XS650 Clutch Pushrod experiment & tidbits

TwoManyXS1Bs

BBQ Hunter
Top Contributor
Messages
14,670
Reaction score
15,273
Points
813
Location
Fredericksburg, Texas
This is mostly a 'thought experiment', and a companion piece to the clutch worm experiment:

http://www.xs650.com/forum/showthread.php?t=31554

The XS clutch has the notorious feature of slackening-up when the engine is hot, and the conventional wisdom blames this on differential thermal expansion rates between the aluminum engine case and clutch pushrod(s). This exercise explores the potential relief if an alternate material is used for a 1-piece pushrod. I have never handled, much less seen, a 2-piece setup, which I understand utilized aluminum in the longer of the 2-piece pushrod setup to overcome this expansion problem.

The computations here are rough ballpark, to determine the feasability of using alternate clutch pushrod materials. These alternate material candidates were selected based on availability, strength and thermal expansion coefficients.

Pic #1 - I have a few 5/16" rods of these materials, which are 304 Stainless, Navy brass, 6061 aluminum and 7075 structural aluminum.

NewPushRods.jpg

Considering whether it's feasable to conduct this experiment. Others, especially ME's, are encouraged to jump-in here and comment on this project. I have this in a spreadsheet, but decided to expand it here to improve understanding.

1-piece stock clutch pushrod:
Length = 9.570" (243 mm), we'll use 10" for our rough sample calculations
Diameter = 8mm (.315"), can use 5/16" (.3125)
Material = unknown Carbon steel

Depending on type of clutch worm and stock/aftermarket clutch levers, total leverage multiplier at clutch lever is about 30-50.

Pushrod stress:
At a tested mid-range compression of 370 lbs (stock XS1B) , a 5/16" rod (1/13 sq-in cross-sectional area) is experiencing about 5000 psi of compressive stress.

Normal compressive load distribution can be used when length/diameter ratios are below 6. Special column loading considerations and limits must be employed when the length/diameter ratios are 6 to 60, and especially at 60-120 (severe buckling phenomena). The clutch pushrod length/diameter ratio is about 30. Without going into all the engineering math here, all the selected materials here (within this stress value) qualify for normal compressive stress/strain calculations, without worrying about buckling.

Decided to drop the 6061 aluminum since it would be a slightly weaker and redundant version of the 7075 aluminum.

Coefficients of thermal expansion (within normal engine temperature ranges). Units are MicroInch per °F (.000001" / °F) per inch of material
Engine case Aluminum = 13.1
Carbon Steel = 7.8
304 Stainless steel = 9.6
Navy Brass = 11.8
7075 Aluminum = 13.1

Compute expansion of 10" of material based on 200°F temperature rise (70°F - 270°F)
Engine case Aluminum = 0.026"
Carbon Steel = 0.016"
304 Stainless steel = 0.019"
Navy Brass = 0.024"
7075 Aluminum = 0.026"

However, these candidate clutch pushrods undergo compression during clutch disengagement, and deform (shorten) under stress. The amount of length change (strain) can be predicted using the modulus of elasticity (Young's Modulus) for each material, when stressed below the material yield strength. The higher the modulus, the stiffer the material. The length change (strain) is: stress (psi) / Modulus, then times the overall length.

Rough values of Young's Modulus (in millions of psi):
Carbon Steel = 30
304 Stainless steel = 30
Navy Brass = 15
7075 Aluminum = 10

Compute length change (strain) for 5000 psi stress on 10" rod
Carbon Steel = -0.002"
304 Stainless steel = -0.002"
Navy Brass = -0.003"
7075 Aluminum = -0.005"

Now, compute overall length change (expansion -strain).
Carbon Steel = 0.016" - 0.002" = 0.014"
304 Stainless steel = 0.019" - 0.002" = 0.017"
Navy Brass = 0.024" - 0.003" = 0.021"
7075 Aluminum = 0.026" - 0.005" = 0.021"

Interestingly, the Navy brass and 7075 aluminum are equal.

Now, compute overall length differential to aluminum engine case (0.026" - rod expansion).
The smaller, the better.
Include 30-50 leverage multiplier to show slackening value at clutch lever.

Carbon Steel = 0.026" - 0.014" = 0.012" (0.36" - 0.60" at clutch lever)
304 Stainless steel = 0.026" - 0.017" = 0.009" (0.27" - 0.45" at clutch lever)
Navy Brass = 0.026" - 0.021" = 0.005" (0.15" - 0.25" at clutch lever)
7075 Aluminum = 0.026" - 0.021" = 0.005" (0.15" - 0.25" at clutch lever)

It's winter, too cold to experiment with this now, just throwin' this out there. Anybody else try any of this? I see no issues in at least trying the 304 stainless, but would appreciate any insight.
 
Last edited:
New 1-piece clutch pushrods have been found to have these dimensions:

Diameter = 0.315" (8mm), but 5/16" (0.3125) will work fine
Length = 9.570" (243 mm)
Outboard/worm end has reduced diameter of 0.262" - 0.267" (7mm)
This reduced diameter is 0.300" - 0.325" (8 mm) long

Received a new MikesXS clutch pushrod. Inspection revealed that the ends are rough, the body is mill-ground, not polished, and rolling it on a flat plate shows that the rod is slightly bent. I'll straighten it, redress the ends, and polish the whole surface.

Pic #1 - Small-end, my polished/dimpled rod on left, new MikesXS on right.

Pic #2 - Clutchpack end, my polished/dimpled rod on left, new MikesXS on right, very rough.
 

Attachments

  • PushRodA01.jpg
    PushRodA01.jpg
    88.8 KB · Views: 832
  • PushRodA02.jpg
    PushRodA02.jpg
    94.8 KB · Views: 674
Another you might consider is the XS500 pushrod. It's a long one piece affair that is mostly aluminum but with steel ends. It has a couple short-comings. First it's a bit too large with about an 8.65mm diameter, but that would be no problem for you with access to a lathe. Second problem is it's about 10mm too short. I thought that might be overcome by just using 2 balls in the motor. Or maybe sourcing a mushroom pusher piece (the part that contacts the pressure plate) that is a little longer.
 
Great! XS500 pushrod, steel-tipped aluminum. I can steel-tip the 7075 rods. That'll solve diameter and length. I wonder which alloy it uses. I'll need to lookup the parts fiche to see if the rod layout is similar to the XS650, i.e. exposure to the drivechain and the elements, and passage thru a similar mainshaft bushing/seal. The length/diameter ratio is 27, not too different, but the cross-sectional stress would be less. Thanx, 5Twins, good info...
 
This has become rocket surgery.

How many nanoatoms are displaced when the thermocupling disengages from the fluxcapaciter causing neurotransmitters to collide with doohickey that rotates the inflated multicompound radial space displacer resultinging in a power ratio of 88mph.

Now thats important stuff.

Twomany.....sweet write up.
 
This has become rocket surgery.

How many nanoatoms are displaced when the thermocupling disengages from the fluxcapaciter causing neurotransmitters to collide with doohickey that rotates the inflated multicompound radial space displacer resultinging in a power ratio of 88mph.

Now thats important stuff.

Twomany.....sweet write up.

Don'tforget your blinker fuild and exhaust bearings lol:laugh:

Posted via Mobile
 
How many nanoatoms are displaced when the thermocupling disengages from the fluxcapaciter causing neurotransmitters to collide with doohickey that rotates the inflated multicompound radial space displacer resultinging in a power ratio of 88mph?

Hahaha, last time I checked ..... three!

Thanx, EvenmoreXS. More fun stuff to follow.

Next week we'll figure out how many pancakes fill a doghouse.

5Twins: "Here's an XS500 rod on eBay. Yes, the outer 70mm is steel. It's probably done that way so steel mates with the bushing in the case. That would wear better than aluminum."

Thanx, 5Twins, then that's the path I'll take. See the next post...
 
Here's a clear pic of the XS500 clutch pushrod. The main body is aluminum (unknown alloy), with a short steel tip at the clutchpack end, and a longer steel shaft to pass through the pushrod seal and bear against the clutch actuator worm. According to postings on our sister xs400 forum, these parts appear to be press-fitted together.

I feel like I'm reinventing the wheel, the XS500 engineers implemented this over 35 years ago. Pity this particular design didn't transfer over to the XS650, unless there was a secret reason to not do it this way.

So, I'll be making a couple of rods like this XS500 rod, probably with 304 stainless and 7075 aluminum, interference fit, assembled hot/cold in an alignment jig.
 

Attachments

  • XS500-ClutchPushrod.jpg
    XS500-ClutchPushrod.jpg
    10.9 KB · Views: 680
Here's a clear pic of the XS500 clutch pushrod. The main body is aluminum (unknown alloy), with a short steel tip at the clutchpack end, and a longer steel shaft to pass through the pushrod seal and bear against the clutch actuator worm. According to postings on our sister xs400 forum, these parts appear to be press-fitted together.

I feel like I'm reinventing the wheel, the XS500 engineers implemented this over 35 years ago. Pity this particular design didn't transfer over to the XS650, unless there was a secret reason to not do it this way.

So, I'll be making a couple of rods like this XS500 rod, probably with 304 stainless and 7075 aluminum, interference fit, assembled hot/cold in an alignment jig.
For your information, here are some of the measurements of the XS650 2-piece push rod:
steel rod: 85mm
Aluminum rod: 150mm (over-all length, including the steel tips)
Steel tip on each end of aluminum rod: 4mm
ball bearing: 8mm
You can look at it closer (and have a set) when you see me.
I think the ones you will be making will be a great improvement, and I'll take ten please.
I do think I see a reason for the 2-piece design, aside from the expansion of the 2-piece rod better matching the expansion of the engine:
The rod spins as it contacts the rotating clutch. A spinning rod accelerates wear on the push rod seal and bushing. The ball bearing between the two rods reduces or eliminates the spinning of the outer rod, thereby increasing seal and bushing life.
 
I feel like I'm reinventing the wheel, the XS500 engineers implemented this over 35 years ago. Pity this particular design didn't transfer over to the XS650, unless there was a secret reason to not do it this way.

So, I'll be making a couple of rods like this XS500 rod, probably with 304 stainless and 7075 aluminum, interference fit, assembled hot/cold in an alignment jig.

hard to believe isn't it !:doh:

It does set you wondering how much other stuff has been upgraded or improved on other Yam models over the years that could be utilised on our XS650's doesn't it.!

I am looking forward to when you get to analysing and testing the clutch springs because my feeling is that they are going to prove to be the biggest source of temperature related change on uniform clutch pressure . :thumbsup:
 
...measurements of the XS650 2-piece push rod:
steel rod: 85mm
Aluminum rod: 150mm (over-all length, including the steel tips)
Steel tip on each end of aluminum rod: 4mm
ball bearing: 8mm

That gives 85mm + 8mm + 150mm = 243mm overall length, matches the 1-piece, good info.

... you will be making ... and I'll take ten please...

Hahaha, you're my best customer! We'll give this a try and see if it works.

...reason for the 2-piece design, aside from the expansion of the 2-piece rod better matching the expansion of the engine:
The rod spins as it contacts the rotating clutch. A spinning rod accelerates wear on the push rod seal and bushing. The ball bearing between the two rods reduces or eliminates the spinning of the outer rod, thereby increasing seal and bushing life.

I see you there. I was also wondering about the worm/case/mainshaft misalignment mentioned in the 'worm experiment' thread, thinking that Yamaha expected this 2-piece design to reduce or eliminate side loading in that seal/bearing area. That's why I advocate a perfectly flat adjuster tip, removing misalignment sideloads by letting the ball seek its own center.

To restrict buckling of the rod-ball-rod combo, the passage through the mainshaft would need to be a close fit, but that would introduce another problem of oil restriction, as oil is delivered through this pushrod/mainshaft passage. I wonder if this mainshaft hollow changed diameters over the years?

Now I'm wondering if I may have introduced a problem when I made/fitted a close-fitting non-split pushrod bushing. Maybe the split in the bushing is necessary to guarantee oil flow to the mainshaft?
 
hard to believe isn't it !:doh:

It does set you wondering how much other stuff has been upgraded or improved on other Yam models over the years that could be utilised on our XS650's doesn't it.!

I've seen things, man... (from a Seinfeld episode?)

Having been involved in numerous computer hardware/firmware/software development and release cycles, I can personally attest the hundreds of thousands of 'bandaids' and mis-communications in that industry. Kinda sensitive to it, but also expect it as a natural part of product evolution. Some of them are real headshakers.

A lot of the fellows out there have been mix/matching bike parts for quite a while, their own version of 'technology transfer'. Fun stuff. Reminds me of when I used to hop-up h*nda 125's by resleeving with 305 sleeves and boring out to fit a cb750 piston, makes it a 145cc, using just factory parts.

I am looking forward to when you get to analysing and testing the clutch springs because my feeling is that they are going to prove to be the biggest source of temperature related change on uniform clutch pressure . :thumbsup:

Me too. Yep, doing a small crawl up through this clutch system, hopefully ending at the clutch pack. Will need to build a spring tester (always wanted one, but need certain features)...
 
Decided not to bore you with the math...


If a BiMetal clutch pushrod were made from a combination of 4" of 304 stainless and 6" of 7075 aluminum, compute length changes, add to the list, and make it easier to read:

Carbon Steel ====== 0.012" (0.36" - 0.60" at clutch lever)
304 Stainless steel = 0.009" (0.27" - 0.45" at clutch lever)
Navy Brass ======= 0.005" (0.15" - 0.25" at clutch lever)
7075 Aluminum === 0.005" (0.15" - 0.25" at clutch lever)
304/7075 BiMetal == 0.006" (0.18" - 0.30" at clutch lever)

A 304/7075 BiMetal rod would cut the heat expansion slop by a half compared to the stock steel clutch pushrod.
 
Any oil in the hole the rod passes through is just pulled through by the action of the rod as it pushes in and out. As long as it gets some oil it will be fine.
I guess you could polish the hole out if you feel it's to much friction.
Leo
 
XSLeo, if you study your crankcase oil distribution schematic, you'll see that gallery-pressure oil is delivered to the small space behind the pushrod oilseal (that's why the seal has a flange). From there, it travels thru the mainshaft, some is released thru small oiling holes in the mainshaft, the remainder passes alongside the pushrod mushroom shaft, to emerge into the clutch hub, where it is 'flung' thru the clutch plates thru those holes in the hub.

On the H*nda cb750s, we used to modify (per bulletin/warranty) the oil restrictor that fed oil to the clutch hub to oil the plates, to prevent clutch 'chirp'. Those were dry-sump, and needed that oil.

The XS650 wet-sump bathes the clutch in oil, so the need for oil (and how much) to be fed into the hub is an unknown to me. I've often wondered if some reported clutch problems could be sourced to problems with that mushroom opening oil delivery.
 
The oil does get pumped into the area behind the seal. This oil is to lube the bearing. Most of the oil flows out through the bearing.
Looking closer I can see some oil gets pumped through the transmission main shaft. Most goes through the bearing and into the hollow shaft the shift forks ride on where it sprays out to help lube the tranny gears. On the other end of the shaft oil gets pumped to the transmission output shaft.
I haven't been inside to many different bikes engines but this is a very good system. A lot of transmissions only get oiled by splash.
Leo
 
Thanx, Leo. The oiling down that mainshaft may not be an issue after all, we don't hear of any burnout problems in the fitment of the gears to the mainshaft, and don't hear of any 'chirp' insufficient-oiling problems in the clutch.

The oil delivery thru the shift-fork guide rod does interest me. In the pursuit of neutral while stopped, you're either trying to come up from first, or down from second. Those gears are at the ends of the shafts, and the shift drum is trying to shove a shift fork over to disengage one of those dogs/cogs. The geometrical arrangement of the drum-slot, fork-pin, guide rod, and fork ears is awkward, and I've been suspecting some kind of gouge/binding when those are put under load. But, will be experimenting there much later during the overhaul...
 
Back
Top