1 Steam Cars : Phorum
General Steam Car topics
Connecting Rod Roller Replacement Re-visited
Posted by: Howard Randall (IP Logged)
Date: November 18, 2002 10:08PM
<HTML>I was not sure that folks went back and checked old topics, so I have re-posted this topic with a question for Pat F.
Pat, what supplier has individual bearing rollers?
Re-doing bottom end on 30 HP. Races are fine, rollers are not! Not ready to take the plunge for new modern (self aligning) bearings!
Thanks!</HTML>
Pat, what supplier has individual bearing rollers?
Re-doing bottom end on 30 HP. Races are fine, rollers are not! Not ready to take the plunge for new modern (self aligning) bearings!
Thanks!</HTML>
Re: Connecting Rod Roller Replacement Re-visited
Posted by: Jim Crank (IP Logged)
Date: November 19, 2002 02:05PM
<HTML>Howard,
Most bearing manufacturers will sell their stock bearing rollers separately.
If you need a custom size, try Bremmen Engineering in Bremmen, Indiana. They made the custom sized rollers for me when I made those ten Doble crankcase kits. No one else would do it anywhere in the U.S.
Custom bearing balls, yes, bearing rollers, no.
You would need to buy a mill run from them; but maybe the 30 hp Stanley engine uses the same size as the 20 hp, I don't remember, and you could sell new rollers to the rest of the Stanley owners that need to rebuild their engines.
Jim</HTML>
Most bearing manufacturers will sell their stock bearing rollers separately.
If you need a custom size, try Bremmen Engineering in Bremmen, Indiana. They made the custom sized rollers for me when I made those ten Doble crankcase kits. No one else would do it anywhere in the U.S.
Custom bearing balls, yes, bearing rollers, no.
You would need to buy a mill run from them; but maybe the 30 hp Stanley engine uses the same size as the 20 hp, I don't remember, and you could sell new rollers to the rest of the Stanley owners that need to rebuild their engines.
Jim</HTML>
Re: Connecting Rod Roller Replacement Re-visited
Posted by: Pat Farrell (IP Logged)
Date: November 20, 2002 01:41PM
<HTML>Howard, I could not find individual rollers the correct size. There was a longer roller available at the bearing supply house that I could have put through the chop saw. Other than that, check out Jim's suggestion, or go with modern bearings.</HTML>
Re: Connecting Rod Roller Replacement Re-visited
Posted by: Jim Crank (IP Logged)
Date: November 20, 2002 02:30PM
<HTML>Howard,
I forgot one thing.
I don't know if you are making new bearing races or not; but I will pass on what I did with the new Doble crankshafts and connecting rods.
Doble used square rollers, 5/8" X 5/8", same OD as the length. Absolutely bad engineering, and I think Stanley did the same thing. Square rollers have a great tendency to cock with any possible angle variation due to bending of the crankcase, not possible with the new Doble crankcases; but CERTAINLY the Stanley engine bends all over the place with high power, I have seen this myself and it can be eye watering, they squirm all over the place when they are really working hard, like hill climbing.
Also the smaller rollers let me enlarge the bearing diameters on the crankshafts, giving even more strength and rigidity. Plus, like the Stanley, the Doble bearing races, rod and shaft, did not overlap, a further bad piece of engineering. Now the new Doble shafts have good overlap. And, in those days the hardening was highly variable, now they are precise with the new rollers. Rockwell 60C. I did Rockwell testing on everything before assembly.
The fine people at Bremmen suggested smaller; but more rollers, giving me more in contact at any one time and greater radial load capacity. Also they ran a life time calculation with the new and smaller rollers. Over 100,000 hours at full torque before 10% could fail due to cracking, their criteria for failure. Needless to say, they are inspected once a year, anyhow.
Something for you to think about.
Jim</HTML>
I forgot one thing.
I don't know if you are making new bearing races or not; but I will pass on what I did with the new Doble crankshafts and connecting rods.
Doble used square rollers, 5/8" X 5/8", same OD as the length. Absolutely bad engineering, and I think Stanley did the same thing. Square rollers have a great tendency to cock with any possible angle variation due to bending of the crankcase, not possible with the new Doble crankcases; but CERTAINLY the Stanley engine bends all over the place with high power, I have seen this myself and it can be eye watering, they squirm all over the place when they are really working hard, like hill climbing.
Also the smaller rollers let me enlarge the bearing diameters on the crankshafts, giving even more strength and rigidity. Plus, like the Stanley, the Doble bearing races, rod and shaft, did not overlap, a further bad piece of engineering. Now the new Doble shafts have good overlap. And, in those days the hardening was highly variable, now they are precise with the new rollers. Rockwell 60C. I did Rockwell testing on everything before assembly.
The fine people at Bremmen suggested smaller; but more rollers, giving me more in contact at any one time and greater radial load capacity. Also they ran a life time calculation with the new and smaller rollers. Over 100,000 hours at full torque before 10% could fail due to cracking, their criteria for failure. Needless to say, they are inspected once a year, anyhow.
Something for you to think about.
Jim</HTML>
Re: Connecting Rod Roller Replacement Re-visited
Posted by: Peter Brow (IP Logged)
Date: November 21, 2002 06:04AM
<HTML>Jim,
What would be the maximum adviseable diameter-to-length ratio for rollers? I have never seen square rollers. On the other end of the scale though, I have seen some (Torringtons?) with really long thin rollers -- needle bearings, really -- dia/width ratio of probably 1/10.
If it's not too proprietary, what ratio did you use in the new Doble bearings?
Peter</HTML>
What would be the maximum adviseable diameter-to-length ratio for rollers? I have never seen square rollers. On the other end of the scale though, I have seen some (Torringtons?) with really long thin rollers -- needle bearings, really -- dia/width ratio of probably 1/10.
If it's not too proprietary, what ratio did you use in the new Doble bearings?
Peter</HTML>
Re:Rod Roller D-L ratio, cages
Posted by: C Benson (IP Logged)
Date: November 21, 2002 11:25AM
<HTML>Hi,,,Some of the rollers in m/cycle use are .25dx/ .5 L ///Crankpin more likely would use ''square at .25 x .25 ,,,,or .375 x.375 w/ cages on the crankpin prefferd,,,,Loose rollers seem ok if thrust plates are used to control the skewing of rollers,,,Indian used .375 rolls w/ cast iron cages from 1913 to 1951,,,,and back then advertised they held more records than all other makes combined ! ! When fitting a new [assy] we always honed the sleve to get a good fit,,,the variation in the press of the sleeve would be more than the tolerance,,,so this was essential,,,We also graded[ threw out] the rollers,,,NEVER were they all the same size,, never ,,,,order a extra bag of rollers These engines run for years and are abused by idiots daily and stand the abuse,,,,,All those big harleys[ small h on purpose yeh ]]]]that passed U on the freeway ,use this roller system,,,1913--to date something must work,,, Don't quote me on this part from memory,,,, Cast iron cages to 4000 rpm,,,,alloy above,,,,Loose rolls can be used to 4or 5000 but small rollers do better at high speed i think ,,,Vincent used .120 x .200 rolls on crankpin to 5800rpm,,these were loose ,, that was the limit on racing these,,,,,a 30 inch motor prod' 30-40-hp,,, we ran a 30 '' motor to 8400rpm in 1970,,,above 7000 rpm we had FLOUR of CAGE glistening in the oil,,,,but always finnished,,,,,From all this I would guess square rolls loose will work if ya keep a thrust plate in there to hold alignmet,,,Jim has superior experience and engineering knowledge ,,But I have been working in the auto / bike field since 1951 or thereabouts,,so I feel this input may have value to add to the overall view,, Cheers Ben</HTML>
Re: Connecting Rod Roller Replacement Re-visited
Posted by: Jim Crank (IP Logged)
Date: November 21, 2002 12:38PM
<HTML>Peter,
Not proprietary at all.
After the available sizes were run through their computer, we settled on 3/8" diameter and 5/8" long. Bremmen said that 1.25 to 2.0 ratio is just fine. Best load capacity and maximum life at the loads they would see at maximum.
In the Doble the rollers are all uncaged.
Square rollers are really bad practice, yet I have seen many old machines that used them. I don't think they really knew in those days.
Jim</HTML>
Not proprietary at all.
After the available sizes were run through their computer, we settled on 3/8" diameter and 5/8" long. Bremmen said that 1.25 to 2.0 ratio is just fine. Best load capacity and maximum life at the loads they would see at maximum.
In the Doble the rollers are all uncaged.
Square rollers are really bad practice, yet I have seen many old machines that used them. I don't think they really knew in those days.
Jim</HTML>
Re: Connecting Rod Roller Replacement Re-visited
Posted by: George Nutz (IP Logged)
Date: November 21, 2002 06:08PM
<HTML>Howard,
Small diameter rollers are fine and have a greater load capacity than larger rollers on the same diameter---just that small rollers have less misalignment capability for the same width and a much lower speed capability. This has to do with the roller RPM limitation that can lead to roller "skidding", somewhere around 30-40,000 roller RPM they can start to skid. Most motorcycle people would not recommend small uncaged diameters as they lead to failure at high RPM's but are fine for a 1000RPM steam engine, entirely different conditions. A full roller complement gives the greatest load capacity and needle rollers more dynamic load capacity than larger diameters. Of course the rolling surfaces must be hard;Rc=60 a minimum. Sometimes two narrow roller bearings will be used in place of one long one as it helps with the misalignment problems. Nothing like ultra cheap Torrington needle drawn cup bearings!
Good luck, George</HTML>
Small diameter rollers are fine and have a greater load capacity than larger rollers on the same diameter---just that small rollers have less misalignment capability for the same width and a much lower speed capability. This has to do with the roller RPM limitation that can lead to roller "skidding", somewhere around 30-40,000 roller RPM they can start to skid. Most motorcycle people would not recommend small uncaged diameters as they lead to failure at high RPM's but are fine for a 1000RPM steam engine, entirely different conditions. A full roller complement gives the greatest load capacity and needle rollers more dynamic load capacity than larger diameters. Of course the rolling surfaces must be hard;Rc=60 a minimum. Sometimes two narrow roller bearings will be used in place of one long one as it helps with the misalignment problems. Nothing like ultra cheap Torrington needle drawn cup bearings!
Good luck, George</HTML>
Re: Connecting Rod Roller Replacement Re-visited
Posted by: Jim Crank (IP Logged)
Date: November 22, 2002 02:32PM
<HTML>George,
Correct. The key item in the selection of roller diameters was the roller speed.
As this is a very slow rpm engine, 3/8" was optimum.
Also, as you cannot sit there with a cold engine, winging the throttle, like on a motorcycle or a Bugatti, roller skidding didn't enter into the picture.
Jim</HTML>
Correct. The key item in the selection of roller diameters was the roller speed.
As this is a very slow rpm engine, 3/8" was optimum.
Also, as you cannot sit there with a cold engine, winging the throttle, like on a motorcycle or a Bugatti, roller skidding didn't enter into the picture.
Jim</HTML>
Re: Connecting Rod Roller Replacement Re-visited
Posted by: Peter Heid (IP Logged)
Date: November 22, 2002 06:59PM
<HTML>In many motorcycles, needle bearings do not see much rotation and are very small diameter, maybe 1.5 to 3.0mm needles in most instances. Case in point, connecting rod/wrist pin bearings. The bearing speed is less of a concern than making sure the needles make at least one full revolution in their travel to keep even wear and prevent indenting at one contact point.
BMW motorcycles used roller lower connecting rod bearings in models previous to 1970 and they don't recomend synthetic oils or bearing skid may occur. After running vintage BMW race bikes on BMW brand synthetic oil we found this statment to be without merrit.
I think rollers under 1/8 inch are commonly called needles. All rollers or needles are better caged unless the load demand prevents it due to the drag of rolling against each other.
Peter Heid</HTML>
BMW motorcycles used roller lower connecting rod bearings in models previous to 1970 and they don't recomend synthetic oils or bearing skid may occur. After running vintage BMW race bikes on BMW brand synthetic oil we found this statment to be without merrit.
I think rollers under 1/8 inch are commonly called needles. All rollers or needles are better caged unless the load demand prevents it due to the drag of rolling against each other.
Peter Heid</HTML>
Re: Connecting Rod Roller Replacement Re-visited
Posted by: Peter Brow (IP Logged)
Date: November 24, 2002 07:58AM
<HTML>Ben, Jim, George, and Peter:
Thanks for the useful nuts & bolts info on roller bearings! I'd prefer to avoid homebuilt bearings, but have an idea I'd like to test for low-cost experimental roller eccentrics. Nice to know the workable roller ratios, skid speeds, and hardness. Is hardness matching crucial here -- or is (slight?) hardness mismatch OK as long as both rollers and races are above 60? Now if I can find an affordable Rockwell hardness tester ...
I have also heard tell that outer races need to be slightly out of round, and _not_ absolutely smooth (no mirror finish), for best results. A friend whose job involved replacing antifriction bearings in aero engines way back when, stated that one factory bearing manual instructed them to _drop_ outer races onto a hard surface from a certain height (presumeably mounted on the end of an arm or with some kind of guide), to get the proper out-of-round. Obviously this is not done with modern unit bearings.
Peter</HTML>
Thanks for the useful nuts & bolts info on roller bearings! I'd prefer to avoid homebuilt bearings, but have an idea I'd like to test for low-cost experimental roller eccentrics. Nice to know the workable roller ratios, skid speeds, and hardness. Is hardness matching crucial here -- or is (slight?) hardness mismatch OK as long as both rollers and races are above 60? Now if I can find an affordable Rockwell hardness tester ...
I have also heard tell that outer races need to be slightly out of round, and _not_ absolutely smooth (no mirror finish), for best results. A friend whose job involved replacing antifriction bearings in aero engines way back when, stated that one factory bearing manual instructed them to _drop_ outer races onto a hard surface from a certain height (presumeably mounted on the end of an arm or with some kind of guide), to get the proper out-of-round. Obviously this is not done with modern unit bearings.
Peter</HTML>
Re: Connecting Rod Roller Replacement Re-visited
Posted by: Jim Crank (IP Logged)
Date: November 24, 2002 02:46PM
<HTML>Peter,
Depends on what you are looking for when you do have to rebuild the bearings.
I chose Rockwell 60-61 C for the rollers and 63 C for the races, upon their advice. Then the rollers wear first and can be replaced once every 40,000 miles.
Jim</HTML>
Depends on what you are looking for when you do have to rebuild the bearings.
I chose Rockwell 60-61 C for the rollers and 63 C for the races, upon their advice. Then the rollers wear first and can be replaced once every 40,000 miles.
Jim</HTML>
Sorry, only registered users may post in this forum.