ESE's works engine tuner

[Vannik]

Frits, what is the actual problem with the 50cc racers crank reliability.
We run the 125 KZ engines to 14800 all day, and the basically same configured OKN 125 direct drive engines are rev limited at 15,000 with no reliability issues at all.
That level is of rpm with a 54.5 stroke is 27M/s, with a 40 stroke in a 50cc its only 20M/s.

I have the same question.

I do know that things do not scale linearly with engine size, maybe the issue is the big-end bearing inertia? And in a 50cc it is a much bigger percentage than on a bigger engine. The mode of failure is that the rollers start skidding as they are accelerated and decelerated per revolution. Maybe the big end pin on a 50cc has too big a diameter. That leads to a bigger cage and more inertia. I have never done cage acceleration value calculations but maybe that, more than mean piston speed is the real differentiator?

[koenich]

Gents, you are comparing high quality TM race parts developed as a package since ages for a specific race engine to Derbi aftermarket stuff. While you certainly can buy nice 85/90 mm conrods or good pistons none of this was ever developed as a package to achieve 24+ HP...

[Flettner]

You are all in trouble now, retired as of this week, parts are being sourced as we speak (read) for the 'ultimate' 100cc twostroke. Haha, I'm full of shit, but I have retired and 'we' are building a 100cc twostroke. Based loosely around a TF 125, very loosely. (And legally)

[F5 Dave]

Congratulations Neil

[Wos]

I remember when I was at Orbital and we were initially dealing with Mercury, this being prior to the Optimax engine. The engineers were talking about the Rouen 24 hour boat races. To prevent both seizures and failed bearings, they filled the carb float bowls with straight petrol so the engine would start without loading up. Then, after starting, the fuel was fed from the tanks…..this was on an 8:1 ratio. Oil is good (and now expensive).

Some of the aftermarket expensive 50cc " racing" cranks are very pure in quality...pure shit

[lohring]

We ran our 26 cc, 28 mm stroke Zenoahs up to 25,000 rpm during testing. That's around 23 m/sec. It was only for a short time on the water, though. Big end bearing life is limited at racing with 20,000 rpm maximum for a few seasons, even with over 16:1 oil mixes. Ask me how I know.

Lohring Miller

[Frits Overmars]

I do know that things do not scale linearly with engine size, maybe the issue is the big-end bearing inertia? And in a 50cc it is a much bigger percentage than on a bigger engine. The mode of failure is that the rollers start skidding as they are accelerated and decelerated per revolution. Maybe the big end pin on a 50cc has too big a diameter. That leads to a bigger cage and more inertia. I have never done cage acceleration value calculations but maybe that, more than mean piston speed is the real differentiator?

50cc big end pins are Ø15 or Ø16 mm; that is roughly equivalent to 22 mm for a 125cc engine. The RSA had a 24 mm big-end pin, which posed no problems; an RSA crankshaft survived a full season on the test bench. Wob, what is the big end pin diameter in the KZ engines?

What does fail on the 50cc engines, are poor-quality big end and small end cages and poor-quality axial thrust washers.
My mate emot.nl is concentrating on developing more reliable stuff for the Freetech racers. But I don't think they are going to rev to 20.000 rpm any time soon. Some things, like boundary layers, are not scalable which means the smaller the engine, the harder it becomes to create sufficient angle.area.

[Wos]

50cc big end pins are Ø15 or Ø16 mm; that is roughly equivalent to 22 mm for a 125cc engine. The RSA had a 24 mm big-end pin, which posed no problems; an RSA crankshaft survived a full season on the test bench. Wob, what is the big end pin diameter in the KZ engines?



What does fail on the 50cc engines, are poor-quality big end and small end cages and poor-quality axial thrust washers.
My mate emot.nl is concentrating on developing more reliable stuff for the Freetech racers. But I don't think they are going to rev to 20.000 rpm any time soon. Some things, like boundary layers, are not scalable which means the smaller the engine, the harder it becomes to create sufficient angle.area.

Wobbly and Frits,

Not Diameter of Bore ore Bolt Big end, some use 17mm..."quality " of components ...foregive me if i dont want to talk about to much ... dont use these "racing" cranks out of the box ...


Liedolsheim shows if you did everything for reliability

Grüße

Wolfgang

[wobbly]

I believe three things were in play with the KZ engines. Firstly was the big end pin diameter, this was like many engines of the 125 size was 20mm ( Rotax was the same ) before 1980.
Quality control of the M cage rollers ( junk ) was atrocious, be the needles from INA or SKF, I used to buy twenty big end bearings and sort the rollers into 4 micron size groups, using a 4 decimal digital micrometer.
Lastly was the oil, this was the period when unleaded fuel became popular and or mandatory, and coincided with many companies producing fully synthetic race oils for two strokes.
Reliability , even early on was marginal, only pulling just over 13,000 on a regular basis.
The fix came in three steps as well.

Big end pins went from 20mm solid to 22mm hollow, this considerably increased the cantilevered beam strength.
The cages gradually changed from the M style to flat dividers between the rollers, this increased the cage strength, and prevented bad wear of the rod bore inner from the narrow " hoop " contact
area on each side - and the rollers specifically to be used in big ends were factory sorted into size groupings.

Lastly was a change in oil base stock. The full synthetics were OK when used in unleaded race engines at the time, as this fuel, very unlike leaded high octane race gas, made more power when run rich, with more timing
and less compression.

What was discovered in the two stroke world ( unlike the current on line shit storm it creates ) was that the results gained from the Timken/Falex oil test rig held true in the 2T race engines.
On that rig a pure synthetic can have a load to fail film strength over ten times that of a castor based oil, but the instant the film is broken the synthetic " burns " as it breaks down into its chemical constituents
that have no inherent lubricant properties.
This then resulted in the spinning test drum surface, being badly torn up at the point load, ball contact area.
A Castor oil " fails " at a much lower load on the ball, but when the oil film is broken, the wear generated is vey smooth and shiny, as the base chemical structure is still a good lubricant.

Thus many semi synthetic, ie those with a castor added to a synthetic base stock began to be produced. The Castor being very cleanly refined to prevent the old issue of burnt combustion carbon.
ELF 909 ( as used in GP by Aprilia ) Maxima 927, Vrooam, Castrol A747 etc were found to give superior performance ( both power wise, and wear reduction ) when used at very high ( 0ver 600*C ) EGT numbers.

I tested a huge range of oil ratio's and oil types for a customer using the KT100 air cooled engine up at 650*C.
The short of the result was that 20:1 gave more power ( more wasn't better ), the semi synthetic Elf 909 and a couple of others, gave more power and much less piston/bore wear in identical run times.
The popular Motul 800 fully synthetic made the least power, and gave the most wear - by a big margin.

Thus all these factors combined at about the same time, and the results enabled full reliability well past 14,000 in the karting world, from 100cc air cooled to 125 50 HP KZ's.
Specifically that for a period even the flat cage , larger diameter pins would still wear away the silver, and turn blue - alot later, but as soon as we switched away from fully synthetic oils such as Motul 976
the whole issue disappeared.

[pete376403]

Honda (4 strokes, sorry) were running tiny bore and stroke roller crank engines 125cc fives, 250cc sixes) at 18 - 20,000 rpms back in the 60s. I believe they ran castor based oils exclusively.

[Wos]

I believe three things were in play with the KZ engines. Firstly was the big end pin diameter, this was like many engines of the 125 size was 20mm ( Rotax was the same ) before 1980.
Quality control of the M cage rollers ( junk ) was atrocious, be the needles from INA or SKF, I used to buy twenty big end bearings and sort the rollers into 4 micron size groups, using a 4 decimal digital micrometer.
Lastly was the oil, this was the period when unleaded fuel became popular and or mandatory, and coincided with many companies producing fully synthetic race oils for two strokes.
Reliability , even early on was marginal, only pulling just over 13,000 on a regular basis.
The fix came in three steps as well.

Big end pins went from 20mm solid to 22mm hollow, this considerably increased the cantilevered beam strength.
The cages gradually changed from the M style to flat dividers between the rollers, this increased the cage strength, and prevented bad wear of the rod bore inner from the narrow " hoop " contact
area on each side - and the rollers specifically to be used in big ends were factory sorted into size groupings.

Lastly was a change in oil base stock. The full synthetics were OK when used in unleaded race engines at the time, as this fuel, very unlike leaded high octane race gas, made more power when run rich, with more timing
and less compression.

What was discovered in the two stroke world ( unlike the current on line shit storm it creates ) was that the results gained from the Timken/Falex oil test rig held true in the 2T race engines.
On that rig a pure synthetic can have a load to fail film strength over ten times that of a castor based oil, but the instant the film is broken the synthetic " burns " as it breaks down into its chemical constituents
that have no inherent lubricant properties.
This then resulted in the spinning test drum surface, being badly torn up at the point load, ball contact area.
A Castor oil " fails " at a much lower load on the ball, but when the oil film is broken, the wear generated is vey smooth and shiny, as the base chemical structure is still a good lubricant.

Thus many semi synthetic, ie those with a castor added to a synthetic base stock began to be produced. The Castor being very cleanly refined to prevent the old issue of burnt combustion carbon.
ELF 909 ( as used in GP by Aprilia ) Maxima 927, Vrooam, Castrol A747 etc were found to give superior performance ( both power wise, and wear reduction ) when used at very high ( 0ver 600*C ) EGT numbers.

I tested a huge range of oil ratio's and oil types for a customer using the KT100 air cooled engine up at 650*C.
The short of the result was that 20:1 gave more power ( more wasn't better ), the semi synthetic Elf 909 and a couple of others, gave more power and much less piston/bore wear in identical run times.
The popular Motul 800 fully synthetic made the least power, and gave the most wear - by a big margin.

Thus all these factors combined at about the same time, and the results enabled full reliability well past 14,000 in the karting world, from 100cc air cooled to 125 50 HP KZ's.
Specifically that for a period even the flat cage , larger diameter pins would still wear away the silver, and turn blue - alot later, but as soon as we switched away from fully synthetic oils such as Motul 976
the whole issue disappeared.

wobb

Quality of Parts! tolerances ... roundnes ...axis...diameter...parallelity

Todays motul 800 for street/racing is often used in bikes of winners...sems not to be main problem

[wobbly]

The only help with tolerances etc I can give I learnt from Curt at Hot Rods, the big end clearance should be 1 micron per pin mm ie 20mm pin = 0.02 clearance.
When I spent hours measuring everything I would stack up the pin, rod bore and roller diameter to achieve this.

And any roller bearing surface must be Rc 58 min to a depth of 0.5mm.

Re winning with Motul 800 - if thats the case then they are not pushing the mechanical/thermodynamic limits, but in every case they are leaving power on the table.

[ApolloMotoMoto]

While we are on the subject of cranks and "reliability" ....

These are "motorized bicycle" cranks for a 90cc single speed direct drive kart track race bike engines.

42 stroke x 52 bore

Air Cooled running VP MRX02 (converting to E85 for next race event....)

Cylinder Reed induction with Minarelli AM6 reed's.

Portmap designed in JanBros with balanced blowdown/transfer STA for peak torque at 9,500 to match the pipes being used (any 80-85cc MX pipe you can find cheap on eBay...)

BMEP prediction in JanBros is 12.9; BHP prediction 24.6...

Based on how they run I would expect to see 16-20 RWHP on a dyno.

Spinning up to 13,800-14,200 RPM, but even one example engine with an overly long tuned length pipe which taps out at 12,250-12,500 (regardless of jet tune or ignition map adjustments) is also snapping these cranks, so it doesnt appear to be terminal RPM piston speed in my case that is the problem.

These cranks use a 16mm OD crankpin and ....15mm OD..... output/drive shaft.

They are mold-forged from "forty pound steel" (40cr?) a rather common chinese grade of steel; they come out of the forge mold red hot and are near immediately dropped into a water quench

Final hard machining occurs after this.

There is ALSO a custom crankshaft manufacturer who produces essentially this exact crank starting with 4140 prehard material, machining them in the hard state.

I am aware of exactly 1x of these cranks which has suffered the EXACT carbon copy crank output shaft failure. I call it a snap, but the clearly visible "beach lines/tide lines" on the face of the break zone clearly show that its a crack which progresses through the material over time;

And the brown/orange oxidation/corrosion on the outer perimiter of the break zone clearly shows it has been cracked open long enough for moisture to creep in and start corroding/oxidizing the steel.

And again, the 4140 prehard crank have 1x known failure that looks EXACTLY like the close-ups attached to this post...

The break profile has a novel "dimple-and-dome" or "crater-and-dome" shape where the crankwheel face is dimpled/cratered and the broken shaft end is domed.

At first I thought this confirmed that the ...horrible... heat treat procedure used on the stock chinese crankwheels was the "problem"; but the failure occuring in the 4140 prehard cranks, which looks exactly the same, suggest maybe thats not the root of the cause.

I dont know if this particular vendor of aftermark cranks made from 4140 prehard is doing any post-process hardening which might explain the carbon copy break profile, but its interesting nonetheless.

So, "what gives"?

Have I just exceeded the yield strength of 15mm shafting with the given specific power output?

[ApolloMotoMoto]

Pics of the dead 4140 prehard crank.

[Wos]

The only help with tolerances etc I can give I learnt from Curt at Hot Rods, the big end clearance should be 1 micron per pin mm ie 20mm pin = 0.02 clearance.
When I spent hours measuring everything I would stack up the pin, rod bore and roller diameter to achieve this.



And any roller bearing surface must be Rc 58 min to a depth of 0.5mm.

Re winning with Motul 800 - if thats the case then they are not pushing the mechanical/thermodynamic limits, but in every case they are leaving power on the table.

Good question where the limits are in diffrent Races

Liedolsheim 8H wide open Throttle...think the limits are diffrent to freetech50

Maybe the Motul 800 changed, since you have testet it?