jumping out of 3rd gear

[knighty]

Just built a TS1 225 with 150 gearbox, readspeed 5 plate clutch - clutch is heavy and g/box jumps out of third gear. All good in other gears but as i wind on power in third I can feel the twistgrip moving to pop out of gear - if I hold pressure on to keep it in gear it stays in????? any ideas please. Shimming is good, new selector dog, spring, fork, dogs etc Gearbox gears are used but look perfect - no damage no corners missing etc

[Solid Air]

Cable routing maybe?

[Scooterdude]

Are you certain that you installed 3rd gear the correct way round?

[St George]

Sounds like the third is not fully engaging, possibly due to cable as stated.

[Warkton Tornado No.1]

Presuming the gear is the correct way around & the inner drive faces are good & 'square' then suspect the sliding cursor. Basically, the Indian ones (including SIL) are mis-machined & not as hard as OEM.

I've taken to grinding back the drive faces of the sliding cursor on a jig that I cobbled together. I knew that I had kept dozens of old cursors for something

[missing lynx]

As above I would suspect the sliding dog if it was the cables the problem would be found in 1st or 4th

[Scooterdude]

My first thought would be the sliding dog but OP said that had been replaced which let me to my comment about incorrect installation of gears, however if he’s replaced with a cheap poor quality replacement that could be the issue. How much did you pay and were did you get your replacement sliding dog mate? Just to give us a clue like.

[knighty]

I think from all of your comments I am beginning to doubt the quality of the sliding dog. I bought it from one of the well known traders via ebay but it was not the most expensive available!!!. It is of course possible that I loaded 3rd gear the wrong way round but I doubt it as I am normally quite methodical. I think I will strip the box and have a good look at it - I have a spare LI 150 motor complete so perhaps I should nick the box from that as I know it works - I will take the layshaft as well so it will only need shimming.
I will take the opportunity to see if I can improve the clutch feel as it is a 5 plate readspeed set up and is bloody heavy - I have a Rayspeed cassette in another scoot and the readspeed one feels heavier and I am using the reedspeed springs.
Thanks a lot for your comments and I will post an outcome when I fix it.

[HxPaul]

Inner cable might be to tight.

[Warkton Tornado No.1]

I think from all of your comments I am beginning to doubt the quality of the sliding dog. I bought it from one of the well known traders via ebay but it was not the most expensive available!!!. It is of course possible that I loaded 3rd gear the wrong way round but I doubt it as I am normally quite methodical. I think I will strip the box and have a good look at it - I have a spare LI 150 motor complete so perhaps I should nick the box from that as I know it works - I will take the layshaft as well so it will only need shimming.
I will take the opportunity to see if I can improve the clutch feel as it is a 5 plate readspeed set up and is bloody heavy - I have a Rayspeed cassette in another scoot and the readspeed one feels heavier and I am using the reedspeed springs.
Thanks a lot for your comments and I will post an outcome when I fix it.

The sliding dog cursor has an incredibly hard job to do & innocenti's efforts to get them as good as they could should never be underestimated.

If you check any of the dozens of old ones that I have, dimensionally, they will be consistent to such accuracy, that alone should tell you how important Innocenti & it's licensed partners must have considered manufacturing to drawing specification!

So, if you then compare any other manufacture, you will find the Indian type, even SIL are not even on the same page!

Therefore, by my train of thought, how likely is it that the material specification & other processes, particularly hardening, will be anything like what should be required in a standard Lambretta engine?

The dogs will wear on any cursor, but the way I now see the issue, it is best to use an old, OEM (not Indian) component & dress the working faces of the dogs so that some measure of consistency can be relied upon.

Admittedly, it has taken me many attempts to get my own procedure accurate enough to work, but I think that eventually, I have effected a 'cure' that works in 25+ BHP motors.

I'm glad that I never scrapped all of my old OEM cursors, though I must have @ least a dozen 'one race event only' alternative cursors taking garage space. They are SO wrong that I have a number of them that have cracks along the length, because the manufacturers know sweet F A about material choice, hardening, tempering & surface finishing.

As for clutches, I have built some road motors with beefed up clutches in my time, but the last one was in a ported TS (70 mm DT250 re-pegged piston x 60 mm stroke long rod crank estimate about 30 BHP) yet to be dynoed, that I squeezed a six plate CamLam clutch into, & that is great......so far. Mind you, I spent some time selecting from my stock of stronger springs, measuring compressed length & wire gauge. However, I haven't been using the ST90 type oil for nearly a decade now & have to say there are much better alternatives IMHO

[nickw]

I haven't been using the ST90 type oil for nearly a decade now & have to say there are much better alternatives IMHO

What gearbox oil do you use please? I did try some synthetic oil years ago when the carbon clutches were made and what a waste of time they were. I never continued trying the synthetic oil.

[Warkton Tornado No.1]

I haven't been using the ST90 type oil for nearly a decade now & have to say there are much better alternatives IMHO

What gearbox oil do you use please? I did try some synthetic oil years ago when the carbon clutches were made and what a waste of time they were. I never continued trying the synthetic oil.

Despite over four decades continuous ownership of Lambrettas, I try to remain open minded when it comes to alternative ideas. My belief is that philosophy is even essential to function as a design engineer, as procedures 'set in stone' in one industry may be costly & restrictive, so there are occasions when I have championed an alternative that has worked well.

As the result of some times working within Motorsport, I have had lots of chances to try alternative oils & even fasteners etc.

As the consequence of having changed the ATF in a Mazda Bongo, & being aware that it is widely used in MX, I thought it worth a try in a six plate clutch, torquey engine (TV200 top gear ratio) that annoyingly slipped just a tad.

So I thought "What the Heck?" & just drained out the old oil, & replaced it with ATF. I thought it would be inevitable that it would result in a strip down, so didn't even replace the clutch.

That was seven years ago & I have put ATF in all the motors I have built ever since, the latest being the self ported TS 70 x 60 mm which has had no clutch issues @ all.

The best thing is that it is cheap as chips.

There will be prejudice against it because of people's pre-occupation with ATF 'slipping' but slippage only occurs because in an auto, components that transmit the torque, does so via the viscous nature of the fluid. They are not physically in contact with each other! ATF does 'grip' & can also take a lot of abuse. You can buy it far more easily than normal gearbox oil. Bantam racer's with three speed gearboxes (therefore ridiculously high first gears!) use steel on steel clutches, ATF & go indecently quickly.

I hope some of you might be as pleased as I am having tried it.

[rossclark]

Latterly when I was last running Vethpas in the 90s I regularly used ATF in those and the components aren't all that different - I see no reason why it shouldn't work.

[nickw]

Thanks for that. Won't say anymore as I don't want the thread to drift.

[Scooterdude]

I’ve heard it said somewhere before about using ATF , thing is if you strip an auto box you will find two clutch packs, one front and one at the rear, now these don’t look dissimilar in appearance or size to those in a Lambretta plus of course all the bearings and other stuff they have to lubricate so it would make good sense to use it, only thing being different grades ie; dexron 2,3, 6 etc I wonder which would be best to use

[Warkton Tornado No.1]

Dexron III is the stuff I use, but there may be something even better.

I meant to say that in use, my clutches fitted with ATF have no tendency to 'creep' with the clutch lever pulled in, not even from cold with the generic Lambretta type with six plates squeezed in standard components, the only modification being to the crownwheel slot depths if necessary. I'm also fussy about the spring gauge & fully compressed height of the set, to ensure getting coil bound is avoided & consistency'of them in action.

However, my usual practice is to set the clutch release arm toward the rear of the engine to give more leverage. Setting it 'square' gives the least movement if you think about it

[Scooterdude]

Dexron III is the stuff I use, but there may be something even better.

I meant to say that in use, my clutches fitted with ATF have no tendency to 'creep' with the clutch lever pulled in, not even from cold with the generic Lambretta type with six plates squeezed in standard components, the only modification being to the crownwheel slot depths if necessary. I'm also fussy about the spring gauge & fully compressed height of the set, to ensure getting coil bound is avoided & consistency'of them in action.

However, my usual practice is to set the clutch release arm toward the rear of the engine to give more leverage. Setting it 'square' gives the least movement if you think about it

yes mine creeps when cold using SAE90 but within 2 minutes of riding is fine at the junction when I have to stop and set off again, but it’s the initial start up from cold and clonk into gear I don’t like so am keen to give ATF a go.

As i look at the clutch cable adjuster it strikes me it would work better if the position of the block was further inbound with the cable routed behind the engine mount and with that kink taken out of it so the inner doesn’t rub on the adjuster if you see what I mean.

[Warkton Tornado No.1]

As i look at the clutch cable adjuster it strikes me it would work better if the position of the block was further inbound with the cable routed behind the engine mount and with that kink taken out of it so the inner doesn’t rub on the adjuster if you see what I mean.

With the clutch arm, the more degrees of rotation will be achieved, the more acute the angle.

Difficult to explain, but if you think how many degrees of rotation that a crankshaft goes through with the piston approaching TDC & going beyond, the angle is huge for very little linear movement of the piston. If that principle is applied to the clutch arm (most degrees of rotation per linear cable movement) then the arm should be angled as far back as practicable for those of us requiring more running clutch clearance. That is to say, six + plate clutches/warped steels/ST 90( ) filled gearboxes can all do with whatever extra clearance you can gain.

In extreme cases such as old school nine plate clutches, we used to cut the arm & re-weld the forked part back on making it a smaller radius to gain more rotational movement. In any case, that arm is often better suited bent 'upwards' to gain a bit of space when working on the engine, what with that area being s 'congested'