Zero bearing interference on Rotax DSB?!?!

FMR

Member
Got a problem i would like your thoughts on.
I’m in the process of renovating a Rotax DSB. When tearing it down i discovered that the crank bearing on the sprocket side slips in and out of the seat without even heating the half up. As far as i can tell it has 0 interference at room temperature, which cant be right, right?

So my question is: How do i solve this?

Problem is that the amount of material from the bottom of the seat to the intake port is maybe 1-2mm so to bore it out and drop in a sleeve doesn’t look like an option as it would probably crack or break thru to the intake port.
 

Attachments

  • 60FEEDB2-9F3B-4E65-912D-2FB98E6FE8CC.jpeg
    60FEEDB2-9F3B-4E65-912D-2FB98E6FE8CC.jpeg
    170.2 KB · Views: 48
  • C426893A-AA5C-4B4C-B788-FD98531B705D.jpeg
    C426893A-AA5C-4B4C-B788-FD98531B705D.jpeg
    99.3 KB · Views: 44
  • 43B8201A-AC34-46F0-9A42-6AB04FF2238C.jpeg
    43B8201A-AC34-46F0-9A42-6AB04FF2238C.jpeg
    99.7 KB · Views: 44
Got a problem i would like your thoughts on.
I’m in the process of renovating a Rotax DSB. When tearing it down i discovered that the crank bearing on the sprocket side slips in and out of the seat without even heating the half up. As far as i can tell it has 0 interference at room temperature, which cant be right, right?

So my question is: How do i solve this?

Problem is that the amount of material from the bottom of the seat to the intake port is maybe 1-2mm so to bore it out and drop in a sleeve doesn’t look like an option as it would probably crack or break thru to the intake port.
What brand bearing are you using? I had the same problem with a KT100 several years ago, I was using bearings I bought off EBAY which turned out to be counterfeit. Once I bought a set of SKF bearings from my local supplier, they fit with the proper interference.
 
FMR,

I've repaired many rotary valve cases where the sleeve ended up slightly protruding into the intake port. I just touch them up with a die-grinder and a bit of light polishing afterwards, and it ends up being barely noticeable. It's not an issue at all.

Is that a 6304 bearing or a 6203? (52mm or 47mm OD)? My normal sleeve wall thickness for the 52mm bore is .090", however you could bring that down a touch to minimize intake port "protrusion". I would go no lower than .075" wall on the finished sleeve though (finished, as in... when the bore is to final size).

Have you checked bearing interference on the opposite side? If it's "loose-ish", it might be worth sleeving both sides and doing an accurate line-bore on the case, (and also skim the rotary valve pocket and valve cover mounting face precisely 90 degrees to the crank centerline while you're at it).

I would *highly* recommend avoiding any sort of loctite or retainer compound. The reason is... those "chemical solutions" make things tight by swelling as they cure, and they can actually exert enough force on a bearing to act like it has *too* much interference. I've seen bearings spin extremely poorly after being "retained" with Loctite or other chemical compounds.

Fire away if you have questions.


PS: that bearing appears to have been loose for a bit, judging by the back face it sits against. :(

PM
 
Dan: 99% i use SKF bearings from trusted supplyers. The bearing that came out of this engine is a no-name one, but measuring the OD i get the same measurement as on a new SKF.

Pete: Thanks for the info. I thought that sleeving would be a problem, mainly because i thought it would crack easily. But if that’s not a problem, then sleeving it is! The bearing on the valve side is 6304 (47mm) and on ignition side it is 52mm with plastic sleeve (and guess what i found when i opened it up, yep this engine also has had a crank pin failure with welding around ignition side bearing, as we discussed previously). Therefore i plan to sleeve the ignition side also with aluminium sleeve so line boring it may very well be an option.
If you have the time to explain to me in detail how you would go about line boring it i would really appriciate it.
 
Yes, I would definitely re-do the ignition side with aluminum.

Let me put something together on what my process is, and I'll post it here by this evening. I may even have tooling here for holding a Rotax that I can photograph for you.

PM
 
FMR,

When I make sleeves, I try to use 6061-6511 tubing if possible (it stays "rounder" when you part it off compared to normal "T6"). Ideally, I try and make sure that sleeves end up round within .0003" or better. It not only makes them easier to install, but also makes them easier to measure.

I bore the sleeve blank to about .020" under the finish size of the bearing bore, then turn the OD to your preferred size (I'd go 2.230" for the 52mm bearing sleeve, and for the RV side you might be able to get away making the finish OD of the sleeve 2.000" (so you can make it out of 2" OD tubing or 2" bar stock).

Use the same feed rate on your lathe to cut the OD of your sleeves as you will use to finish-machine the the bore the sleeve drops into. (fairly fine feed: .005" per rev or less)

Part the sleeves off about 1/16" longer than the width of the bearing: it gives you something to hold on to when you drop them in the case half. Make sure the entry end of the sleeve has a nice decent chamfer on it, and then touch the corners of the chamfers with a fine file so there is NO sharp edge (where chamfer meets OD) that could hang up when you drop in the sleeve (where the red arrow points in the photo below). I also chamfer the bore a decent amount on that same end.

For interference, I would recommend .0028"-.0030" on boring for the ignition side, and .0025" on boring the RV side (47mm sleeve).

I personally use PCD tooling (polycrystalline diamond) to machine everything: making the sleeves, as well as all the finish work on the cases.

To install... everything squeaky clean. I *highly* recommend using an oven to heat the cases, not a torch, hotplate or toaster oven. If possible, put a plate of aluminum in the oven that you will set the case halves on (prevents the very direct heat from going on the case, and provides a good heat sink to set the cases on). Preheat oven to 350F. Set the case halves on your aluminum plate... typically if the oven and the plate is up to temp, 15 minutes will do, but you can leave them in 30 minutes if you want... just don't set the oven higher than 350.

Sleeves in the freezer.

Pull a case half out, set it on something that doesn't suck heat out, grab the sleeve out of the freezer and wipe it dry with a paper towel (quickly). You don't want any condensation on the outside. Grab it by your fingertips on the very end, and drop it in. This whole process takes much less time than it took to read the last two paragraphs.


So that covers making the sleeves and dropping them in. I'll get back to this tomorrow and explain how I bore the cases for the sleeves, and then how I do the finish bores so everything is lined up.

PM

sleeve2.jpeg
 
Last edited:
Part 2 of repairing bearing bore(s):

I will just explain how I go about aligning things on pretty much any engine case I’ve ever reworked.

Most of the time, my aim is to base crank centerline on where the seal bores are. I know… one would assume that seal bores and bearing bores should all be concentric, but it’s not “always” true, unfortunately. Also, if a bearing bore has had a loose bearing spinning in it, then it’s likely the bore did not “wear” perfectly concentric to the original bearing bore. Using the seal bores as the reference centerline is the best method, since those are the two bores we do not want to move or touch.

OK… so assumption #1: Line up the finished bearing bores concentric with the centerline between the 2 seal bores. (in other words, all 4 bores should be perfectly concentric when done).

My next “goal” is to do my best to remove the absolute minimum amount of aluminum on any other surface I am going to machine.

Now we choose a place to start, and that’s going to be the center/sealing face of each case half. On the rare occasion that I will machine that face, it’s because someone has asked for a very precise width between bearings.

So for this case, we’ll make assumption #2: We are not going to machine any material off the sealing faces of each case half.
_ _ _ _ _ _ _ _ _

“Usually”, what I will do is remove any alignment pins from the case halves (IF they can be removed), and I will lightly lap each case half on a piece of 400 grit sandpaper on the surface plate. More than anything, this will give a confirmation that the face is flat, which will be the starting point of things.

On my lathe, I use faceplates to do everything. My approach is to have a 1” thick disc/circle of aluminum that is typically around 8” in diameter, and it’s attached to the faceplate on the lathe via 3 “stand-offs”. I just use 3 pieces of 1” aluminum bar-stock with a ⅜” hole in the center, and the faceplate has 3 counterbored holes for SHC screws (heads need to be well below flush). Once this is mounted on the lathe, it can be skimmed flat. I usually skim the OD and also drill and tap a ½-13 thread in the center (more on that later). Each of my faceplates has hole patterns for different engines: obviously I have one for the KT100, and others can sometimes be used for more than one engine make or model. My lathe has a D1-6 spindle nose, so in my case…. every faceplate is mounted on its own D1-6 steel faceplate… I never take them apart.

The next step is deciding how each case half will be held. I have a cabinet full of various “adapters” I’ve made over the last 40 years for holding all types of engine cases. You only need one for each side of your Rotax, so at worst, you’ll need to make a two small parts. Generally speaking, there will be some sort of “face” on each case half that I choose to mount to for holding each case half. One side almost always has a face and register for an ignition stator, so that will take care of how to mount one of the halves, and the PTO side can sometimes be a bit more challenging.

Ultimately, the goal is to be able to…. mount each case half with the center/sealing face against the faceplate (which runs dead true since it’s been machined), and the lightly skim whatever surface you want to use as the mounting point when the case half is flipped over to do machine work on the bearing bore. Any adapters to be made should be dead parallel, so once the case half is mounted with bearing bore facing out, the face of the case runs dead true.

At this point (once all the above is done), the case should be indicated in so the seal bore runs dead true, and the bearing bore machined out to accept the sleeve. I try never to touch the back face during the process… I save that for when doing the finish machining on the bearing bore.

I should stress one final point, which really applies to any machining process being done on case halves or the complete case: ALWAYS use the least possible clamping force during every operation. If a case half is being held to the face plate with a few studs and nuts, tighten everything very lightly. Some cases can be very flexible, and simply do not have enough material to make them "robust". A KT100 case is a perfect example, but so is something like a vintage B-Bomb case. Everything being machined for ultimate accuracy should be held/clamped carefully and with very light clamping force.

The previous post should cover how to make the sleeves and drop them in once the case halves are machined, but fire away if you have questions. Next step will be how to do the finish work so everything is lined up perfectly.

Pics of a faceplate added. This happens to be one I made that will handle "most" older IAME cases. Bolt patterns for top, bottom and sides of cases.

1.jpeg



2.jpeg




PM
 
I´ve been swamped with work and havn´t had the time to read until now.

But WOW! Pete, I really appriciate the work you put in to share your knowledge.
I think i get most of it, but i´m having a hard time to understand some of the steps of the process.
A couple of questions:

1. I don´t get the use of the stand-offs, what is the reason for using them?

2. Say you want to use the ignition stator face to hold that case halve. When you flip it around (to do the bore), do you mean you then make a face plate formed as an ignition stator to mount the case on?
Do you have any more pictures on this particular process or face plates?

3. I mostly use reed engines and the sprocket side case halve has no machined surfaces on the outside of the case. What do you think of making three drilled-thru machinist jacks to mount the case with (using the bolt holes in the case halve) against the face plate? This would mean you would need to indicate it instead to get the face true. Do you think that would work or do you already have a better solution for this problem?

Again many thanks!
 
Question 1:
95% of the time, a case half will be attached to the face plate using studs screwed into the case. Standing off the plate with all the bolt patterns in it allows you to get behind it with your fingers and a wrench to screw on and tighten a nut.

Question 2:
No, the face plate will always be flat. An adapter is required. For adapting to an ignition stator face, that will be a round piece of aluminum, faces of it perfectly parallel, and a bolt pattern in it that allows it to be attached to the using the same threaded holes used to attach the stator. The one I have for Motoplats actually has 3 elongated slots that allow it to bolt to any engine (not all engine manufacturers put the 3 tapped holes in the same spacing/pattern). If you're only making an adapter for one engine though, drill/counterbore the 3 holes is much easier.

This adapter needs to seat against the face that has been "pre-machined" dead flat, and should also loosely pilot on the register that is there to center/align the stator.

The adapter will have a ~1/2" hole through the middle (and probably a counterbore from the "engine" side of this adapter). You screw a 1/2" threaded stud into the face plate (center of the face plate tapped 1/2-13), and now you attach your case half (with adapter bolted to it) to the faceplate by reaching through the seal bore and tightening this "assembly" to the faceplate with a 1/2-13 nut.

I'll post some photos.

Question 3:
I have never done that. I *always* find a way to machine "something" on the outside face that is parallel to the center face. In extreme situations, I also have a selection of very thin-wall steel parts that will slip into the seal bore (from the bearing side), and allow me to get a nut in there and pull the case against some type of adapter that the outside face of the case is sitting on (and onto the face plate, of course) . These small steel parts have to be counterbored so a nut can sit down inside it, and must have a very small/thin lip to pick up a tiny bit of the back face of the bearing bore. There needs to be enough room in there so the bearing bore can still be machined (and the back face as well) without touching this small steel part. This can get a bit "delicate" to do, depending on the case type, but I've done the machine work many, many times this way when there is no other choice.
_ _ _ _ _

One other thing I'll mention regarding face plate and/or adapters: Somewhere in that stack-up (whether it's case half to faceplate, or case half to adapter to faceplate), there has to be a way to shim or tweak things when it comes to doing the finish machining. There is absolutely no guarantee that when the case assembly gets bolted up to the faceplate (for final machining), that the seal bore centers will be perfectly lined up. I'll get into the details on this in the next post.

And once again, I'll mention how important it is to hold everything lightly ( just tight enough so nothing moves when machining). Many case halves are easily distorted if any sort of significant torque is applied to any of the mounting screws.

PM
 
FMR,

Follow up on previous message. Here is a photo of "some" of the adapters I have for doing various engines (I have more, just didn't feel like fully unloading that shelf of the cabinet).

Some that I used frequently, I took that time to make them very nice... machined all over. Others might have been to do one or two rare engines, and usually it ended up being something I could make quickly (but always machined both sides dead parallel, even if it was just a band-sawed part out of plate).

I also did a quick count on faceplates for the lathe, and I have 8 different ones, though two of them were used for probably 85% of my work (KT100 and IAME specific face plates).

I've been slacking a bit, so need to get back on this thread and fill in the details on how to do the finish machining. I'll get to it shortly!

1.jpeg
 
Thanks for clearing that up Pete.
I think i get most if it now. What i need to do is to get some material and start making a faceplate and a couple of adapters. As i understand it your faceplates are made to center perfectly on your spindle, right? So the steps to sleeving that case i showed in my earlier post would be:

1. make face plate with register for adapter
2. attach case, with rotary side out
3. lightly skim rotary side
4. make adapter to register between faceplate and case rotary side
5. attach case to adapter
6. attach adapter to faceplate
7. start work on bore

Did i miss or missunderstand something?
 
Thanks for clearing that up Pete.
I think i get most if it now. What i need to do is to get some material and start making a faceplate and a couple of adapters. As i understand it your faceplates are made to center perfectly on your spindle, right? So the steps to sleeving that case i showed in my earlier post would be:

1. make face plate with register for adapter
2. attach case, with rotary side out
3. lightly skim rotary side
4. make adapter to register between faceplate and case rotary side
5. attach case to adapter
6. attach adapter to faceplate
7. start work on bore

Did i miss or missunderstand something?

What's important about the faceplate is that the face is machined dead true. But yes... when I make a new faceplate, I typically bolt it on, face it, turn the OD, and drill and tap a ½" x 13 hole in the center. IF I remove the faceplate for any reason, Then it's simple to bolt it back on, indicate the OD, and tighten things up. Not a bad idea to make the 3 standoffs precisely the same length too, of course... as that makes it more likely that the faceplate will bolt back on the same as it was initially.

1) No register. Faceplate should just be dead flat. Need to have the ability to bump things around a bit "somewhere" in your setup.

2) Yes. Make sure to lightly lap the center/sealing face first though so you know it's sitting dead flat on your faceplate.

3) Well... if necessary. Some of this depends on the "big picture" -- how you plan on holding the case to the faceplate for the finish boring of the bearing bores. The first thing to do is attach the case half to the faceplate and indicate it in so the bore of the RV recess runs true. At that point, I would indicate the face of the counterbore (where the valve sits) and the face of the lip that the RV cover bolts to. See how those run. If everything happens to run dead true, then no machining of this "setup face" is required (this almost NEVER happens though). For me, I "generally" take a very light skim on the lip face that the RV cover bolts to. (make sure you use a depths mic to measure how deep the opening for the RV is before machining, as you'll want to make sure when you're finishing the job, you are back to that dimension).

4) Yes.

5) Well.... yes. Sometimes on a RV case, I am just "pulling up" the case half to the faceplate with a center stud and a small counterbored "bucket" with a lip on it to catch the very edge of the center bore. It all depends on the configuration of the case half. Lots of ways to "skin the cat" so to speak. Whatever allows you to lightly attach the case half to the faceplate, and indicate in the seal/center bore (that's our reference). If course if you attach by pulling down in the center in some way, a bit of measuring should be done ahead of time to make sure that seal/center bore runs true to "something" else that you have indicated. This might sound convoluted, but it's not. I'll try and make a quick/short video on this.

6) Depending on how you've made your adapter, yes.

7) Yes. I generally make the sleeves first so I have the target size for the bore in the case half.

PM
 
FMR,

How is your Rotax project going?

I just blueprinted a KT100 case yesterday for someone going to New Castle next month for the VKA event. I'm actually building the full "long block" for a friend, and all they will have to do is install carb and ignition.

An interesting side note regarding this case: there are some KT100 cases that can actually be line-bored without sleeving. They are typically very early permanent mold cases (late 70's to probably mid-80's vintage). The reason it's "possible" (though not guaranteed) is because that era of case had very tight interference on the bearing bores (approaching .0020"). If the case has not been abused, and is very low time (very few or no bearing changes), then with some work... things can be lined up and the bearing interference can be set to an ideal number by just using the tiny bit of material that is available.

On this particular case, one bearing bore had .0020" interference, and the other was a bit out of round with around .0014" to .0018" interference. The photo below is the step *before* I machine the bearing bores. The case has been flip-flopped a few times... holding by the PTO side and holding by the ignition side, with a bit of manipulation until I can get both existing bearings bores running true. This is the final "skim" of the ignition side after getting the existing bores to run dead true. So after this face is skimmed clean, the ignition side case half is removed and final machine work is done on the PTO side bearing bore and the back face (that the ball bearing sits against). Once that's done, I can attach the ignition side to my faceplate, indicate in the bearing bore and take it to size (and just clean the back face of the bearing bore).

Needless to say, it requires a little bit of care since on this case I was only machining about .0007" out of the bearing bore on one case half, and around .0002"-.0005" out of the bearing bore on the other half (to get to the desired bearing interference).

It's actually far easier to sleeve and line-bore a case, as then there is "material to work with", but this is something I do on the rare occasion when I can get my hands on cases that are of the correct casting and condition.

PM

facing.jpeg
 
Great info Pete. I just needed to post on this thread so I can find it easily in case I need to later on.
 
Pete, I've read your posts for years and been blown away by your knowledge, skill, and the clarity of your writing.
Have you ever considered writing a book? I'll happily sign up for a copy or two.

John
 
Back
Top