I can only hope Morgan gets the uploading issue under control. It's been pretty frustrating.
Originally Posted by Shane in Oz
You probably already know this, but felt oil seals should be pre-soaked in oil, then allowed to drain so that they're damp rather than wringing wet.
The same principle as, you can't wipe up a spill with a dry sponge.
Originally Posted by chaterlea25
Ok adding to the work load ...
Hey, fella, the work load already is heavy enough.
Originally Posted by BSA_WM20
I assume you use 000000 grease on that silk
Infused with MoS2.
Thanks for the comments and suggestions. However, before I was able to read them I finished assembling the crankcases.
I put together a stack of shims that was only slightly too thick and managed to polish a half-thou. or so off each side of each of them to get it down to where the end float was 0.008". I then decided not to use Yamabond but instead to use Permatex Permashield (Hylomar). Dave-NV recommended this sealant a few years ago and I've been using it on my Gold Stars since with great success. Yamabond would have been an excellent permanent solution, but Permatex will allow the cases to be pulled apart again much easier. Not that they'll need to be pulled apart again, but better safe than sorry.
I applied the Permatex, let it dry for 20 min., then bolted the cases together. I used more clamping force than when I tested the shim stack and ended up with 0.008" end float. So, the pieces are starting to go back together at last.
Next is either the cam (hard Cr plating followed by grinding to size) or boring the cylinder. I'll probably start with the cylinder.
Hi MM, I forgot to mention in my last message about shims that the outer diameter of the timing side shims needs to be small enough not to block the oilway from the timing side bush that feeds oil to the flywheel groove
A few years ago I bought a J.A.P engine that seemed to be in really nice condition Paranoia made me dismantle it to check every part The shims on the timing side were blocking the oilway from the bush to the crank !!!!
I forgot to mention in my last message about shims that the outer diameter of the timing side shims needs to be small enough not to block the oilway from the timing side bush that feeds oil to the flywheel groove
Since I already had glued the cases together yesterday I'd be mad at you for only giving me this information today, and I would be even more angry after I took the cases back apart to check and found the shims weren't too large. But, I had checked this myself so all is well.
Your too-late caution prompted me to do something I had intended to do earlier, see if I could inspect the passage with the smallest probe of my borescope. The passage is 3/16" (0.188") while my smallest probe is 0.179" so it fits. Watching the screen while I inserted the probe it looked like a scene from 'Fantastic Voyage'. The "tophat" end of the mainshaft journal partially blocks the straight line passage to the crankshaft, but the passage widens out at that point so there isn't a flow restriction. Anyway, I can actually see the machining marks on the crankshaft (vs. the smooth surface of a shim) as it rotates. It's pretty cool.
Turning to the cylinder, I centered the torque plate on it using the mill to hold the indicator and torqued the four 3/8" bolts to the 25 ft.lbs. that bolts of that size should have. I then installed the assembly in the four-jaw chuck of the lathe, adjusting the tilt ever so slightly to give a runout of less than 0005" for the mounting base as I adjusted the jaws to give less than 0.001" runout at that end. Precision better than this isn't actually needed based on my other measurements of the cylinder and muff. I then wrapped the fins with two heavy bungee cords to damp any vibrations that were tempted to develop. I had planned to use an old inner tube under the bungees but I don't seem to have one. After firmly tightening the jaws I placed the two keys in a baggie with a note reminding me to skim the base when I had finished the boring. That should stop me from removing the cylinder from the lathe before remembering to do this. This will make the base precisely perpendicular to the bore.
I have a 1-1/2" boring bar for the lathe so rigidity won't be an issue. I installed a carbide bit and made sure the bar projected far enough to reach past the far end of the cylinder. I had honed the bit to a knife edge and made a very short cut to be sure it cut without problem, which it did. But, I ran out of time because we had to get across town for a musical performance the granddaughters were in.
When I got home there was a little time before dinner so I did a crude test of the leakage rate of oil through the mainshaft journal. Filling one of the two oil holes with oil, it emptied into the journal in a little over 15 min. A rough measurement of the volume of that oil hole is 0.021 in.3 (0.34 mL). There would be a greater pressure head on the oil when the timing chest is at the proper level, but at a leakage rate of 0.34 mL/15 min. it means approx. 16 mL would leak from the timing chest into the crankcase during a 12-hour overnight stop. The timing chest holds 1/4 pint (118 mL) so a transfer of 16 mL wouldn't upset the Balance of Nature, but it's worth me doing a more careful measurement once the timing chest is attached. There's nothing I can do about the leak rate, but knowing what it is will tell me if draining some fluid from the crankcase needs to be part of the morning ritual.
Last edited by Magnetoman; 02/14/185:08 pm. Reason: added photos
Since +60 is as far as I can go with this barrel I decided it prudent to do a few more measurements before starting to cut. The cylinder is currently slightly tilted with respect to the base and top flanges and I'd like to "optimize" the amount of material removed when I bore it. I could bore it along the current bore axis, in which case I'd have to remove material from the flanges to make them perpendicular. Or, I could bore it perpendicular to the flanges, in which case it would end up with slightly thinner walls on one side near the top and the other side near the bottom. Anyway, having a full set of measurements before I cut would let me decide the best way to approach this. So off the lathe, and on the surface plate.
An hour of measuring this and that, adding a few shims, and double-checking, and it is now back in the chuck, and the keys back in the baggie to keep me from removing it before skimming the base. It would have been OK to have bored it as it had been, but it's now better than OK. And I won't have the queasy feeling I might have forgot some important measurement as metal is flying from the tool bit.
Absent credible cautions (before I bore it, chaterlea25, not after...) I plan to bore it to 0.005" under the final size and hone it the rest of the way. For this I have really coarse 80 grit stones for removing material way too fast and Healy-approved 180 grit for the final finish.
As for the clearance, an Omega tech person responded to my inquiry to say to use a clearance of 0.0024" (0.095 mm). However, the guy who sells the for them for the Ariel club says he has them in several bikes and uses the clearance recommended in the original manual, i.e. 0.002"-0.004" measured front-to-rear at the skirt. So, taking both recommendations to heart, I plan to use 0.005", hoping to trade some possible piston slap for reduced chance of seizing.
This is just my opinion, it is free, so take it for what it is worth.....
With the slightly better quality oil available today vs. 1928......I would go for .0025 - .004 range. I have never liked starting loose, I can make them loose on my own. With some care on start up/break in, i don’t feel you are going to have issues with seizing.
Offline chaterlea25 suggested 0.003-0.004". He further made a suggestion worthy of Solomon if seeing of there was a compromise value that would allow the Gardini to be used as a backup. Many careful measurements ensued. The Gardini is 0.0014" larger than the Omega so if I use +004 for the Omega, and later needed to replace it with the Gardini, the Gardini would have a +0026 clearance. That falls inside the bounds, but would be even further inside the bounds after running a glazebreaker through the cylinder.
No matter what the final diameter was to be my plan was to bore it to the diameter of the piston and then remove the final few thou. with the hone. So, today it was time to bore it. The first thing I did was to confirm the cylinder hadn't shifted in the lathe (it hadn't). Then, because I was about to machine an essentially-irreplaceable 90-year old item, I took it exremely carefully. Every step of the way I measured with a micrometer, bore gauge, calipers, and the lathe dial.
As an aside, I have a direct reading dial on my lathe, i.e. it reads the change in diameter directly, not the radius. It's not like multiplying or dividing by 2x is a big deal, but a direct reading dial means there's one less thing that can go wrong.
Based on the size of the bore and use of a carbide bit I set the lathe for 250 rpm and a feed of 0.004"/revolution. Being cautious, my first cut was 0.0025" deep (0.005" on the dial) and I could hear it was an interrupted cut at the top ~1/3 of the cylinder, consistent with the 0.003" offset I expected there. I continued removing the necessary ~0.017" by taking cuts of the same depth every time except the last. I checked the result with the bore gage every time. The combination of the 1-1/2" boring bar and bungees on the cylinder resulted in no hint of vibration. Also, I used Tap Magic Aquacut and it didn't smoke, indicating the cuts I took were pretty gentle.
I used the piston as a bore gage after the final cut and, consistent with my measurements, it went in ~2/3 of the way so the bore is currently 0.0007" under the diameter of the piston. Running the bore gauge from one end of the cylinder to the other there is no wavering in the diameter. It's now up to the hone to remove the remaining ~4.5 thou. of the material.
I replaced the boring bar with a standard bit and skimmed the base to be precisely perpendicular to the bore. This is something wouldn't be possible with a commercial boring machine. Including the paint, I had to remove ~0.004" so my guess is the base itself was out of true by ~0.001-0.002".
I'm glad I made the decision to bore it myself. A lathe is the right way to make something round. It's not the fastest way to make an engine cylinder round, so it's not for commercial shops, but it's the right way.
However, the best part of this is if anything goes wrong with the Ariel at any point on the Cannonball, obviously it will be entirely the fault of Rich B and chaterlea25 for their clearance recommendation.
Last edited by Magnetoman; 02/14/185:03 pm. Reason: added photos
I finished skimming through all my previous posts in this thread and uploading photos where I saw the need for images. I know I missed some places, and probably uploaded at least a few unrelated ones, but at least the thread now consists of more than just words.
No real progress to report on the Ariel the past two days due to having to take care of other things, plus having mis-remembered that I had honing fluid. I discovered what I thought was a gallon of it was a gallon of something else, but thanks to McMaster-Carr it and a few other things are due to be delivered tomorrow. Honing the cylinder is going to be messy, but there's no way around that.
While I could take care of hard Cr plating the camshaft spindles, that's a big enough task that I'm going to wait until the cylinder is done and I have the head apart and, presumably, am waiting for parts for it to arrive once I know what parts I'll need. Meanwhile, we were notified that the official Cannonball route sheet holders are now available so one is on its way to me (cheap at half the price...), and the final installment of the entry fee is due March 1.
I made a bit of progress today. I try to keep the garage fairly neat, although not up to OCD standards, but recent machining had left its mark so I first spent some time recovering. Tools and gauges got put back, fixturing replaced, and chips vacuumed up. However, I left the cylinder in the lathe. I don't intend to machine it any further, but until I need the lathe for something else, or it's time to hone it, leaving it in place keeps my options open as long as possible.
Rules for the Cannonball call for rear view mirrors on both sides and a red reflector on the back, so today I added a second mirror and the reflector. Both were straightforward but work like this still takes time. The handlebars are 1" whereas the mirror brackets I had were for 7/8" so it took a few trips back and forth to the vise and file before I had it installed in a location where it wouldn't interfere with the tank (I have both mounted under the bars, not above). Similarly, the reflector required making a simple bracket and painting it so it took time as well.
The route sheet holder arrived in the mail today. My initial impression is I'll have to machine my own bracket to hold it out of the way of all the other clutter on the handlebars.
The honing oil was delivered late afternoon so tomorrow is going to be a messy day. Today I decided to mount the route sheet holder. It has to be mounted at some point, and doing so now gets it out of the way.
It turned out to be easier than I expected thanks to the Ariel having a bracket for routing cables that I repurposed. Still, it took time since I had to machine a 3/4"-wide slot drill and tap four holes, and drill and countersink a fifth hole.
The cable bracket is shown in the photograph at the left, below. As can be seen it's made of fairly thin metal and is held against the top link of the forks by the bolt that holds the spring.
The Al bracket I made for the route sheet holder is shown in the other photograph. The piece with the black ball came with the holder and is what allows the position of the holder to be adjusted. What isn't apparent is that the thickness of the Al makes it a slip fit between the fork bracket and the handlebars so the weight of the holder actually will be supported by the handlebars, with the fork bracket just keeping it from slipping out of position. Because of other constraints the two 1/4-20 bolts only can have four threads in the Al bracket so I drilled and tapped the "black ball" for a 1/4-20 cap screw that is countersunk in the Al bracket.
I don't have a photograph of the complete assembly yet because I painted the Al black to match the bike and it has to dry overnight.
Good update MM. These small but essential jobs can really eat up time so you are wise to fit them in when you have time in between the bigger stuff.
Obviously I cant see the entire setup (and I did have a look to see if I could find more picture on the web) but that bracket looks quite puny to be retaining the top of the fork spring. I am not saying its wrong or should be changed, I am sure it is as Val Page designed it, I am just making an observation that I would have expected it to be a bit beefier.
that bracket looks quite puny to be retaining the top of the fork spring.
Ah, I see the perspective of the photograph in my previous post is deceptive. The one below from when I was taking the forks apart should help. As this one makes clear, the bracket isn't securing the top of the fork spring, and is only an unwitting bystander trapped by the bolt that holds the spring assembly in place.
A big step forward today. Now sitting in the garage is a honed cylinder ready to install.
Throwing wise advice to the wind, I decided I wanted 0.005" clearance to reduce the possibility of seizing. However, having never bored or honed a cylinder before (other than glaze breaking, which isn't the same at all) I was afraid material might be removed too fast and I would overshoot, so I started with the 180 grit. I soon learned I had nothing to worry about, other than exhaustion, so I switched to 80.
Honing is now by far my most hated job on a motorcycle. I can't even think of what's in second place, but it's a long way back. But, if I ever have to do it again please remind me to bore the cylinder a little bigger than I did this time so the hone has less to take off.
I clamped the cylinder in a 3-jaw chuck that I have mounted to a base to hold items vertically on the mill and placed it in a pan on the floor so I would be standing over it. That was a mistake. Next time I'm mounting a cylinder horizontally at a height that allows me to sit. By the way, I knew there would be a lot of grit, but it's a no-name chuck that isn't appropriate for high precision work and I would be thoroughly cleaning it when done.
I used a Lisle hone driven by a 1/2" air-powered drill. I'd never used that drill for more than 5 min. at a time before so something I hadn't experienced is when you run it long enough the air expansion gets it pretty cold.
I would hone for a while, stop to slather on more honing fluid, stop a while after that to measure the progress (or lack thereof) with the bore gauge, slather on more fluid and continue. Repeat ad nauseum.
After what seemed like hours the bore had increased by 0.004" so I switched to the 180 stones. At that point I noticed the 80s were nearly worn down to their metal holders so it's a good thing there wasn't much more material to remove. After a fair amount of time the bore gauge registered just under 0.005" so I stopped to wash the cylinder in my pail of Gunk/diesel followed by soap and water. I thought with all the grit gone I might get a slightly larger reading but it stayed the same.
The stones in my hone are 5" long and the cylinder is 7" so the center region sees the hone the entire time while the ends don't. However, as the ends of the stones come out either end of the cylinder the pressure on the remaining portion of the stones goes up so it cuts more aggressively (P=Force/Area and the area in contact goes down while the force remains the same to the pressure goes up). This would say that the bore at the two ends of the cylinder and the middle should be a little bigger while the regions around 1/4 and 3/4 should be a little smaller.
Switching to a more sensitive gauge, the maximum difference between the largest bore (at the ends) and the smallest is just under 0.0004", but the difference over the portion of the cylinder that will be experienced by the rings is ~0.0002", i.e. 0.0048"-0.0050"
The day was getting late so I lightly coated the cylinder with oil to keep it from rusting. I left the torque plate in place since I want to spend some more time repeating my measurements. And in case I have second thoughts and decide I'd rather have even larger clearance.
I spent some quality time with the bore gauge today. The cylinder is more, um, cylindrical than my initial measurements yesterday indicated.
Other than the top 1", over the entire rest of the cylinder, " the clearance varies within a range of + or - 0.0002" from a mean value of 0.0043", i.e. 0.0043" +/-0.0002". However, the top 1" tapers larger by 0.0017",
The clearance affects two things, the tendency of the skirt to seize, and the rings to have to flex periodically. With it larger by 0.0012" at the position of the top ring when at TDC it means that ring will have to expand and contract by that amount on each revolution. Which isn't a lot to ask of it. That said, I plan to use a compression plate, which will drop the piston somewhat in the bore, so the amount of movement will be less. However, because I've skimmed some from the bottom of the cylinder I won't know exactly where the piston sits in the bore at TDC until it is assembled.
For present purposes, the top of the upper ring is ~0.2" below the top of the cylinder and the bottom of the skirt is 2.19" below the top of the cylinder. Since the stroke is 3.7402", at the bottom of the stroke the bottom of the oil ring is 4.4" below the top of the cylinder and the skirt is 2.19"+3.74"=5.93" below the top of the cylinder (i.e. ~0.6" above the bottom).
Turning to the skirt, when the piston is at the top of the stroke the skirt will be at a position where the clearance is 0.0043", and at the bottom of the stroke it is 0.0042". In between those positions it makes it as far as 0.0045" and 0.0041".
You might remember that Omega recommends 0.0024", the guy at the Ariel club 0.002"-0.004", and two guys here 0.0025"-0.004". I had decided on 0.005" but the above shows I fell short of that larger figure by ~0.0007". So, I have to decide whether to stay with what I now have, or get all messy again and coax a little more material from the cylinder.
Last edited by Magnetoman; 02/20/183:40 pm. Reason: added graph and corrected typos
Hi MM, I think that the widening clearance at the top was because all the honing was done from that end If you can work form the other end of the cylinder to take the extra 0.0007 out this will minimise the "bell" effect
I think that the widening clearance at the top was because all the honing was done from that end
Actually, 95% of the honing was done from the other end. The bell shape was caused by a rookie mistake (it was the first cylinder I had ever honed). I know what I did and won't make that mistake again (there are plenty of others for me to make instead...).
I honed a bit more today and now have it to the point where I'm declaring the cylinder done.
Since the top of the cylinder was slightly bell mouthed I installed a stop inside to keep the hone from reaching closer than 1" from the top. My goal was a clearance of 0.005", by which I mean the largest diameter of the piston (the skirt) has at least that much clearance everywhere in the bore. Ideally, the cylinder would be perfectly cylindrical when done with that 0.005" clearance everywhere, but all that's actually needed is it be "reasonably" cylindrical and with the skirt never seeing a clearance smaller than that. As the attached graph shows (in red), the clearance at T and B DC after today's work are now both 0.005".
The 0.0054" clearance the piston will have in the middle of its travel is 0.003" larger than the clearance specified by Omega so there might be some piston slap. But, I'm more than willing to trade some of that for seizure resistance. Also, the rings will see the bore decrease in size by 0.0008" and then increase again by 0.0007" as the piston travels downward, so the endgap will decrease and then increase by π times those amounts, or ~0.0025" each revolution. This also will result in each point on the rings moving in and out of the piston lands by ~0.0004". These motion amounts don't seem problematic in light of how much the rings move in a well-used bore, where the wear is larger near the top of the stroke.
Those of you with good memories might remember that last July I made a torque plate, the need for which a few readers were skeptical of. So, was making it a waste of time and steel? After boring and honing I measured the cylinder to be cylindrical, which it certainly should have been. I then removed the torque plate and remeasured. Approximately 1/4" below the top flange, where the upper ring will reach when at TDC, the cylinder is oval by 0.0011". Or, equivalently, had I not used the torque plate the top of the cylinder would be distorted by this amount after the head was bolted on. Since the rings would have a difficult time accommodating themselves to this much distortion, at the point where the gas is hottest and pressures is highest, the rings likely would have leaked hot combustion gasses past them.
While the distortion is 0.0011" 1/4" below the top flange, it reduces to 0.0002" by 1/2" and is gone by 1". From here forward all my rebuilds will have torque plates, even if the engine is a century old.
I hadn't painted the bottom flange after skimming it a few days ago so a thin coat of primer is drying overnight. Tomorrow I'll hit it with a thin coat of gloss black and the cylinder will be ready to install (after the black dries).
A good read and some very useful tips on undertaking a long trip (4,100km in that case) on one of these bikes. I am pretty sure there must be lots of things that you can pick up from their trip that you can use on the Cannonball. It seemed like the bike was in much better shape for the first half of the trip however, now I have read that thread, I am sure your preparation along with being able to see how one of these bikes performed on a long trip will ensure that you Black Ariel will perform well throughout the Cannonball.
thread on the AOMCC site that was linked from the VMCC site.
John, I saw the link on the VMCC site this morning and, indeed, there were a number of useful tips in that thread (that I noted in a post on that site after reading the thread). Several times the rider mentioned how much he like his Airhawk seat cushion so I looked on Amazon for one that might fit. It looks like their Dual Sport model might but, unfortunately, the reviews are evenly spread from 1 to 5, i.e. there wasn't just one dissatisfied customer, there were many. So, given the ~$100 price, I decided not to buy one just yet.
With the black paint drying on the cylinder (for another ~20 hours) I turned to the head. Not surprisingly, both of the guides and both of the valves have seen better days, with the valves rattling around with ~0.01" clearance. Interestingly, the intake and exhaust valves and springs are identical. Thankfully, the seats aren't recessed and appear to be in good condition.
The stems of both valves are scratched and show signs of galling so I can't tell what their original ODs were. Stem diameters range from 0.3343" at the smallest to 0.3402" at the largest, which makes 11/32" (0.34375") the closest possible fraction. The nearest "reasonable" metric dimension, 8.5 mm, is smaller than the largest of these at 0.3346". The present valves are 4.05" long (~103 mm), have heads 1.767" in diameter (44.9 mm), stems ~0.337" dia., and faces cut at 45o. However, the seats are 1.75" OD (0.125" wide). Anyway, while the actual head is Imperial, the wear on the items is enough that I can't tell if the present valves and guides also are Imperial or are repurposed metric components.
Goodson supplies reamers for cast iron guides in increments of 0.001" over the range 0.340-0.348", implying new or reground valve stems can be found in all those sizes. However, no matter what, I'll have to install new guides so I'll get to ream them to whatever size of valve I can find that has the correct length and head that matches the seats. The guides are ~0.626" OD, with the intake 2.15" long and the exhaust 2.40" long. Both project outside the head by ~1".
There are two springs on each valve, both ~2" long. They have different diameters to fit inside each other, and use different size wire, but both turn out to have the same spring constant of ~68 lb./in., i.e. equivalent to having a single spring of ~136 lb./in.
So, the hunt is on (or, it will be in a couple of days) for valves and springs. Since they are of the same size I'll be looking to put exhaust valves in both positions. In case anyone has good leads, the valves can be Imperial or metric as long as they have the following dimensions: Valves Imperial head: 1.75" length: 4" stem dia.: ~5/16"-23/64" (0.3125"-0.359") seat angle: 45o
Metric head: 44-45 mm (1.732"-1.772") length: 100-105 mm stem dia: ~8-9 mm (0.315"-0.354") seat angle: 45o
Springs: Inner 3/4" ID 1" OD 2" long ~65-70 lb/in.
Outer 1.1" ID 1.39" OD 2" long ~65-70 lb./in.
Last edited by Magnetoman; 02/23/187:37 pm. Reason: corrected the decimal value for 11/32"
There are two springs on each valve, both ~2" long. They have different diameters to fit inside each other, and use different size wire, but both turn out to have the same spring constant of ~68 lb./in., i.e. equivalent to having a single spring of ~136 lb./in.
Assume that would be to avoid spring resonance leading to valve flutter.
Diesel valves are the usual fall back for turning down to fit old bikes as they are both larger in size and can take the heat of an air cooled engine but I assume you would already be aware of that.