Saturday, 7 December 2019

Multigrade Oil

Old instruction books frequently tell you which oil to use in the summer and which oil to use in the winter, so that when our engines are hot, we can still retain sufficient viscosity in summer heat, and that when the weather is cold we can still get the oil moving quickly enough to avoid engine or ankle damage from reluctant kickstarters from oil that refuses to flow. Summer weight oils are of a higher viscosity grade than winter weight oils, for a given engine, and yo change the when the weather starts to get cold.

This is obviously not ideal, and in 1948 an American called Bob Lancaster looked to improve the performance of oils in the Ariel Square Four raced by Gus Johnson. 

Gus Johnson

Bob took a polymer from a company called Enjay Chemical, and blended it with SAE 20 oil to boost the viscosity when hot to near that of SAE 50, while retaining the SAE 20 properties when cold. It was so successful that Bob started a company to manufacture and market the oil, and Torco is still trading today.

So we know that the temperature range the oil is exposed to in many vehicles can be wide and that the difference in viscosities for most single-grade oil is too large between the extremes of temperature to provide the lubrication performance we need. To bring the difference in viscosities closer together, as Bob Lancaster showed us, special polymer additives called viscosity index improvers (VIIs) are added to the oil.

These additives are used to make the base oil a multi-grade motor oil. The idea is to cause the multi-grade oil to have the viscosity of the base grade when cold and the viscosity of the second grade when hot, enabling one type of oil to be used all year. The viscosity of a multi-grade oil varies logarithmically with temperature as does a straight oil, but the slope representing the change is less steep. This slope representing the change with temperature depends on the nature and amount of the additives to the base oil.

The SAE designation for multi-grade oils includes two viscosity grades; for example, 10W-30 designates multi-grade oil. The first number '10W' is the viscosity of the oil at cold temperature and the second number is the viscosity at 100 °C. The two numbers used are individually defined by SAE J300 for single-grade oils. Therefore, an oil labelled as 10W-30 must pass the SAE J300 viscosity grade requirement for both 10W and 30, and all limitations placed on the viscosity grades.

Breakdown of VIIs is a concern in modern motorcycle applications, where the transmission may share lubricating oil with the motor - though obviously that doesn't concern those of us with separate gearboxes. For this reason, synthetic oil or motorcycle-specific oil is sometimes recommended.


In addition to the viscosity index improvers, oil manufacturers include other additives such as detergents and dispersants to help keep the engine clean by minimizing sludge build-up, corrosion inhibitors, anti-wear additives and alkaline additives to neutralize acidic oxidation products of the oil. Today, we are very concerned with old Square Four engines that have not had their sludge traps properly cleaned - we don't want modern detergents loosening particles and blocking oilways, so we clean them out on rebuild and use a detergent oil with a filter, to keep the sludge in suspension and trap it in the filter..

It is these additives which start to make owners of classic motorcycles think carefully about their oil. For example, most oils produced today are for modern vehicles and have a minimal amount of zinc dialkyldithiophosphate (called ZDDP, so you don’t have try and pronounce dialkyldithiophosphate) which is an anti-wear additive designed to protect metal surfaces with zinc should they come into contact, which is obviously desirable for our camshafts and followers and you might think that you want a lot of zinc in you oil. The problem is, engineering is full of compromises and one is that the quantity of ZDDP is limited because it damages catalytic converters; other extreme pressure additives can limit the performance of wet clutches (clutches running in engine oil) so we have ‘classic’ oils, manufactured to older specifications which contain the additives we need and reflect that not many classic motorcycles have catalytic converters – but a lot of us, Bantam owners included, have wet clutches. Another win for Ariel separate clutches though!

Other EP additives, like calcium sulfonates, are available for additional protection under extreme-pressure conditions or in heavy duty performance situations. Calcium sulfonate additives are also added to protect motor oil from oxidative breakdown and to prevent the formation of sludge and varnish deposits.

Some additives contain molybdenum disulfide which reduces friction, bond to metal, or have anti-wear properties. MoS2 particles can adhere to steel surfaces and some Trabant engine liners were even treated with MoS2 coatings during manufacture to reduce friction.


Now you might be thinking this is all a bit bewildering. OK, I understand viscosity, but how do I know which additives are good and which are bad, and how do I know whether the supermarket 20w/50 I just bought has them in?

Engineers to the rescue!

The American Petroleum Institute (API) governs most of the activity in the industry which has paid my wages for the last 25 years. In addition to the SAE classifications for viscosity, we have API specifications for oil performance

The API standard includes alphabetical ‘service classes’ for lubricants, which really refer to the ability of the oil to perform a given task (like lubricating a sliding tappet against a camshaft, for example). These have two general classifications: S for "spark ignition" and C for "commercial/compression ignition" (diesel engines to you and me). Engine oil which has been tested and meets the API standards may display the API Service Symbol (also known as the "Donut") with the service designation on containers sold to oil users.

This started in the 1930’s with the SA classification; today we are up to the ‘SN’ standard. The SN standard refers to a group of laboratory and engine tests, including the latest series for control of high-temperature deposits. Current API service categories include SN, SM, SL and SJ for petrol engines. All previous service designations are obsolete, although motorcycle oils commonly still use the SF/SG standard reflecting the need to control EP additives to protect our clutches and our camshafts.

Since the SG standard, all categories for petrol engine lubricants have placed limitations on the ZDDP content due to the effect ZDDP has on catalytic converters. In API SG rated oils, this was at the 1200-1300 ppm level for zinc and phosphorus, where the current SM is under 600 ppm. This reduction in anti-wear chemicals in oil has caused premature failures of camshafts and other high pressure bearings in many older automobiles and has been blamed for pre-mature failure of the oil pump drive/cam position sensor gear that is meshed with camshaft gear in some modern engines.

Synthetic oils

Synthetic lubricants were first synthesized, or man-made, in significant quantities as replacements for mineral lubricants (and fuels) by German scientists in the late 1930s and early 1940s because of their lack of sufficient quantities of crude for their (primarily military) needs. A significant factor in its gain in popularity was the ability of synthetic-based lubricants to remain fluid in the sub-zero temperatures of the Eastern front in wintertime, temperatures which caused petroleum-based lubricants to solidify owing to their higher wax content. The use of synthetic lubricants widened through the 1950s and 1960s owing to a property at the other end of the temperature spectrum, the ability to lubricate aviation engines at temperatures that caused mineral-based lubricants to break down. In the mid-1970s, synthetic motor oils were formulated and commercially applied for the first time in automotive applications. The same SAE system for designating motor oil viscosity also applies to synthetic oils.

Higher purity and therefore better property control theoretically means synthetic oil has better mechanical properties at extremes of high and low temperatures. The molecules are made large and "soft" enough to retain good viscosity at higher temperatures, yet branched molecular structures interfere with solidification and therefore allow flow at lower temperatures. Thus, although the viscosity still decreases as temperature increases, these synthetic motor oils have a higher viscosity index over the traditional petroleum base. They're specially designed properties allow a wider temperature range at higher and lower temperatures and often include a lower pour point. With their improved viscosity index, synthetic oils need lower levels of viscosity index improvers, which are the oil components most vulnerable to thermal and mechanical degradation as the oil ages, and thus they do not degrade as quickly as traditional motor oils. However, they still fill up with particulate matter, although the matter better suspends within the oil and the oil filter still fills and clogs up over time. So, periodic oil and filter changes should still be done with synthetic oil; but some synthetic oil suppliers suggest that the intervals between oil changes can be longer, sometimes as long as 16,000-24,000 km (10,000–15,000 mi) primarily due to reduced degradation by oxidation.

Friday, 6 December 2019

What's that ominous noise?

Last weekend, my Sunday morning jaunt ended after 1.5 miles with a screech/rattle (strange sound in between the two) that has appeared while accelerating up a hill. I think it's transmission - out of gear the engine appears to be fine. Turning the rear wheel by hand with the clutch out and the bike in gear sounds rattly - I think I will have to take a look at the clutch. The noise changes with clutch operation, less harsh perhaps and appears to be related more to road speed than engine speed.

Of course, as I write this I sound like I have a misplaced confidence that no Square Four owner has any right to - you are always worrying that it is going to lift it's skirts above it's knees and show you a connecting rod when you were least expecting it. This time, I arrived home with the engine ticking over sweetly and it continued to do so when I parked up. 

Running it on the stand, I couldn't replicate the sound, but I did notice the rear wheel is a lot more wobbly than it was when I built it.

A week or so later, after some busy time at work and contemplating that I might get the CX500 on the road, I found a bit of time to pull it apart. It's worth remembering that older bikes are often easier to fix than newer ones - my Square is a lot less complex than the Honda and it's much easier to work on.

All I need is this handy kit that I found in the bike's toolbox:

Or - not. Falling at the first fence, I found that I did not have a 3/8" AF box spanner in the tool roll. This odd ball size fits not only the clutch centre screws but crucially the exhaust pipe clamps - so I couldn't even start the job. I made a note to buy one for the tool roll.

With the pipe off, I slackened off the brake pedal and disconnected the brake light - that lets you remove the clutch dome. I might see if I can get a better shine on that patina:

With the right tools, the clutch plate came off perfectly:

And it revealed a loose clutch centre nut, just as had been predicted by the AOMCC forum panel of experts. The nut was loose but still in place, but it probably wouldn't have come loose at all if I had been a bit more thorough with the tab washer:


The clutch plates are getting thin, but I have no wish to re-cork them right now, nor any cork; they need pulling out with this bit of hooked wire I keep in the tool box:

Slightly sticky. That's grease from the centre bearing:

Here it is again - we will have to be sparing with it on reassembly. The splines are no worse than when they went in back in 2014:

And again. Those screws are in place but very loose:

As can be seen in this video:

Thankfully, all looks well under the primary chain cover:

The shock absorber nut is exactly as it was when I fitted it back in 2014:

Sprocket looks fine:

Having removed the chainwheel and carefully removed its bearing (with the twelve needle rollers) I set about removing the rear half of the chaincase. When it wouldn't come free, I remembered that the rear upper chainguard is bolted to the chaincase and that the bolt lives under the battery. This has not been off for ages as it has been no trouble at all:

Here's the bolt I'm talking about. You have to remove the coil as well:

Once behind there all looked fine & dandy:

Oil filter head looks OK:

At close of play, the rear chaincase was back in place, sealed with Threebond 1215:

The next day I tensioned the primary and put the chainguard and the coil back in place, and the following day I put the primary case back together:

Next came the clutch rebuild. This time I paid much closer attention to assembling clutch basket to the sprocket, and the centre to the mainshaft retaining both with Loctite 243.

The mini-lathe is a useful beast. Here I'm turning up a new spacer to fit between the primary chaincase inner half and the frame in stainless steel. It's 1/2" long, 11/32" ID for clearance on a 5/16" bolt and 3/4" OD:

This is the truly awful side stand. I never use this, as it is far too short and consequently precarious. The tie wrap is there to make sure the spring cannot become disengaged from the lug on the leg, which would leave the leg flopping about:

Assembling the battery carrier and it's spacers:

And all back together.

Wednesday, 20 November 2019

CX500 - almost there

As the CX edges closer to being rideable again, I have lost my daily ride - the SQ4 clutch is making a strange shrieking rattle that sounds like a loose centre so I am bikeless at the moment.

On the upside, the CX has its radiator filled - I have replaced the O rings I chewed up on the coolant pipe:

I've also discovered that Cillit Bang makes a great cleaner for the old Japanese Aluminium - look at the rear engine cover above the starter, and the coolant pipe clamps:

Coming back to the radiator, the captive nuts that Honda used to retain the front radiator grille to the shroud are knackered, rusty, and cost £4.50 each. I have these rubber M5 expanding nuts from eBay which fit the 9.5 mm holes beautifully (with a smear of silicone grease)

Here's one

I've discarded the original M5 screws and replaced them with stainless cap heads & washers. In this mock up, you can see how they expand the rubber nut to retain the front grille:

I've shined up the grille with a smear of silicone grease:

Moving forward, the calipers have been removed to have their Helicoils fitted:

I've also fitted the front wheel to get it out of the way while I wait for the Helicoils to come:

That's all for now - hopefully I can repair the calipers later this week.

Friday, 15 November 2019

Let's fix those duff threads

Updated: First published January 2014

Well, I knew I shouldn't have looked too closely at those calipers. I replaced one nipple successfully:

The other one is too far gone and I am going to need to Helicoil it. Time to resurrect an old post that I published a long while ago:

So you've all seen how the trial fit up of the primary cases revealed some duff threads that I really should have seen earlier - well now we have to fix them.
I bought this nice little set from Chronos:
It  comes in a nice little steel box with all the tools you need and 20 or so helicoils, for less than £15 delivered. These are 1/4" BSW.
Here are the drill, tap & a helicoil insert:
Here's the insert tool, which has a depth stop collar supplied with a hex key to adjust it, and the punch for knocking out the tang:

The first step is to drill out the hole to the correct tapping size for the outside diameter of the insert. It is vital to use the correct drill bit size, as the outer thread of the helicoil will match the newly tapped hole, and will dictate the size of the internal thread such that it matches your fastener.

You need the screw to ultimately go into the work at the proper angle, so if you can, set the work up in a drill press.

Next you have to cut the thread for the outside of the insert using the tap that is provided with the kit. The tap has a short taper (called a 'second' tap) to help get the thread started; use a tap wrench (90 degrees forward, 120 degrees back to clear the swarf) with some lubrication to cut the new thread. Cut the thread until you no longer feel any resistance as this means the hole is now completely tapped with the new thread. Gently wind out the tap with the wrench. Make sure you don't try to bend it at any time - you really don't want to have to deal with a broken tap.

Next take one of the helicoil inserts and put it over the end of the installation tool, with the tang of the inset in the slot and in what will become the bottom of the hole - so the tang goes in the hole first. Set the depth stop so the insert will end up 1/2 a pitch below the surface.

Wind it in. You shouldn’t feel any resistance as you do this but if you do you may just need to clean the thread up a little. Note that as you get near the bottom of the hole and the friction increases you will feel the insert start to wind up, spring like, against the tool.

One the coil is in place you use the little punch that should come with the kit to knock the little tang off the end of the helicoil that assisted with winding it in. If you are rethreading something relatively thin, as is the situation with this primary chaincase, you can grind off the spare length of insert with the Dremel:

Job done:

Monday, 11 November 2019

Plastic Nipples...

Way back in the '70's, Honda used these horrible plastic sleeves around their cable nipples to reduce friction as the nipple turned in the control levers. All well and good, I hear you say and it is so.

The sleeve slips around the nipple, and the nipple swivels nicely in your clutch lever without you developing that gorilla grip. All fine, unless you repeat the application on the other end of the cable in a claw arrangement like this:

Alright, I know someone has mangled it but it wasn't much better when it was new. Guess what that claw does to those nice plastic sleeves? Mangles them as well!

Here's my solution. The nipple is 6 mm OD x 10 mm long, and the sleeve is 8 mm OD. I've used the mini-lathe to turn these brass sleeves to fit, and added a little notch with a needle file to accommodate the cable:

Here's how they fit. This is just a test at the handlebar end - the plastic sleeve will be staying here.

Here's the engine end - the two halves are fixed with a high-viscosity cyanoacrylate:

Sitting neatly in the claw. I'll have to remove the clutch cover to straighten that out:

There's a lot more on cables and cable making here

Change of subject - a bit of painting. The chrome on this coolant pipe was past it's best and missing in a few places, so it has had a coat of UPol wheel silver:

As you might have gathered, I am aiming to get this bike roadworthy as a winter hack at minimal cost.

This is not how you fit O rings...