrive for the new build.

6-tooth 2.86 "drive wheels from Timbersled. The wheel in the background is 7-tooth that we run on our previous build.
Smaller driver wheels has both advantages and disadvantages.
They make more roll resistance as the track is forced around a tighter radius. Fewer teeth engaged with a higher risk of slipping.
Why we choose to test the smaller wheel is to maintain, or possibly even lift the drive shaft despite a track with higher lugs. The whole package is compact and there is limited space. Higher axle gives at the same time a steeper approach angle , which to some extent reduces deep snow performance but at the same time higher ground clearance. We believe that this overrides the disadvantages that smaller wheel does.
b89790ae05aa77df27d486ffb969da55.jpg

130 mm brake disc is smaller than previous.
The location of the brake disk on the jackshaft is the limiting factor in how much the suspension movement can be allowed of the tunnel until the brake disc hits the plastic parts on the rear of the motorcycle.
For that reason, we will mount the brake on the drive shaft instead, a more neater solution and safer in the case of chain failure.
We will in any case be forced to cut the lower part of the "airbox" in order to get any travel.
115bb5c7408a64d9801c12f45834c42e.jpg


Mino caliper to a rear brake of a KTM SX 85 cc. Should match well with the bike's OEM brake main brake cylinder. Earlier caliper had too large volume and we have been forced to switch to a larger cylinder in order to get it to work.


The track is a Maxtrack from Yeti. Looking good!
b2115ee0bc5e7ae8b4ae37f9b4c532e6.jpg



Stay tuned.....
face-icon-small-tongue.gif
 
Last edited:
M
Jan 14, 2004
3,079
1,390
113
I think you are on the right track with the cut aluminum frame members vs a welded cage, IMO it will give more consistent results when building multiple kits. The jury is still out for me on the need for the TSS at all, good geometry with minimal weight and simple design is what I personally am shopping for in a kit. Are you sticking with the cut aluminum plate sides for the tunnel?

Always enjoy your threads. Thanks for posting. If you post your drawings I just may build one here. LOL

M5
 
D

DValentine

New member
Nov 27, 2007
53
4
8
Laurel Montana
You guys are awesome. I very much enjoy reading about your production, trials, redesigns and what you have learned. Keep up the good work.
 

swedenturbo

Well-known member
Lifetime Membership
Dec 4, 2009
330
397
63
Sweden
Thanks guys.
Hope I will post some build photos soon but right now we're in the design process that we would like to share with you
face-icon-small-wink.gif


Are you sticking with the cut aluminum plate sides for the tunnel?
M5

We will continue to use a tunnel similar to previous build, but in a slightly different way.
We will try to use the tunnel as a load-bearing member. Snowmobiles uses the tunnel in that way, right?
With this idea, the frame can be made shorter and hopefully with less weight.

One more thing we have been looking at is different aluminum alloys.
We have previously used aluminum quality 6082.
Looking at the properties aluminum 7075, the yield strength is more than 80% stronger than 6082.
Not sure if these properties change dramatically in the lower temperature range.
So far, the only disadvantage with 7075 is that it has less corrosion resistance. This will not be any mayor issue as all parts will be powdercoated.

We will not be able to bend any of the aluminum parts as the 7075 is too stiff.
Instead, the tunnel will be made out of three parts. Angle aluminum profiles, pop rivits and heat resistant glue will be the way to assemble the tunnel prior powder coating.

All our ideas are hard to explain but hopefully this will be more obvious as soon we start to make som actual parts...
face-icon-small-wink.gif
 
Last edited:
M
Jan 14, 2004
3,079
1,390
113
I understand what you are saying. Most sled tunnels that I have built or seen are typically 5052 which is easy to work with but nowhere near as strong as a 6000 or 7000 series material. I've thought of doing a build where the tunnel is one piece and bent much like a sleds and the front is much like what you are planning. It would offer some advantages and also some disadvantages but basically wouldn't be much different than a sled and its bulkhead.

M5
 
S
Jan 27, 2012
27
2
3
SnoMoto Xtreme

Here's a pic of the one Im currently building using a bolt together frame if your looking for ideas. The center pieces have 1/2" Dia bosses that go a 1/8" into the 1/4" main frame so Im not relying on just the bolts. Also Im running the chain inside so theirs no chain case cover to deal with. Only down fault is the frame is a little wider. I did have the brake on the main drive on my prototype but moved it on top because of the smaller drivers Im running. Still going to run 2.52 pitch track 2" paddle 121" with 7 tooth driver 5.25" OD. Doing research all the top drag sleds still use the 2.52 pitch. The rear tunnel is going to be carbon fiber maybe carbon/colored Kevlar. The rear suspension is a M10 with a modified rear arm and lighter rear spring 10-3/4" wide track.

Main Frame Model.jpg
 
S
Jan 27, 2012
27
2
3
Here's a different approach to adding a third shock similar to the Honda uni link used on the sport bikes. There's a simulation video somewhere on the net.

honda-cbr-1000-rr-fireblade_rear_suspension.jpg 12Oct_Tech_ProLink_03_gallery_large.jpg
 
A
Jan 4, 2015
245
129
43
Canada
Here's a pic of the one Im currently building using a bolt together frame if your looking for ideas. The center pieces have 1/2" Dia bosses that go a 1/8" into the 1/4" main frame so Im not relying on just the bolts. Also Im running the chain inside so theirs no chain case cover to deal with. Only down fault is the frame is a little wider. I did have the brake on the main drive on my prototype but moved it on top because of the smaller drivers Im running. Still going to run 2.52 pitch track 2" paddle 121" with 7 tooth driver 5.25" OD. Doing research all the top drag sleds still use the 2.52 pitch. The rear tunnel is going to be carbon fiber maybe carbon/colored Kevlar. The rear suspension is a M10 with a modified rear arm and lighter rear spring 10-3/4" wide track.


You should stay away from the 2 bolts flange bearings. The cast one break and the sheet metal one bends. The sheetmetal 3 bolts are probably fine due to the more uniform loading.

The best solution is making billet bearings mounts or using billet mounts from Karts.
 
S
Jan 27, 2012
27
2
3
The reason for the stamped flanged bearing is if there's any misalignment these bearings help with that and if there's any flex in the shaft the bearings will move. I might go to the 3 bolt but I have the bore on the side panel within .005 of the OD of the flange so the load will transfer thru that and not the bolts. With have a fixed bearing in a billet housing causes a lot of load to the bearing causing premature wear. T-sled found this out that's why they went to the double row to handle the load. Also you can find the stamped flange bearings really easy if in a pinch also can carry a few with. The ones I'm using are double sealed Timken
 

swedenturbo

Well-known member
Lifetime Membership
Dec 4, 2009
330
397
63
Sweden
We have successfully used billet machined aluminum bearing housings on the drive chaft on our previous build. C3-rated bearings has wider tolerances that allows some flex.
We think adding rigidity with less flex is the trick to make this design to work.
Our solution is to make a really strong rim around the chaincase. This makes the latheral flex on the tunnel plates to a minimum. This has proved to work really well.
0be53bc483dad82fa6f6575e72892611.jpg


Currently there are two different solutions to tension the chain on the production kits. TS uses a tension ildler sprocket. Yeti and CMX adjusts tension by moving the whole kit rearwards. A really neat solution compared to TS. Tension sprockets like TS has a limited tension range that in some cases end up in some gearing issues.

A movable jackchaft is the third solution, a really nice idea to adjust chain tension. This keeps the kit in the correct position in relation to the bike regardless how the chain is tensioned.
The challenge with this design is to keep bearings alligned when the jackchaft is moved.
Currently we have two competing solutions in this topic.

My cobuilder Jonas made a billet housing that connects inner and outer bearings with the sprocket between. A complex housing to machine but uses regular cheap bearings.
296dcb2785be7c5aa2e2d714ae471a41.jpg

Our other solution uses two separate housings. Double row sperical roller bearings can handle misalignments easily. Roller bearings can take a lot more static and dynamic load than ball bearings but the downside is that the bearings themself are more expensive.
For the manufacurer of a production kit, this may be something to consider regarding manufacturing costs and at the end price.
b15153ea68fe8e8f73885532de87bc8b.jpg

a767e2a2d1689ff70c6bb44c6b57b4ef.jpg

1c68388bde9b12bf5b4d81875245df2e.jpg

548dba9d5e48c8ff8289a14b8f13eaed.jpg
 
Last edited:
M
Dec 29, 2007
64
11
8
Have you done any testing with the mountain bike shock yet? I was thinking about trying this out as well. Can pick up a fox air shock with 2.25" of travel for very cheap. My only concern is the available spring rate, max pressure is 300 psi. Does anyone know what the spring rates timbersled or yeti has used in their third shock setups?
 

swedenturbo

Well-known member
Lifetime Membership
Dec 4, 2009
330
397
63
Sweden
Have you done any testing with the mountain bike shock yet?
Not yet. This project is currently in the process of making all the drawings.
Not ideal time to start a new build in the middle of season. Most of our free time goes to having fun riding our current snowbikes.
face-icon-small-wink.gif



We think that the spring rate in this tiny shock will be sufficient thanks to the linkage that we will use.
By having a lever ratio 2:1 the load is cut in half.
Of course the final travel gets shorter but we dont think that super long travel on the strut is that important.


Here's a pic of the linkage that we will use.
As said, it will have a 2:1 ratio that will end up in 8 degrees of travel.


ca2f9328e2377744c9605d30c4c52ea0.jpg
 

swedenturbo

Well-known member
Lifetime Membership
Dec 4, 2009
330
397
63
Sweden
Now we're almost be done with all the drawings of our GEN 3.
Just left to scrutinize so everything is right.
Then we can start waterjet cutting all the parts.


Took some time to get all the geometries in place. Also double check clearances so no collisions on the bike, such as exhaust, etc.


We had at first a targeted 12 degree swing motion but realized that linkage ratio was 1.6 to 1. We believe this is slightly too little for such a small shock we´re planning to use.
Instead we chose the linkage ratio of 2:1 which ended up in 7.8 degrees of swing travel when using 55 mm stroke of the mtb-shock. The shock has actually 63 mm stroke, but we want to leave some safety margin.

On the drawing below you can get some idea in how short the frame will be. This is thanks to the new tunnel design that will be a stressed load carrying member.
We draw in a simple 2D software that works fine but not as fancy or comprahensive like a 3D solidworks



gen33_zpsifle8etl.png
 

swedenturbo

Well-known member
Lifetime Membership
Dec 4, 2009
330
397
63
Sweden
Begun the work of machining the drive shaft.
This one will not be out of steel, but aluminum.
Since regular keyways seems a less reliable solution on a aluminum shaft we choose flanged mounted sprockets.
The aluminum quality is 7075 T6

0b20d4fcbd2fd7e6e9664094fef82c85.jpg

d9ff3264927b9a5826a723f34a87f9db.jpg

Have acquired a better lathe. Previous lathe worked okay but felt a little too weak. The new one is much stronger. Weighs approx. 2.5 tons. The new drive shaft is based on a 60 mm aluminum stock. Previous lathe had 38 mm passage. The new features 80 mm passage which in some way simplifies chucking. Feels really strong and solid.
15cfbb4e6059ba7f981a371f03377891.jpg

The drive shaft done. Just left to drill and tap bolt holes for the flange mounted sprocket.
d9d6fd9764a643e3295d691171b3b0dd.jpg

Milling hexagonal part with a circular feed table and tailstock.
d38cc10045dc8a15240bbe9d370e4313.jpg

A lot of material have been machined if you compare to the stock that I started with.
ae8b2c80d1efb91a34821d55a6b70a6d.jpg
 
A
Jan 4, 2015
245
129
43
Canada
The reason for the stamped flanged bearing is if there's any misalignment these bearings help with that and if there's any flex in the shaft the bearings will move. I might go to the 3 bolt but I have the bore on the side panel within .005 of the OD of the flange so the load will transfer thru that and not the bolts. With have a fixed bearing in a billet housing causes a lot of load to the bearing causing premature wear. T-sled found this out that's why they went to the double row to handle the load. Also you can find the stamped flange bearings really easy if in a pinch also can carry a few with. The ones I'm using are double sealed Timken

I build one with the 2 bolts flange and they bend. I will build a billet housing but will keep the spherical bearing used in the 2 bolt flange.
 

swedenturbo

Well-known member
Lifetime Membership
Dec 4, 2009
330
397
63
Sweden
Test assembling the bearing housings on the aluminum drive shaft.
3943cf234cc549232f1ff85f5cb8460e.jpg

e7587a73cba14540f9ade3bc17c492f5.jpg

2abaca0e7624e1d84380b50afb1924be.jpg

Bogie rails to GEN3.
This time we will run a direct attachment without any sliding mount in the rear.
We saw in our mockup that we will keep track tension on a short track.
468b291d97db319da58f16490c1741d8.jpg



Chamfer edges mostly for the look.
6c1aac20177df3e3eefddaef8b0b1211.jpg

3ceeac372d5821e4fde7fde83bed3436.jpg

We have chosen one of the strongest Mtbdownhillshock we managed to find.
7ddc3be781d462c4a7d83520fba0eae4.jpg

The shock has a lock feature. We believe that a locked shock provides better off trail performance and unlocked the better on trail.
This control lever will be mounted on the handlebars which operates the lock on the damper.
451bc3df884ed56151bbb7437acbba2c.jpg
 
Last edited:

swedenturbo

Well-known member
Lifetime Membership
Dec 4, 2009
330
397
63
Sweden
New redesigned suspension arm. The previous GEN2 arm was hourglass shaped and run inside the bogie wheels. Unfortunately, the new shocks would not fit because the side-mounted EVOL camber.
Therefore we choose instead to let tubes run straight back, just outside the bogie wheels. Becoming both an easier and cleaner design.
320c5ea4939c23518607fa19bd710d7b.jpg

The previous welding jig was used after some modifications.


7c658aa83f4ee84412fffe49d03fa814.jpg

Notching tubes in the milling machine. Very good fit and precision.
1368627d583a707efad22ed8b4b99771.jpg

42a74c33d81eff6dbb8c2202ab7cf1c0.jpg

Bogie rails are completed.
639149e2b33a769131f9b027587f82a6.jpg

We test a new solution to fasten the lower rail. Instead of milling a slot into rail and a T-profile, we ordered the watercut rails with a guide pin and predrilled holes.
Self-tapping countersunk screws, as well as two-component adhesive. This glue is designed to withstand the heat when curing powder coating.
Turned out to work out really well. Relatively quick and easy to assemble.
b094c64bca0e43f3edc76a2944f9116f.jpg

43f4dbc11750392af42cf08e92525785.jpg

9a38beca052a45930b1417c1d158c04c.jpg

ab1dd36d3e33031dfbc4c7853c21a239.jpg

606fee0e19bf70759d5b787f0c79cffc.jpg





Drive shaft pre-assembled with the brake disc and hub.
5965f53b7b95b89502b115746b41d448.jpg

bac69f3a523aebc311f97a529c406ad5.jpg

82ef19ad8615e35e5694d703be2714f6.jpg
 
Last edited:
Premium Features