Honda-SOHC
SOHC.co.uk Forums => The Black Bomber Board => Topic started by: AshimotoK0 on June 24, 2020, 08:39:39 AM
-
I went to see a bloke recently who sells me parts from time to time and has Honda classic race machines and most of the remaining original /UK-stocked spares for them.
He told me that a lot of CB450 racers and some owners of standard machines have converted to conventional valve springs, as the torsion bars degrade after 50 odd years in torsion and are hard and expensive to come by NOS. I had never heard of this before but it seems he is correct.
https://www.caferacer.net/threads/cb450-valve-spring-conversion.11894/
-
I've been reading through that thread Ash, but it's like being blasted with a scatter gun of variable quality reasoning ;D much of it born of misunderstood engineering principles.
The subject is so dispurse because of the above it's hard to make conclusive decisions around it, other sites the posters seem to hold the view that there are no springs in there at all !!
https://www.hondatwins.net/threads/new-or-used-torsion-bars-test-of-6-used-cb450-doch-heads.37856/ looking at this one, I can't visualise just how he's checkng the load (physical arrangement) but the numbers appear repeatable and interesting.
Initial thoughts are that the interpretation of the data, specifically for B torsion bars, in showing that variance, are a concern. But he has no reference of their original (obviously) performance to say that they've changed. There exists the possibility that the original samples had significant differences straight off the shelf.
The A bars look more like you'd expect from initial metallurgical theory. Can't be sure of exactly the material used, but general principle suggests "creep" ( bar loaded continuously without respite) of 0.1% per year, making 5% from new to now if left under load for entirety.
Have you got facilities to test them?
Would have thought it's possible to reproduce something like this currently, F 1 routinely use torsion bar in suspension. Possibly the scale may be within someone's capability to batch produce.
-
Seems to be fairly common in the 'states in vintage race classes. Team Hansen (http://teamhansenhonda.com/parts.htm) is one place to go for spring conversions. Sounds like they used to do a drop in conversion kit at one time.
There's a bunch of other places that do camshaft regrinds for race 450's, and there's a number of profiles that are accompanied by the warning only to be used with valve spring conversion heads.
-
Have you got facilities to test them?
I don't have anything set up yet but I have about 5 sets of torsion bars and I need to select the best set for my engine very soon, so I will have to make something up. From my understanding, the Honda Twins guy merely assembles the torsion bar in the head as normal (the bars are not splined all of the way around the diameter so are effectively 'keyed' in position). He then applies torsion with an accurate torque wrench until the locating hollow dowel (knock pin) in the head is in line with the hole in the torsion bar 'arm'. He then notes the torque figure. It must be quite tricky to determine when the dowel in excactly in-line with the hole though because there is only a small amount of angular deflection even on a new torsion bar. So even a small amount of misalignment of the pin relative to the hole will make a big difference to the torque value measured.
-
I've not seen their method, but assumed it was something like you've described.
With the engine/ head held in a jig of sorts, you could apply effort to the torque wrench with something like a rope loop that you put a "stick" into and twist to keep raising torque level until an able assistant ;D can slot the pin home.
Or, in the same fashion, use one of those double ended yacht rigging tensioner to gently wind on more load and hold while you read off torque when pin locates.
-
I
Or, in the same fashion, use one of those double ended yacht rigging tensioner to gently wind on more load and hold while you read off torque when pin locates.
I might do that Nigel (I always knew them as 'turn-buckles' :)) and put an accurate tension digital load cell i have, in line, to measure the force and calculate the torque from the length of the bar (bar at 90° to the tensioner).
-
Why would a torsion bar be any more susceptible to ageing than a spring?.....after all, a torsion bar is essentially an unwound spring and twists in exactly the same way.......unless someone knows different?
Mick.
-
Why would a torsion bar be any more susceptible to ageing than a spring?.....after all, a torsion bar is essentially an unwound spring and twists in exactly the same way.......unless someone knows different?
Mick.
Coil springs weaken with age too. That's why you measure the spring length.
-
Why would a torsion bar be any more susceptible to ageing than a spring?.....after all, a torsion bar is essentially an unwound spring and twists in exactly the same way.......unless someone knows different?
Mick.
Springs easy to source ... torsion bars NOS, very expensive and NLA from Honda
-
There's just such an almost impenetrable fog about these engines in particular that really obscures any clear judgments if they are out there.
It'll be interesting to see what your testing yealds Ash as like you I suspect it's difficult to get a good read of the load while accurately positioned. If you can get repeatable numbers at least you'll have a more factual foundation to work with.
Why would a torsion bar be any more susceptible to ageing than a spring?.....after all, a torsion bar is essentially an unwound spring and twists in exactly the same way.......unless someone knows different?
Mick.
As you Mick, I don't see as much change as is generally put forward by discussion of these components. Many are projected as nothing more than because it's old, then it must have changed "stands to reason innit " ;D out in the great internet.
Metallurgical reasoning gives that steel is prescribed for spring components specifically to alter their long term (permanent) physical composition, such that defined heat treatment, hardening, subsequent stress relieving manufacture, then leaves the components able to maintain this condition almost indefinitely.
The metallurgy doesn't change, all the heat treatment lies way out of engine temperature operating range all reasearched and field tested to levels significantly beyond layman's terms of appreciation or reference. Many have no hope of understanding it let alone give informed comment, but that doesn't stop them presenting their "facts".
-
But when torsion bars are used as suspension on a car they can "settle" over an extended period so there must be some sort of permant twist as it were
-
But when torsion bars are used as suspension on a car they can "settle" over an extended period so there must be some sort of permant twist as it were
Good point, I'll elaborate my thinking.
Although we are using "torsion" as overall blanket descriptor, the material and final process can make a large difference to the component we are discussing. Other applications I feel aren't the best comparitor as a means of viewing performance of something so specific as the valve springs.
https://www.engineeringenotes.com/metallurgy/steel/heat-treatment-of-spring-steels-metallurgy/26992 this is worth a read if just to demonstrate how intricate are the choices made by the original design in selecting the most resilient performance.
-
Porsche 911s sit on torsion bar suspension rather than coil springs, and though I've never heard of it as being a problem, I did notice this on the widely respected Pelican Parts website, however they make a living selling new spares, so do have a vested interest:
While many normal cars use coil springs, the 911 suspension uses the torsional stiffness of the metal bars in order to provide the necessary spring. Like traditional coil springs, as the cars age the torsion bars have a tendency to become weak, or even snap and break.
My CL450 barn find sat for several decades with an exhaust valve open, causing a mouldy piston crown and a rusty valve seat. However, there have been absolutely zero problems with setting and running that valve, a check after 660 miles showed zero change in the valve clearance.