ORGS build up

[DRjoe]

I've welded solid drive shafts together on outboard motors before by cutting both ends at an angle (about 60), clamp both pieces in a bit of angle to align them and then weld.

but my so called engineering is a bit rough so best not to do what i do

I've also seen crankshafts extended by putting a matching taper on both pieces then pressed together and welded.
I like this way because in theory the taper should take the load.

[Zebedee]

Fantastic stuff …

Many thanks for the continued updates …



John

[x3300]

I've been down riding the black and yellow R100GS in the Mexican Sierra Madre mountains for the past two weeks so haven't been able to make any progress on the build-up.

I did about 1000 miles (1600 km) of dirt riding that really gave me the experiences and the time to think again about the limitations of the R100GS and what I want in my ORGS. Here's a clip from an easy section that shows what I want this bike for.



fishkens, it is much easier to make a pin accurately fit a hole than it is to make a hole accurately fit a pin, and that is why I didn't use a standard dowel pin, but even drilling the holes and turning the insert ends down to fit was too involved. Also, the void in the holes not taken up by the insert ends can trap air and machining oil which can expand and cause weld contamination and/or weld inclusions. I think holding the parts externally with a v-block fixture is easy and is as accurate as holding them with an internal pin, and the entire cross section can be welded with a butt weld setup.

Airhead Wrangler, friction welding takes a lot of force and a lot of power, it needs some specialized equipment to get the pieces melted then to hold them together until they solidify. Just for anyone interested, check out friction stir welding, a similar technology.

DRjoe, my welder has a pulsed DC feature that I can set a high pulse frequency to get a relatively narrow and deep penetration without having a big chamfer. I found a narrow notch at the joint seems to work OK, and with this method I don't need to add a lot of filler.

-x3300

[fishkens]

Originally Posted by x3300

fishkens, it is much easier to make a pin accurately fit a hole than it is to make a hole accurately fit a pin, and that is why I didn't use a standard dowel pin, but even drilling the holes and turning the insert ends down to fit was too involved. Also, the void in the holes not taken up by the insert ends can trap air and machining oil which can expand and cause weld contamination and/or weld inclusions. I think holding the parts externally with a v-block fixture is easy and is as accurate as holding them with an internal pin, and the entire cross section can be welded with a butt weld setup.

Okay. I'm familiar with drilling a hole and then reaming it out to size to fit a standard dowel as opposed to machining the pin. But I barely know enough to be dangerous and understand how machining the pin could be easier.

Thanks.

Looking forward to more.

[Airhead Wrangler]

Originally Posted by x3300

Airhead Wrangler, ... check out friction stir welding

I like. Hadn't seen that one before.

[rediRrakaD]

^ +1.
S.

[petekeys]

Hi Geoff

Now you are set up with jogs and everything, are you going to offer this as a service?
If so then how much would you charge?
cheers

-Pete

[petekeys]

Originally Posted by Stagehand

wow that is pretty cool-

and yeah, agreed, this is great work. Fun to watch, too.

Is it possible to do this in a conventional lathe? There must be a fair amount of pressure being applied between the two shaft so maybe its not doable in a lathe.

[x3300]

petekeys, this project is about realizing my vision, about creating the machine and executing my plan, and also of course about telling the story. Once I create something I really have little interest in doing it again for someone else, unless maybe if it allows me to be involved in some cool project or enterprise. I don't want to get into a discussion about it, but if someone wants to do such a service, all the photos, drawings, and documents from the project are my own and I am the sole copyright holder, but I release them to the public under the terms of the Fabricators Design License. The spirit of the license is to allow anyone to use the material for whatever use they want as long as any modifications or improments to the original design are made available to the public at no cost.

As I mentioned in a previous post, the u-joint of the shaft I extended was going bad so I intended to replace it. I measured the OE joint as 19x44. Here's the joint in the original shaft.



After some searching I found two after market u-joints I thought might work. One is a 19x44 and the other a 19x48. From the left; the 19x44, the 19x48, and the OE joints.



I was interested in the 19x44 because there's not much clearance between the transmission output flange yoke and the swingarm housing when the swingarm is in the bottomed-out position with my 280mm of wheel travel. I thought with this joint there may be enough room to add a circlip to the yoke to hold the joint in place, but when I got the part I found it to be of very poor quality, and I think intended for steering shaft applications. As seen in the photo the bearing inner shaft has a very small diameter and the needle bearings are much shorter then the OE joint.

Here's a comparison of the OE joint on the left and the 19x48 on the right. The OE joint has larger diameter bearing shafts and longer needle bearings than the replacement 19x48. I think the OE joint would have longer service life in the monolever's splash lube application, but the 19x48 should be sufficient for my application. I am still on the lookout for a better replacement, ideally something near identical to the OE joint.



The bigger 19x48 didn't quite fit into the yokes.



I used a die grinder to take a small amount of material off the yokes and the joint cross.



Here are the shaft parts ready for assembly. The photo shows where I ground two ribs off the joint cross to get the needed clearance for assembly. The replacement joint has a grease fitting and shaft seals, but I didn't install these to allow the gear oil in the swingarm to get to the joint bearings.



The 1st step in assembly is to get the cross in the yokes.



Next is to install the caps. I used a brass hammer and some sockets to get caps in place. The photo shows how I used a socket large enough to pass the cap to support the yoke from below. Some light tapping was enough to get the caps positioned.



I used a depth micrometer to center the caps in the yoke such that there was no play in the joint bearings.



As mentioned, with my increased wheel travel there isn't much clearance between the transmission output flange yoke and the swingarm housing at the bottom-out limit, and with this bigger joint the options for fastening the joint were few. My 1st idea was to make a strap, maybe 5mm wide that would go across the cap and be spot welded to the yokes, but in the end I decided a simpler solution was just to spot weld the cap to the yoke.

I knew that it was somewhat common method, especially in off-road trucks, but wasn't sure if it would work here. To get a better feel for it I cut up a cap from an old joint with an abrasive cut-off tool to see the cross section.



I figured that if I did the welds with minimal penetration there was a lot of material there at the end to take the weld, and also, all the u-joints I've ever seen always wear out at the load bearing sides of the needle bearings, so if I put the weld somewhat perpendicular to that the effect of the weld would be minimized. I setup the welder with a low amperage and the DC pulser at around 5 Hz. I just put one weld at the thicker end of the yoke. I can try putting a second weld at the other side if I find trouble when I put it into use.



There wasn't much else to do other than weld the other caps on. Here's the finished shaft.



And the shaft installed on the bike. This shows the yoke clearance problem. When the shaft turns a few degrees from this position the corner of the yoke will almost touch the arm housing.



I have a spare joint, so when it comes time to replace this joint I'll just need to grind away the spot welds to get the caps out and put in the new joint.

-x3300

[Beemerguru]

Great project and wonderful workmanship.

I'm just starting another G/S..this time going retro with a dual shock version of the original late '70s G/S. Of course the frame needs some bracing and wondered if you had any left over finished gussets or if the vector graphic data is still available?

Thanks

Greg in Foster City


one

Originally Posted by x3300

Hi everyone,

I have really enjoyed the stories of other airhead bike build up projects, so will be posting mine, the build up of ORGS my Off-Road GS here.

I have a good general idea of what I want, but you'll have to follow along as the plan in my head unfolds. My intention is for ORGS to be a bike I can take on the highway and ride all day long for several days to get to a destination, say Moab, Baja, or Copper Canyon, spend a week or so doing some hard riding on unmaintained back roads, then make it back. I want it to be able to take a wipeout without a lot of damage, or cost to repair. That was a drawback of my GSPD. Windscreens, body panels, handle bars, etc. break and cost a lot.

As a base to start with I have bought this '92 GS with 27,000 miles:



And I got this run down '92 PD with 52,000 for parts:



The first thing I'll work on is to get the frame in order. I took a day to strip the parts bike down. Here is what is left that I'll start to craft into ORGS. It looks like a chicken to me:



I studied some photos of the frame upgrades done by various projects, and came up with these paper patterns, then cut them from metal I bought at the scrap yard:



It took a lot of work to get those gussets cut out using a pneumatic cutoff tools.



I made a digital scan of the finished gussets and I'll make some vector graphic data from them which can then be used by anyone interested to program a CAD controled plasma cutter. What took me several days will be able to be done in a few minutes.

-x3300

[maverick]

Just another WOW and thanks for sharing the info. Simple question is the swingarm extention tool you welded up just based on a 90 degree angle when it extends out to the longer sections or at a slightly different angle as the shaft is not entirely centre to start with on the original swingarm if that makes any sense?

[x3300]

maverick, I assume everything is aligned; trans output shaft, final drive input shaft, swingarm pivot, etc. I didn't check that, but I can see my extended shaft and arm seem to have the proper alignment.

Beemerguru and others, I have received a lot of inquiries about the gussets so I created a new thread dedicated to them here: http://www.advrider.com/forums/showthread.php?t=638795

To get the rear wheel travel I wanted with my extended swingarm I either needed to fit a custom shock to work with the original GS upper shock mount or make a new shock mount to fit a shock from some other bike.

Modern off-road suspensions have progressive damping, where the damping increases as the shock compresses. This is done either through a linkage mechanism or through shock valving. Such systems have been well received, so I figured I'd try to get some progressiveness in my design. Since the GS has no linkage I would need to fit a shock with progressive damping. Not too many bikes use such a shock and this requirement would really limit the shocks I could choose from.

As a start to get an idea of what shock length and stroke values would work for my application I made a spreadsheet that calculated the shock length and stroke needed to get my 280 mm of travel using the original GS upper shock mount. I set it up to do the geometric calculations at 2 mm increments along the swingarm.



I used the values to compare what would work with what was available from various shock catalogs. I wasn't so lucky, no stock shock would work with the original GS mount, and the only progressive shocks that seemed like they might work were from a BMW G 450X, or a few of the KTMs. I figured KTMs are pretty common compared to a BMW G 450X, so decided on fitting a KTM shock. After a bit of searching around the Internet this box arrived with an Ohlins TTX44 for a KTM 250 SX-F. The length is advertised as 410.5 mm and stroke 109 mm.



With the shock in hand I could estimate how much the stroke was decreased by the bottoming bumper and how much space was needed between the lower shock mount and the swing arm centerline. Both these need to be considered in the design.

The geometric equations that describe the set of mounting points that would work with this shock are non-linear and difficult to solve for even using numeric methods, so I decided to use a graphical method. I used a large piece of paper to layout a full scale drawing with the swingarm pivot, upper shock mount, and swingarm at the upper and lower limits of travel. I used two thick pieces of paper marked at the extended and compressed lengths of 410.5 and 306.5 (5.0 mm bumper thickness) to find a set of mount points that would work.



Here's the detail near the top mount. As the drawing shows, any lower mount from about 240 mm to 320 mm rear of the swingarm pivot won't work with this shock because the frame tube is in the way of the top mount. The points from around 320 mm on back could give the proper wheel travel but would put the upper mount in the middle of the frame tube and I thought that that would put a lot of bending load on the tube so I chose not to use those points. Around 230 mm seemed good. This would put the shock and mount up above the original shock mount, but I thought it would be OK since I needed to make a custom sub-frame and seat anyway.



Once I got some mount points I thought would work I laid those out on the frame and swingarm with some masking tape to do an initial check of the fitting.



When I was confident I didn't need the upper mount I used a cutoff wheel and angle grinder to remove it, then laid out some mount points on card stock tapped to the frame.



And checked what would give enough clearance between shock and frame.



To accurately check the fitting with the swingarm in the up and down positions I made this jig with holes drilled at 410.5 mm and 306.5 mm.





Then, after all the checking I made this paper template for the upper mount.





Here are the upper and lower mounts I made up. I thought it would be hard to do the bends a one piece upper mount would need so I decided to make it from two pieces that would weld together.



First I welded on the outer piece.



And then the inner.



Here's a view from the inside with the shock in place.



With the upper mount welded on I used the jig to mark arcs at the extremes of swingarm angle at my chosen lower mount point (230 mm from the swingarm pivot). The red shows the upper limit that would not allow the transmission yoke to contact the swingarm housing when the suspension bottomed out, and the green the lower limit that would keep the transmission output flange bolts from tearing up the rubber swingarm boot when the suspension topped out. I would have liked a little more margin for safety, but had to accept what I was faced with. The black cross shows my drill point.



The lower mount would need a spherical rod end bearing to match the shock's lower clevis mount. Some searching in the McMaster-Carr catalog showed two bearings that could work, a 12x30x16 metric bearing and a 1/2"x1"x1/2" inch bearing. I decided on the 1/2"x1" bearing since it was cheaper and would allow a lower profile mount. I made up this mount from some rectangular tubing and round stock.



Here are the internal parts of the lower mount. I bored out the round stock to allow a press fit of the bearing and turned down the edges of some aluminum stock to make two spacers that allow the joint's ball to rotate a few degrees.



Here are the pieces assembled and the joint set at its limit.



And the mount tack welded to swingarm. The shock is installed for checking. I'm thinking to make some short tube shaped seals from neoprene sheet to keep dirt and water out of the bearing.



Not much was left to do other than to finish the lower mount welds. Here's a view from the rear of the finished installation. The reservoir at 45 degrees sticks out a little, but I'll make the sub-frame go around this as needed.



And a view from the side.



I'm really happy with the result. 280 mm (11") of travel and a first class shock. The hollow lower mount gussets didn't work out as well as I thought they would. The rectangle tube wall was relatively thin and difficult to weld to the thicker swingarm. I think some gussets from 1/16" or a little thicker sheet, maybe with some drilled holes would be better.

-x3300

[Mr. Vintage]

Awesome, as usual.

[x3300]

The R100GS takes a huge battery, so I figured I could save some weight and make more room for a bigger tool tray if I fitted a smaller one, with the expectation that I would replace it more often to keep it up to peak capacity.

I looked through the Yuasa battery catalog and found the YTX14AHL-BS which seemed like it might be enough. Its used on some other big displacement bikes. Here's how the specs compare to the 53030:

model        AH  CCA  acid    weight  L   W   H
53030        30  ?    1600ml  7.3kg   186 130 171
YTX14AHL-BS  14  210   660ml  4.1kg   134  89 166

With a little searching I found one advertised cheap locally so picked it up to try it out. Here's how it compares to the original GS battery.



To start I made up a paper model to find the best fit in the frame and get an idea of styling that would look good.



Here's a slightly revised version of the battery tray drawing for the unfolded sheet metal. As seen in the photos following my original design didn't have the full front panel this drawing has.



I used some 16 gage aluminum sheet I had that was left over from the dash, and after a quick layout with a Sharpie marker I used this holesaw and drill press to make the big side holes. These holes were bigger in my original design, and I found they removed too much material which left the sides a bit too flimsy.



I could make the smaller holes with this rotex sheet metal punch.



Here's the plate with the main holes. I don't have a photo of it, but next I punched four very small holes at the intersection of the fold lines then used a corner notcher to cut out the corners. The holes help to make a smooth bend.



I used a sheet metal brake to bend the panels up. Also shown are the mounts I cut from aluminum angle. To be safe I made the mounts a bit longer than I measured was needed with the intension of trimming them down after welding and fitting.



Here's the finished tray with the corners and mounts welded, and mount holes drilled.



The old and new parts compared.



Here's how the tray fits in the frame. It leaves a lot of room for a big tool tray, exhaust pipes, and muffler. I still need to clean up the mount ends and add a hold down strap.



And a view with the battery. This photo doesn't really show it, but the contrast between the dark battery and the shiny aluminum makes the strip and five holes look really good.



As mentioned, this first version is a little flimsy, I think the new version of the drawing will be enough in 16 gage aluminum, maybe with a slightly stronger attachment of the mounts to the tray panels.

-x3300

[x3300]

I'll need some kind of spring to hold up the back of the bike. As I mentioned in an earlier post modern off-road bikes have progressive suspension systems where the damping force and spring rate increase as the shock compresses. Most bikes use a linkage mechanism to achieve this, but some have a linkless design where the progressive effect is accomplished with progressive shock valving and progressive rate springs.

Here's a nice spring force diagram from Race Tech:



With some searching I found two progressive springs on the market that I thought would fit my TTX44 shock. One is the Race Tech 6326 Series.



The other is the Langston Racing Super Progressive.



Just based on these two photos the Langston spring seems to have a much more progressive wind than the Race Tech.

Here's a comparison of the recommended springs for a 250 SX-F from Ohlins, Race Tech and Langston. The stock spring rate for the 250 SX-F is 62 N/mm.

rider weight  Ohlins       Race Tech  Langston
kg (lb)       N/mm (lb/in) N/mm

64 (140)     -            6326P05
69 (152)     -            -          LRS-01 64-132
70 (154)     60 (343)     -
75 (165)*    62 (354)     -
76 (167)     -            6326P10
80 (176)     64 (365)     -
85 (187)     66 (377)     -
90 (198)     68 (388)     6326P15
95 (209)     70 (400)
105 (230)    -            6326P20
109 (240)    -            -          LRS-02 83-176
120 (265)    -            6326P25
136 (300)    -            6326P30

To get a handle on which of these would work with my custom rear suspension I considered how the ORGS rear compares with some linkless KTMs.

bike            style    travel(mm) ratio  weight(kg) rate(N/mm)
KTM 250 SX-F    MX       335        3.07   100        62
KTM 450 SX-F    MX       335        3.07   105        68
ORGS            trail    280        2.57   200        -

This bike will be much heavier than a 250cc MX bike, but it will mainly be used on trails, and the shock leverage is about 80% of the KTMs. I thought I'd rather have the back end a little too soft than too stiff, so I decided on the Race Tech 6326P10. The LRS-01 was another choice, but I thought it might be too stiff near bottoming. I put in an order and this arrived.



Here's what I measured it to be:

Race Tech 6326P10
length	260mm   10.35"
ID	63mm     2.48"
wire	13.0mm   0.51"

Pretty close to 2.5" x 10", a common size of off-road racing truck coil-over springs. I've never seen any progressively wound truck springs though.

To get the spring on the shock I needed to make a spring compressor. Here's an updated version of my ohlins spring plate drawing.



The design is similar to my drive shaft spring compressor. A base plate has two arms welded to it.



And two bolts are welded to the arms that pass through the plate.



Here's the unit in action. I made the arms long enough to allow a block of plastic at the bottom to keep the shock from getting scratched.



This photo shows how the 36.5mm radius cut-out allows the spring clip to be placed in position while the spring is compressed.



And the shock with spring installed. I noticed that the spring hits the corner of the battery. I thought that it might after I got the new tray done. There's enough room to move the battery to the left a little. I just need to drill another set of mounting holes in the tray.



With the shock installed I checked the free sag and ride height. For the 250 SX-F Ohlins recommends 30mm and 110-115mm for those. Converting by 109/335 gives 9.7mm and 35.8-37.4mm at the shock shaft. I measured 10mm and 28mm, which would seem a little too stiff, but I think it will be OK with the heavy bike, and the next lighter spring in the 6326 series is considerably lighter. After I get some trail riding time on the bike I'll be able to judge how well this spring works. I guess I'll need a stiffer spring while carrying traveling gear.

I spent a lot of time researching springs, studying the data, and writing up the report, but the rear suspension is a big part of what this bike is and so I wanted to give selection of the spring proper coverage.

-x3300