Repowering a KTM 640 Adventure

Discussion in 'Some Assembly Required' started by Z50R, Oct 19, 2017.

  1. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    Why?

    Why:



    I have (had?) a 2006 KTM 640 Adventure. I like the bike's concept but I think KTM missed the mark on a few things.


    The good parts of the 640 ADV:

    Carburetor: you can fix it trailside and it needs no battery to run

    7.1 gallon tank: you only run out of fuel if you spring a leak

    Dual disk front end: if one gets damaged, you can remove it and still ride. Until that happens, the bike stops NOW and doesn't wear through pads quickly doing so

    Really nice suspension: The fork does its job well without any love from me except yearly oil changes. The rear shock is rebuildable and gets love every two years.

    520 chain: I don't know who speced a 525 chain on a DR650 but they ought to be shot. You can find 520 chain in a decent hardware store, a 525 isn't in stock in large moto dealerships.

    Kick start: you aren't having an adventure until your battery goes flat. Be it from waking up to ice on the ground and your battery is too cold to get out of bed or maybe from sucking water at a crossing that is just a bit deeper than a good idea, all the good rides require a kick starter at some point. I got good at push starting a DR650 on loose surfaces but enough is enough.

    Cush drive: if you like your engine, damp out the feedback. Honestly most of my riding is commuting. Pavement and no cush drive leads to chewed up output shafts. Even honda 90's have cush drive, why is this a debate for a bike with a license plate?


    The bad parts about the 640 ADV:

    The LC4 engine: it shakes like a paint shaker. I used to think people were being whiney bastards when they complained about this but it turns out, this is bad for the engine too. The shaking causes all the o-rings on the bike to fatigue. Buna N o-rings last about 5,000 miles and Viton last about 7,000. After that, the o-ring leaks. This is fine if you don't mind a drop or two of oil all over your everything but the real problem is the clutch slave cylinder seal fails like clockwork. I changed that o-ring with every oil change and still worried I'd loose my clutch. The engine also apparently has a habit of eating the intake lifter... yah... I got the memo late on that one.

    BST40 carburetor: I get that the bike needs to pass emissions but... yah...

    The low fender was gone on mine before I got it.

    You still need at least a little battery for the bike to run. The CDI box is attached to the main circuit so a completely dead battery is an issue as far as getting a spark. I have read that disconnecting the headlight lightbulb would allow the bike to fire off but despite kickstarting the bike more than electric starting it, I could never kick it over with a completely dead battery.


    Things that annoy me about the bike:

    The rally fairing: I get it, some people are wanna-be Dakar riders. And some people ride their offroad bikes on the highway only. But my happy place is knee deep in a water crossing or on single track that hasn't been ridden yet this year. I want visibility and the fairing is in the way.

    The headlights: These things look cool but don't work for anything. They are body mounted rather than fork mounted so they work better for going straight but don't work well for turning in the dark. They are brighter than stock DR650 lights but still not good.



    So short list of shortcomings with the KTM but replacing the engine is not something that one does on a whim. even with the obnoxious O-ring issue, I had grand plans to follow the ridiculous maintenance schedule this bike demands and keep riding the bike as it was. If I had known about the likelyhood and sevarity of the intake lifter problem, I'd have replaced it as preventative maintenance. In fact, I had just done the water pump about 500 miles prior to engine failure.

    However,

    Once the engine ate its self, I felt like I had finally exhaled after holding my breath for 6 months. This is an opportunity to banish that LC4 and I intend to seize it.
    #1
  2. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    My thought process is "I need to do an engine swap". A bit of research finds that several people have asked about engine swaps but I could find no evidence that anyone had actually done one. The main reason why is KTM's right hand chain drive. To do a swap, one needs to either find a donor engine that runs the chain in the wrong side like the LC4 or one must sort out how to re-engineer the bike to run with a chain on the opposite side. The engine issue could be resolved with a used BMW x challange's rotax engine but I hate BMW with a passion. Other right side chain drive engines are rare and obscure. A chain drive side swap is also non trivial since the rear brake would need to be replaced and the swingarm modified to make this possible. These hurdles could be overcome but they require money with little return on the investment. If no one has done it, yet LC4 engine parts are cheap on eBay (meaning no one else thinks it is a good idea to rebuild these engines either), then we are back to square 1: what to do about my broken bike.

    Well if I sell the mess for what I can get and had cash, what bike would I buy: No one makes a bike like the 640 adventure. (Not even KTM for reasons I don't understand) One can spend money and try to make one similar to the 640 ADV though. So if I were to do so, where would I start?

    I would want a mass produced engine with a cult following for reliability. This means Japanese. I ride single track and highway on the same machine. That means I need to go 80 and lighter is better. 80 requires about 35 hp. So now I am looking at 650 machines. That means KLR, DR650, or XR650L. I throw in the XR650R because they are available with a license plate but pull them back out of the mix because they cost more than I can spend right now. I pull the DR650 out because it can't have a Kick starter added. I add in a DRZ400 when I realize they have the power to do what I want and I decide to pursue the idea when I realize that one can choose carefully and end up with a factory FCR carburetor.

    My next thought is, I need to sell the 640 to pay for parts to make a 640esque DRZ. Wait a sec... I'm going to sell a gas tank to buy a gas tank? And where am I going to find a dual disk front end except the one I have...
    What if I do a fork swap and sell the DRZ fork instead...

    My thought then is, how far could I take this process; what is the farthest it is reasonable to push this idea? If i keep the 640 tank, I'd need to make tank mounts. The front tank mount unbolts from the 640 frame center down tube. The DRZ has a center down tube so it can't be too hard to weld on two tabs to bolt the mount to. The rear tank mount is on the sub frame. If I keep the 640 sub frame then I can keep the side covers, seat, and rack too. The DRZ has a bolt on sub frame. I am sure that the DRZ bolt points don't line up but that can be accounted for by modifying the KTM sub frame. The DRZ doesn't have a cush drive. That is so rediculous that I'll say it again: THE DRZ HAS NO CUSH DRIVE! What was suzuki thinking?! Ok well the KTM has one, can I swap the rear wheel in? Same size axle? Check. Same pitch chain? Check. So last question, disk diameter... you guessed it, same size. That means at most, I have to fabricate spacers. What about the fork?

    Fork swaps from KTM to anything else aren't common because the KTM stem is made for odd sized bearings. KTM speced 29mm id, 50.25 od, 15mm high bearings. Now it is KTM's turn, KTM, what were you thinking? Where did you come up with 29mm? We aren't going to say that other number again because it is so odd that it must have shit on it; you pulled it out of your ass.

    Suzuki speced a more reasonable 30x52x16 bearing on the DRZ. Ok so lets find a 29x52x16 bearing... and yahrightno. Ok howsabout 29x52x15... no there too. 29x52xanything??? Not in tapered bearings, no. So no matter what, a fork swap will not be a bolt on part. That doesn't kill the deal though.

    What would kill this deal? Well if I do this, I need to get money out of the KTM to pay for a DRZ400. So what don't I need and what can I get for it. I don't need a broken engine but no one else does either, maybe $250 as is. I don't need a frame and the title goes with it so that is at least $200. Carb, fairing, and maybe some small stuff for small money. This isn't going well. A friend pipes in. He knows a guy who needs an LC4 frame. Great! Tell him $500. If he takes the bait, I'll spend the time parting this mess out.

    Sold!

    Ok lets make this work.

    I got just shy of $1000 for parts I didn't need.

    And I went DRZ hunting.
    #2
    Te Hopo and invisa-bill like this.
  3. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    So I started DRZ hunting while stockpiling money. My better half says take money and go buy a bike. I say thanks but we really shouldn't spend that kind of money right now. She insists that I need a bike if she's gonna live with me. So she insists that we can part with $3,000 and the money will be free in about a month. Plenty of time to sell parts and get a good feel for the market.

    DRZ research yields the better DRZ is the e model by about 25% horsepower. This is further improved if you get one prior to 2007 that came with an FCR carburetor. The drawback is none came with a license plate meaning any that currently have a license plate likely have a hacked up wiring harness. While reading and watching bikes sell on craigslist, I start scrubbing the dirt off KTM parts.

    IMG_0299.JPG What was a motorcycle becomes a heap of crap.

    Meanwhile DRZ research suggests that it is possible to bore out to 440cc and stroke that to 460. Rear wheel hp (the only kind that actually matters) goes from about 39 stock to a max of 60. I figure I might get lucky on the used market and find something with around 45hp. The other thing to look for is supermoto wheels. Often that means the bike will have two sets of wheels and in my case both could be sold. In any case, supermoto wheels have a market that stock wheels do not.

    I waited my time and found my bike. This: IMG_0333.JPG
    2002 DRZ400e cost me $3100 and came with a 440 bore and Excel supermoto wheels.
    #3
  4. sonic reducer

    sonic reducer Been here awhile

    Joined:
    Aug 19, 2007
    Oddometer:
    460
    Location:
    SeaTac WA
    Look at dr350se/dr650 rear wheels for Cush drive on a DRZ. Not a total bolt on but not far off.
    #4
  5. tHEtREV

    tHEtREV Encouragement award recipient. tEAM iDIOT.

    Joined:
    Apr 14, 2007
    Oddometer:
    10,802
    Location:
    Middle Park, Brisbane, Australia
    Sounds like a hell of a lot of time and effort when a Safari tank and suspension work on the DRZ will give much more practical results with a lot less headaches and I would imagine cost.
    #5
    2 SPOT, airheadDan and Yellow Pig like this.
  6. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    Your prospective is "someone who doesn't own a broken 640 Adventure" where my prospective is "someone who owns a dialed bike with an engine problem"

    Yes, if I were starting with nothing, it would not make sense to start by buying a broken motorcycle for use as parts on a completely different bike.

    I already own all the "farkles" I'd want on a new bike. And all the farkles are factory designed to work well together. The fork is setup for the weight of my fat ass and 7 gallons of gas and a bunch of camping gear. Net cost to set up a KTM fork: $0. Net cost to set up a DRZ fork for that weight: >$0.

    If I wanted a dual disk front end on a DRZ (and I do), how much would I pay for that feature if I chose to work with the stock fork?

    If I buy a DRZ safari tank, I end up with a malfitting tank that looks like an afterthought on the bike...because it is! If I work with the 640 tank then I get a tank that flows into the side plastics, works well with the seat pan, and the tank will take paint without peeling.

    Your suggested approach may be easier. It will not be cheaper for someone in my position. I'd have to sell parts and buy different parts (at best a make work program for FedEx) and it would force design compromises I do not want to make.
    #6
    Mouse likes this.
  7. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    I have a "free" wheel with a cush drive that shares almost all critical dimensions with a DRZ from my 640 Adventure. The only unknowns are disk to sprocket distance and total hub width. Hub width is a trivial matter of turning new spacers on the lathe. Disk to sprocket distance is a bit trickier. The sprocket carrier may need to be turned to bring the sprocket closer to the disk or I may need to shim the sprocket or disk or caliper mount to move them farther apart.
    #7
  8. tntmo

    tntmo Oops, I did it again.

    Joined:
    Jan 31, 2012
    Oddometer:
    892
    Location:
    San Diego, CA
    Did your DRZ come with two sets of wheels? If so, that's $1000 or more back into the project money I would think. You can also sell the front forks which are a pretty good upgrade for DR650 or DR350 guys.

    I sort of agree that it's a lot of work when there are other avenues, but I like the spirit of using what you have to make it the way you want it. Just because I wouldn't do it doesn't mean I don't want to see someone else do it, so keep the updates coming. Looking forward to seeing you overcome the challenges and creating your own masterpiece.
    #8
  9. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    I did not get so lucky to get two sets of wheels. I got the stock bore and the stock exhaust. I am not sure if I can turn that into cash but the supermoto wheels should sell

    I am on the fence but I am thinking of keeping the fork for another project. I am going to see how this project progresses before I make the keep/sell choice on the fork.
    #9
  10. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    So about the fork conversion. As I outlined above, KTM made an odd choice of bearing. Suzuki chose 52mm as an outer diameter and fortunantly for me, that means I have a few choices for the bearings I can use without altering the head tube on the frame.

    I pressed the steer tube out of the lower triple-tree and made a drawing of the stem:
    IMG_0346.JPG
    The really critical dimensions on this drawing are the 30mm diameter on the bottom of the stem where the stem presses into the lower tripple-tree and the 28mm diameter where the stem is clampped by the upper tripple tree.

    These dimensions mean that I cannot use any bearing with an inner diameter over 30mm without causing problems with how the stem is pressed in, I'd need to press the stem from the top down and I'd loose the mechanical safety of the 32mm lip that prevents the stem from walking out. If I choose to use a 30mm ID bearing, I'd need to fabricate an entirely new steer tube since adding material really isn't an option here.

    The 28mm section means that if I choose to use anything smaller than 28mm ID then I need to shim the top clamp. This isn't a show stopper but it should be avoided if possible

    Finally, there is the length component. If the stack height of the DRZ is taller than that of the 640 ADV by any significant amount, then I will either need to make a new steer tube or I will need to steal the steer tube from the DRZ and make it work with the KTM. Ideally I want to either use the DRZ fork elsewhere or sell it. This becomes MUCH more difficult if I "borrow" the steer tube. If I am forced to fabricate a steer tube, I run into the fact that my lathe cannot cut metric threads. I would either need to buy transposing gears or get an expensive tap or settle for SAE threads on an otherwise metric bike. None of these are ideal.

    Fortunantly for me, while DRZ frame compatible bearings are not available in 29mm ID, they are available in both 28 and 30mm. My plan of attack becomes, if the DRZ stack height is equal to or less than the 640 ADV, I turn the existing steer tube down to fit 28mm ID bearings. If the stack height is taller, I fabricate a new stem to fit the stock 30mm ID bearings that come on a DRZ.

    IMG_0338.JPG

    It is difficult to capture in a photo, but the stack height works out almost the exact same between the two bikes. Better lucky than good! So I order new bearings: All Balls Racing part 22-1024 which are 28x52x16 and are normally used on BMW motorcycles.
    #10
  11. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    So onto the lathe work. My lathe is inaccurate. It was used in production for many years then dropped on its face, sold cheap, disassembled, and sold even cheaper to me. When I bought it I measured the bed wear as a .043 drop on the front v-way. This means that I cannot get a perfectly straight cut using the carriage.

    A look through the text books suggests that for press fits in the range of what I am trying to make, I need .001" +/- .0005" interference. My lathe changes in diameter more than .003 over an inch with the carriage so no bueno. Since this is a critical job so it takes a few work-arounds to get a cut accurate enough for what I need.

    Step 1 is bust out the 4-jaw. This part is too large for my 5c collets and my collets have more or less .001" of runout anyway. With enough tinkering a 4-jaw can be adjusted to the accuracy of my tooling or about .0005" of runout in my case.

    IMG_0347.JPG

    Please forgive the oil smudged on the camera phone.

    This is done on two points to assure that not only is the steer tube centered, it is also concentric with the lathe's headstock.

    Next, A quick check of diameter with my questionably worn tooling yields the same diameter reading in two locations on the same section of the stem. Regardless of how accurate my tooling is, the same reading will confirm that KTM did not turn a taper into the tube and I can now use the existing tube to line up my compound. Fun fact: my compound slide is not accurate either! It cuts straight for about an inch then it flairs out about a thousandth. This inaccuracy can be confirmed as we line up the compound but in this case, we can use this inaccuracy to our advantage!

    IMG_0349.JPG

    As mentioned previously the first diameter on the steer tube is nominally 28mm and the inner diameter of the bearing is a lot more accurately ground to 28mm. We want the top bearing to be a snug sliding fit and the bottom bearing to be a press fit. We know the limitations of the lathe and don't trust our tooling.

    The bearing does not currently slide over the 28mm section on the steer tube. A measurement with my questionable tooling yields a reading of 1.1025" which is surprisingly close to what I should read for 28mm (1.1023"). The carriage is locked down and about .030" is taken off the 29mm section. Next, I took a cut starting over the 28 mm section and continuing until the compound ran out of travel. The cut was made so that it removed some material but did not completely clean the surface of the 28mm section. If you recal, we are true within .0005" this means that the compound is now cutting within .0005" of 28mm acording to KTM and really close to actually cutting 28mm.

    This accuracy is however, largely irelivant because it is much more important to have the correct fit than it is to know what diameter that fit is at. Here's where my inacurate compound comes in. I take .001" off at a time until the bearing slides on the steer tube. It only required the removal of .002". A tight sliding fit is .001" undersized from the inner diameter of the bearing. The flare caused by the inacurate compound causes the bearing to get stuck. the spot where the bearing gets stuck is the spot where we can measure the actual diameter of the bearing. Again regardless of the number my tools say, as long as they say that number again, then the diameter is the same. I add .001 to it for the interfearance fit I desire and now I know what diameter I need to read on my tools to know the bottom diameter is correct.

    IMG_0350.JPG

    To get a flat surface the entire length of the upper face, I cover my cut with prussian blue and move the carriage half an inch and lock it down again. I bring my tool into contact with the part so that the tool skims off the prussian blue but does not cut metal then advance the compound. To confirm what I know to be true, the first half inch only skims prussian blue off, the second half inch cuts to the desired diameter, and everything after that brings the surface to .002" too large. This process is repeated until I have cut as far as I can with the part where it is in the chuck.

    Next I pop the tube out, flip it around, and true up the other end. In the same manner as before, I cut 1" at a time off the bottom bearing seat to bring the diameter to what it needs to be for a press fit.

    Finally I swap over to the 3 jaw for the highly inaccurate and unimportant necking down of the center. This can be cut with the regular carriage travel.

    The completed stem now has extensively long bearing seats top and bottom. I press it back into the lower triple tree clamp. IMG_0355.JPG
    #11
  12. gunnerbuck

    gunnerbuck Island Hopper

    Joined:
    Nov 19, 2005
    Oddometer:
    5,974
    Location:
    N.V.I, B.C.

    Sounds a bit like you bought into an already well worn machine.. The intake roller is a part that fails slowly so there is time to notice a problem as the roller will wiggle around as the needles wear... If clearances upon inspection have opened up then that is the sign to change the roller... Did you not inspect them when you had the water pump out 500 miles before? The current intake roller in my main bikes engine has 120,000 KM on it with no wear showing...
    #12
  13. clapped_r6

    clapped_r6 The Spoad Warrior

    Joined:
    Jan 3, 2006
    Oddometer:
    3,250
    Location:
    Ridgefield, WA
    Shoehorning an fz07 motor into that chassis is my unicorn
    #13
  14. Beezer

    Beezer Long timer

    Joined:
    Oct 26, 2004
    Oddometer:
    8,134
    Location:
    Anchorage, formerly Spenard (hub of the universe)
    the biggest gripe on the DRZ is the narrow 5 speed transmission. if I were that deep into it, I would install one of several gear/gear sets that make the range wider. since I heard about that, I want another Z... it's on my list of things to do.
    #14
  15. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    I have owned the KTM for 30,000+ miles. The valves were fine when I checked them last and replacing the water pump was part of damage control for the broken lifter. I should have sidelined the engine and rebuilt it at that point but I had a rally and chose to risk it. I lost.

    The lifter failure happened rather abruptly (500 miles?) and I missed the signs thinking I had carburetor problems. The carb was indeed problematic but it turns out, not the source of my problem.
    #15
  16. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    I'll look into that this winter. I am rather happy with the bike's pep and power as it stands. A wider gear range is always a better thing if the engine can pull the gap between gears.
    #16
  17. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    Installing the fork

    Before disassembly, I checked how the lower bearing interacted with the lower triple tree clamp. IMG_0357.JPG Notice that the bearing is quite a ways up in the frame! This is fine for the DRZ clamp but not gonna work for the new triple tree.

    So now we disassemble the front end:

    IMG_0358.JPG

    Knockout the old bearings and tap in new races. I measured the difference between the end of the head tube and the race as .33". The distance that the bearing sit proud of the race is approximately .16" so I need a spacer .18" to leave me a bit of breathing room. A few minutes on the lathe and I have one!
    IMG_0359.JPG
    Grease and install the lower bearing on the triple tree - don't forget the seal. Grease the upper bearing and test fit the mess.

    IMG_0360.JPG

    Upside down? Well either way, I couldn't have done better on stack height if I tried.
    #17
  18. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    IMG_0362.JPG The rest of the install job is simply bolting KTM parts on the DRZ

    This install job did not address the fork stops. The DRZ frame tab is on the front of the head tube just above the lower triple tree clamp. I intend to make a mount for a headlight that integrates a fork stop this winter.

    Now I have a knobbie front and a street rear, lets see if we can address that problem.

    Pop the old wheel off and slide a KTM wheel in.

    OMG does it fit?!? The brake side fits well:

    IMG_0364.JPG
    But the chain side just bearly doesn't make it.

    IMG_0363.JPG
    The head of the sprocket bolts just grazes the weld on the bottom of the swing arm. If I didn't have a lathe, I'd file the weld and call it good.
    #18
    Te Hopo likes this.
  19. Beezer

    Beezer Long timer

    Joined:
    Oct 26, 2004
    Oddometer:
    8,134
    Location:
    Anchorage, formerly Spenard (hub of the universe)
    #19
    Z50R likes this.
  20. Z50R

    Z50R Not lost yet

    Joined:
    May 11, 2009
    Oddometer:
    1,288
    Location:
    Boston Mass.
    But I do have a lathe!

    If I move the sprocket .100" towards the center of the wheel, my problem goes away.

    Lets throw the sprocket carrier in the 4 jaw... oops, it don't fit! If I had a larger lathe it would. If I had an independent 3 or 6 jaw I'd be able to grab it. If I had the right face plate I could make it work.

    In the end, I grab the carrier with the 4 jaw by removing two jaws, inverting the other two, and grabbing it between two lobes on the cush drive.

    IMG_0365.JPG

    If you think truing a part in a four jaw is tedious, try doing it down a couple jaws!

    Eventually there is more variation in the profile and no decernable error from center. So I skim the dust on the ring and take a tenth of an inch off the face.

    IMG_0366.JPG

    Bolt the mess back together and slap it on the bike. The sprocket now travels through the center of the chain guide and nothing rubs. Success!

    IMG_0367.JPG

    That is as far as this project goes until snow flies. I can ride again and I intend to do so while I am able.
    #20