Showing posts with label Engine and Propulsion. Show all posts
Showing posts with label Engine and Propulsion. Show all posts

Saturday, 22 April 2023

Clutch cone hone, fixing the BANG BANG BANG


Update, April 2023- 10 years later I noticed some clutch slippage so re-honed the cones and replaced the seals.  Pics and hasty write up below.   It is really helpful to make a fixture by hole-sawing a 2x6 to hold the shaft in place.  
You’ll need a puller, pin punch, appropriate sockets, lapping compound  torque wrench, RTV gasket compound, solvent and a tub for parts washing,  aerosol parts cleaner, brushes.  An arbour mounted wire brush and buffer is handy for cleaning and polishing the sealing surface on the output flange.  
You definitely want the yanmar shop manual. Read the notes!
One thing it doesn’t say is to switch the output shaft nuts- inside for outside - and you’ll get a fresh flange for staking the nut to the shaft.  
I can now do this in less than 2 hours with around $30 in parts and materials.
When done, Remember to adjust the shift linkage cover per the manual prior to reinstalling.  
Before starting, note the position of the shift lever on its shaft.  It is not keyed, and it’s position impacts the position of the control lever.  

As removed.  make sure the transmission is clean and the oil has been drained.
First step, remove the dipstick, gear selector and output shaft flange.   You need to bolt a bar to the the flange for this to prevent it turning.   don't forget that the 27mm flange nut and the identical internal shaft nut has a reverse thread.




Second step, remove the hex screws securing the transmission case to the bellhousing.   They are all the same length.  You will remove the output shaft (big one with the gear on the left) and leave the others in place.

Note that there's no gasket, just a very thin layer of RTV sealant.   Clean the mating surfaces carefully, and dont let any crud fall into the transmission housing.  
you can see here how the shift lever mechanism engages the groove in the clutch cone.  It moves the clutch cone into either the forward or reverse gear, where the cone is received by its mate (inside the gear) and either slips, or doesn't. 
The flange is lagged onto a 2X6" fixture to serve as a stand for the next steps.




Shaft inverted and held in place for briefly working from the other side.   No torque needed, so this jury rig is fine.   
Puller pulling.
Gear stack being disassembled.   The yanmar manual covers this.   Have a seperate tray, labelled, each for forward and reverse parts. 
you can see the glazed interior of the forward gear.  Note also two small centre punch marks I made on the forward side of the clutch cone.  The forward gear itself is marked with an "F".  see below.

Abrasive compound has been applied.   then mate the parts and rotate back and forth.  wipe it out and repeat.   do it again.
Then clean it reallly really well. The deglazed cone is clearly visble on the left, below.


Cleaner than clean and drying.  Get all of the grit off.   Then do it again with clean solvent.


Reassembly.   
output flange cleaned with sealing surface polished, then painted.  Don’t paint the flat surface that mates to the shaft flange.



As mentioned above - don’t forget to adjust the shifter cover per the instructions in the manual.  I missed this once on a friend’s transmission and I think it caused the come to glaze again prematurely.   (Sorry Terry). 




10 years prior…..


Windstar suffered a worsening vibration - much more extreme than that really - in forward gear.   It started intermittently but progressed to the point that the boat was unusable.    

After paying a pro to misdiagnose shaft, engine mounts and alignment as causes for a terrible banging  vibration I did some homework online and utimately fixed it myself over a weekend for about $30.   Not for the mechanically inept, but not difficult either.   The specifics are hard to find via google, so I copied the post I found at the time and pasted it below.  Unfortunately I can neither crediit nor thank the original author.   

The Post (not mine)

 
The gearbox is quite a small unit and it is fixed to the engine by 8 bolts around the bell housing, 4 bolts on the prop shaft and releasing the control cable, easing the prop shaft gland we were able to slide the prop shaft back a couple of inches sufficient to allow the gearbox to move back and be lifted out. Not having the right tools on board I was able to put the gearbox into a carrier bag and take home with me.

Cleaning an area on the workshop bench, I drained the oil, removed the 4 bolts that held the control lever in place and removed it. The gearbox is accessed by removing the 8 bolts holding the bell housing to the gearbox, the jointing was just liquid cement so having removed the bell housing I could now see inside the gearbox. You need to be careful to protect the seals at each end, there are 3 shafts an input shaft, an intermediate shaft and the one that we are interested in the output shaft, which is recognised by being the largest and has a locknut on each end. Note: these nuts are left handed threads.

You need to clear the locking tab and using a long handle socket or torque wrench to remove the nuts. You will then need a puller to remove the roller bearings and collar (2) and the bearing inner race and collar (2). You should then be able to lay out the whole shaft on to a nice clean surface, clean the individual components and inspect for damage or wear. The main area’s to look at are the bearings and the drive cone. Having said earlier that the reverse gear seemed ok, I was hoping to reverse the drive cone if possible as I felt this was not as important as the forward gear but I found that the gearbox had been opened up previously and that the cone was worn both sides, however there was some grooving left on the cone with a possibility of reclaiming the contact surface.

The cone is moved in and out of the large gearing by the selector mechanism, the tapered surface of both forward and reverse gear were very shiny, possibly were they had been slipping ? I could not find any referance to this in the manual, so I decided to lap the tapered surface’s with some fine grinding paste ( I was surprised what a difference this made ) When fitting the tapered surfaces together they now locked 100 % and I felt that I had done the right thing.

All parts were then meticulously cleaned before starting to reassemble. You will need a piece of tube to knock the bearings and collars back on. If you have been careful you can reuse the seals and the locknuts, when fitted these need to be torqued up to approx 10 kg/m or 70 ft/lbs, check that all is ok before using liquid sealant and refitting the bell housing. Note: A Large vice is very helpful for stripping down and reassembling and don’t forget to relock the locknuts.

Before fitting the control mechanism, use the aperture as an inspection hole to see that the gears are turning and that the cone moves forward and backwards, you should now be able to fit the control mechanism, Check the O ring is ok and that the shifter which actually locates in the centre of the cone is set to its lowest point (it is possible to be 180 degrees out), loosely fit the 4 bolts, hand tight and fit the control lever to the correct angle.
Note : the bolts holding this plate are over size, allowing for movement to be able to adjust the control lever, so as you can set the same amount of lever movement forwards and backwards, when this is correct you can tighten the 4 bolts.

Before refitting the gearbox, check that the damper plate which is fitted to the flywheel is all ok, then you should be ready to fit the gearbox, lightly grease the spline on the input shaft and slide the gearbox back into position, fit and tighten the 8 bolts, you can then adjust and fit the control cable, put in the required amount of gear oil and secure the filler plug. Slide back the prop shaft and secure the 4 bolts, you should then be in a position to carry out operational tests. At this stage I found that I wanted to readjust the control lever movement and I found it was just as easy to remove the gearbox, lift it into the cockpit and do it rather than struggle in the engine locker. My friend then carried out some tests and found a big improvement, so how long had the cone been slipping ? I quite enjoyed doing the job although as you get older it is more difficult to get in and out of the engine compartments, a bottle of whiskey changed hands which was very nice and it saved my friend putting money into someone else’s kitty.

 

Sunday, 28 July 2019

Yanmar 2gm20f fuel seepage

Update


A look at the Yanmar service manual revealed that the assembly in question has an o-ring at its base, which would be the logical source of the leak.  (#2 in the exploded drawing.)




Thankfully, this turned out to be the problem, and after removing several coolant hoses for access, the repair was straightforward.   Unsurprisingly, after 30 years of heat, diesel and pressure, both O-rings were somewhat hardened and no longer round in section.     Total cost to repair less than $10.   There are springs under pressure that are released in this situation however they don't leap to freedom and a reasonable amount of care is all that's required.  There was a small amount of corrosion evident in the o-ring groove, and i cleaned this and greased the area and the fitting's threads, for benefit of Windstar's future owner who will doubtless deal with this again in 2058.

This made a surprising difference in the livability of Windstar's interior.  Doesn't take much diesel to perfume the entire living space.

Update


 - used ye olde Baby powder trick and the leak is at the base of one of the injector lines .  See below.
This basically means degreasing the area.   Once the engine is basically clean (hopefully it always is...) , use aerosol brake parts cleaner to remove any residual diesel film.   This stuff is effective, leaves no residue and dries quickly.  When the talcum is applied it is white and dry and it will be immediately evident when any fluid dampens it as you can see in the photo.  Thus, if you are watching closely you can determine the source of the leak, even if it quickly flows elsewhere.  Handy.







Windstar’s engine is seeping in the area of the high pressure fuel pump.   Diesel collects around the serial number shown in the photo below and slowly migrates down the starboard side of the engine.

Any suggestions as to source and remedy?

Many thanks!


Monday, 24 October 2016

Gori 2-blade prop

My Martec was tired and I knew I wanted a folding prop to replace it.   Pretty happy with the Gori.  Do beware that the normal shaft section where the castle nut is installed may be different in length than a typical North American prop.  (the gears foul the end of the shaft on the gori, there is more space on the martec.   A long story but it took some doing to install this, and some mods to a brand new shaft.
In any case it works great, no issues so far.   Shipped air from Denmark amazingly quickly, great service from the agent in New England somewhere.







Sunday, 4 September 2016

C&C 33-2 water strainer installation error and 2GM issue

Last week Windstar developed a noisy exhaust note, not unlike an unmuffled motor cycle engine.   After a minute or two, and some higher RPMs, it would settle down to its usual quiet gurgle.   Obviously this was an issue with an intermittent absence of cooling water, but why?   I checked underway and the water pump was turning, its hose seemed pressurized.  I also checked the water strainer (as best I could underway) and clearly water was being pulled through.   This left the mixing elbow and the heat exchanger to consider. 
I read over the very well-written shop manual, and went back at the boat where I disconnected the short hose between the heat exchanger and the mixing elbow, connected a length of 5/8" heater hose to the mixing elbow, and blew through it just to check for resistance.  The pleasant gurgle flashed me back to high school, and the ease of blowing showed that the passages were free of serious restrictions.   I put the hose on the elbow exiting the heat exchanger, and started the engine.  No water, checked the water pump and noticed it was now not turning.    Tightened the belt and problem solved. 
I had set it quite loose and it worked fine all season, only now getting glazed to the point that it slipped at low rpm or until it got warm.  I had dismissed the most likely source of trouble - the pump - because I service it every year, and because I had assumed any problem would involved a completely failed impeller.      Anyway, lesson learned...or probably not.   But, that's not the point...  I found some other stuff.

First Issue:
The 33-2 has its raw water strainer mounted in such a way that it cannot be easily checked or cleaned.   See the photo below where I managed to crop out the heater hoses above, but, in order to lift and remove the strainer basket, it was necessary to remove the coolant reservoir, by sliding it upward off its bracket, but in order to that, it was necessary to unscrew the two P clamps securing the hot water hoses.  (you see those in the photo below)

The solution (below) is simply to move the coolant reservoir bracket to starboard, and move the coolant hose p-brackets upward slightly.   There was enough play in the hoses that this was no problem.  


Second issue:  
When removing the hose connecting the heat exchanger to the mixing elbow, the small elbow/fitting on the heat exchanger came off in my hand, despite what appeared to be a securing set-screw holding it in place.   See below.

That screw (the Philips head, clearly visible) engages a groove in the elbow and this holds it and its sealing o ring in place, allowing a secure seal with 360 degrees of rotation.   On mine, the screw was pitted and flush with the bore, and did not penetrate into its receiving groove in the fitting.   Over time it had eroded or corroded, had become effectively shorter and no longer secured the elbow.   In the photo above, if you grabbed that elbow and pulled hard, straight up, the elbow  would lift out of its bore.   When running, coolant pressure is not adequate (I think) to overcome the tension placed on the joint by the hose, however the screw obviously needs replacement.  Yours probably does too.