Gleniffer Model RB reversing gearbox

[RLWP]

As there are not many of these around (three according to Mike, two in boats and one in a museum), the Vintage Engine section seems most appropriate for this

 

This is rather a different box to the PRM 160 I described recently. It's a lot bigger for a start, and more robust

 

First off, a diagram!

 

 

This is useful as it labels the different parts of the two clutches, we will come back to them

 

Shown here are (most of) the working parts of the gearbox, with the casing and change mechanism not fitted:

 

 

This box gets forward or reverse by using a bevel gear cluster in the centre:

 

 

The bevel gear at the bottom is keyed onto a shaft on the end of the crankshaft:

 

 

This stub shaft is sat in the middle of a huge, conical drum - part C in the diagram above.

 

The bevel gear at the top of the cluster is attached to the driven shaft. Between them sit a pair of bevel gears on a spider:

 

 

Notice that this spider has two trunnions that sit in holes in parts A and 41

 

Part A is a huge drum with a series of spiral grooves cut on the outside:

 

 

On the outside of this sits and slides an enormous clutch disk (Parts 51, 52, 53) with two cones cut on the outside labelled D and F. You can see this assembly at the bottom of this stack:

 

 

The final important component is that there is a cone machined on the inside face of the gearbox casing, labelled F:

 

 

How does it work? OK, before you get lost, the important thing is that the clutch disk is attached to the bevel spider, and that the clutch disk can slide backwards and forwards. Hang on to that thought at all times

 

To engage forward gear, the clutch disk slides forward and the cone D on the outside engages with the cone C on part 50, which is driven by the engine. This fixes the bevel spider to the driven shaft and the locks the bevel assembly up. As the input bevel and the transfer bevels are locked, the output bevel is also locked and everything turns in the same direction as the engine

 

To engage reverse gear, the clutch disk slides backwards and the cone F engages with the cone E in the gearbox housing. This fixes the bevel spider stationary in the gearbox. The input bevel is still attached to the driving shaft and rotates, driving the transfer bevels around. These in turn drive the output bevel around in the opposite direction to the input bevel, and drives the output shaft in reverse.

 

The spiral, labelled B, is cut so that in forward gear, the spiral drives the clutch disk forwards, and backwards in reverse. This helps to hold the box in gear.

 

Notice that there are less parts being used in the drive train in forward than reverse, good design in a marine gearbox

 

Richard

Edited May 14, 2012 by RLWP

[cereal tiller]

As there are not many of these around (three according to Mike, two in boats and one in a museum), the Vintage Engine section seems most appropriate for this

 

This is rather a different box to the PRM 160 I described recently. It's a lot bigger for a start, and more robust

 

First off, a diagram!

 

 

This is useful as it labels the different parts of the two clutches, we will come back to them

 

Shown here are (most of) the working parts of the gearbox, with the casing and change mechanism not fitted:

 

 

This box gets forward or reverse by using a bevel gear cluster in the centre:

 

 

The bevel gear at the bottom is keyed onto a shaft on the end of the crankshaft:

 

 

This stub shaft is sat in the middle of a huge, conical drum - part C in the diagram above.

 

The bevel gear at the top of the cluster is attached to the driven shaft. Between them sit a pair of bevel gears on a spider:

 

 

Notice that this spider has two trunnions that sit in holes in parts A and 41

 

Part A is a huge drum with a series of spiral grooves cut on the outside:

 

 

On the outside of this sits and slides an enormous clutch disk (Parts 51, 52, 53) with two cones cut on the outside labelled D and F. You can see this assembly at the bottom of this stack:

 

 

The final important component is that there is a cone machined on the inside face of the gearbox casing, labelled F:

 

 

How does it work? OK, before you get lost, the important thing is that the clutch disk is attached to the bevel spider, and that the clutch disk can slide backwards and forwards. Hang on to that thought at all times

 

To engage forward gear, the clutch disk slides forward and the cone D on the outside engages with the cone C on part 50, which is driven by the engine. This fixes the bevel spider to the driven shaft and the locks the bevel assembly up. As the input bevel and the transfer bevels are locked, the output bevel is also locked and everything turns in the same direction as the engine

 

To engage reverse gear, the clutch disk slides backwards and the cone F engages with the cone E in the gearbox housing. This fixes the bevel spider stationary in the gearbox. The input bevel is still attached to the driving shaft and rotates, driving the transfer bevels around. These in turn drive the output bevel around in the opposite direction to the input bevel, and drives the output shaft in reverse.

 

The spiral, labelled B, is cut so that in forward gear, the spiral drives the clutch disk forwards, and backwards in reverse. This helps to hold the box in gear.

 

Notice that there are less parts being used in the drive train in forward than reverse, good design in a marine gearbox

 

Richard

Superb pictures and an easily understood explanation of how the reversing gear works.

 

I have never come across a GLENIFFER,were they a scottish company?

[RLWP]

Superb pictures and an easily understood explanation of how the reversing gear works.

 

I have never come across a GLENIFFER,were they a scottish company?

 

Thank you.

 

Yes, Gleniffer Engines, Limited; Anniesland, Glasgow W.3

 

This was an fascinating exercise which lead to dealing with a design problem with the box. I'll put up some more photos either later or more likely tomorrow

 

Richard

[MtB]

Excellent post Richard!

 

Gleniffer made larger marine engines for ferries, lifeboats etc AIUI and the DB2 In Reginald is the smallest they ever made. They got into financial difficulties a few decades ago AIUI and where bought out by Bergius & Co, AKA Kelvin.

 

And to answer your question in another thread Richard, the rebuilt box engages reverse reliably now PROVIDED one waits for the engine speed to settle done to tickover. If one engages reverse at any engine speed above tickover then the old problem persists. It spins the prob lazily with huge clutch slippage and only the slightest amount of reverse thrust. Much to the amusement of a mother and two kids as I whacked the bank trying to execute a three point turn on the Wolverhampton Level on Sunday!

[MoominPapa]

Excellent post Richard!

Seconded.

 

 

 

And to answer your question in another thread Richard, the rebuilt box engages reverse reliably now PROVIDED one waits for the engine speed to settle done to tickover. If one engages reverse at any engine speed above tickover then the old problem persists. It spins the prob lazily with huge clutch slippage and only the slightest amount of reverse thrust.

Something to do with the fact it's got all those meaty springs to provide engagement force on the ahead clutch, but nothing comparable for reverse?

 

MP.

[matty40s]

Seconded.

 

 

Something to do with the fact it's got all those meaty springs to provide engagement force on the ahead clutch, but nothing comparable for reverse?

 

MP.

 

Brilliant pics and almost understandable workings, I am impressed at the complexity and also simlicity of its operation.

 

Does being pink compromise the reverse engagement?

[Chertsey]

I enjoyed that too (I was asking a few months ago how my Parsons reversing box worked) but I must confess that when I got to 'Notice that this spider has two trunnions' I came to my senses and realised I had been online for too long.

[Timleech]

As there are not many of these around (three according to Mike, two in boats and one in a museum), the Vintage Engine section seems most appropriate for this

 

 

Richard

 

I don't know which is the other one 'in a boat', but I sold one of these quite a number of years ago, as a pile of bits but with a complete new set of ball bearings, no idea whether it was eventually fitted into a boat. I bought it attached to a Gardner 2L2, looked very much as though it had been an original fitment from the Gardner factory.

 

Tim

[Starcoaster]

It works much better for being pink.

[RLWP]

Seconded.

 

 

 

Something to do with the fact it's got all those meaty springs to provide engagement force on the ahead clutch, but nothing comparable for reverse?

 

MP.

 

Actually, that isn't what those meaty springs do - I'll explain tomorrow

 

Richard

 

And pink is the only colour for a proper gearbox

[RLWP]

So, what was wrong with this box? Let's go back to this assembly:

 

 

Those big bevels top and bottom are each mounted on a big ball race:

 

 

These bearings fit into castings top and bottom.

 

Meanwhile, at the tail of the gearbox is the main thrust bearing:

 

 

Strangely, this was the cheapest bearing in the box to replace. To take out any endfloat in the shafts, this bearing sits in a housing with a screw thread on it. It's very big, and very fine, and unfortunately makes an excellent bearing press. Overtightening this housing pushes the output shaft bevel and bearing forwards out of it's housing until it binds with the transfer bevels. Then the spider pushes the spiral drum back, pushing the input bevel bearing out if it's housing as well. This is a Bad Thing! The bearings are no longer properly engaged in their housings, the shafts butt against each other and, more importantly, it stops the clutch disk sliding backwards to engage reverse. This was the original problem

 

Here is the end of the output shaft:

 

 

Those notches either side are actually a hole drilled to take a split pin, which would originally have been about 1/8" from the end!

 

The bearing housing in the spiral drum was very worn (0.016" oversize ), so I had to improvise a fix:

 

 

I cut six pieces of 0.008" shim steel to fit between the bearing and the housing.

 

I also drilled and tapped the housing for six cap screws:

 

 

 

The whole lot was assembled with a good dose of Loctite 638:

 

 

Inspecting the clutch disk and spiral drum showed that the disk was being stopped by part of the bearing housing:

 

 

I suspect that this was because of the output shaft being forced forward, but just in case I cut back two of the teeth on the clutch disk:

 

 

Those springs. This disk is actually an assembly. There is a part with the spiral teeth on the inside and the studs for the springs. There is a part with the two cones machined on it and there is a third part. The flange of the coned part is clamped between the other two parts and held there by the springs. This is a shock load device, If the propeller should catch or lock up on something, this assembly can slip, preventing damage to the box

 

Richard

Edited May 15, 2012 by RLWP

[Timleech]

 

Meanwhile, at the tail of the gearbox is the main thrust bearing:

 

 

Strangely, this was the cheapest bearing in the box to replace.

Richard

 

What type of bearing is that? The one that I had, which I bought all new bearings for, used a ball thrust race with a split outer, these are even more obsolete than the ones with the split inner. I can't remember for certain, but it was probably the most expensive bearing by a long way.

 

Tim

Edited May 15, 2012 by Timleech

[RLWP]

What type of bearing is that? The one that I had, which I bought all new bearings for, used a ball thrust race with a split outer, these are even more obsolete than the ones with the split inner. I can't remember for certain, but it was probably the most expensive bearing by a long way.

 

Tim

 

It's just a plain ballrace. Most of the bearings were easy to find, although some ingenuity was needed. These are Vespa crankshaft bearings:

 

 

This was the hardest one to replace, I could only find one with the original manufacturer and it came in at well over £100:

 

 

I didn't put back the original locking device:

 

 

Richard

[David Mack]

To engage forward gear, the clutch disk slides forward and the cone D on the outside engages with the cone C on part 50, which is driven by the engine. This fixes the bevel spider to the driven shaft and the locks the bevel assembly up. As the input bevel and the transfer bevels are locked, the output bevel is also locked and everything turns in the same direction as the engine

 

So if I understand that correctly, in forward gear the bevel gear on the input shaft, and the spider carrying the transfer bevels, are both locked to the input shaft. That means that the drive is transmitted from the input bevel, to the two transfer bevels to the output shaft bevel, with the whole assemby rotating, but none of these bevel gears are rotating on their shafts. That means that once forward gear is engaged the drive is transmitted by tooth to tooth contact between particular teeth, rather than being transferred from one tooth to the next as would happen if the gears were rotating (as indeed happens in reverse). I guess this is OK in a marine application, but if used in a vehicle I would have thought there would be scope for impact damage to the load bearing faces of individual teeth, which would then affect the smooth running of the gearbox.

 

David

[MoominPapa]

So if I understand that correctly, in forward gear the bevel gear on the input shaft, and the spider carrying the transfer bevels, are both locked to the input shaft. That means that the drive is transmitted from the input bevel, to the two transfer bevels to the output shaft bevel, with the whole assemby rotating, but none of these bevel gears are rotating on their shafts. That means that once forward gear is engaged the drive is transmitted by tooth to tooth contact between particular teeth, rather than being transferred from one tooth to the next as would happen if the gears were rotating (as indeed happens in reverse).

 

My understanding of Richard's explanation is that the drive in forward goes via the cone on the input shaft and the cone on the sliding part to the large drum, so not via the teeth of the bevel gears.

 

What you describe is true in Lister epicyclic gearboxes, where the forward clutch locks up a couple of straight gears in the rotating assembly. These gears are the only thing in contact with the input shaft (actually a pinion on the end of it) so drive is transmitted through whichever teeth happen to be engaged with the pinion. It seems to work and be long-lived.

 

MP.

[Chris Pink]

Superb pictures and an easily understood explanation of how the reversing gear works.

 

I have never come across a GLENIFFER,were they a scottish company?

 

That's twice recently you've done that. Stop it please. There is no need to quote the whole of Richard's post with pictures. It makes for a long and unwieldy page. Fine though Richard's exposition is, and it is rather fine, once is adequate.

 

 

 

Does being pink compromise the reverse engagement?

 

Excuse me?

[Timleech]

It's just a plain ballrace. Most of the bearings were easy to find, although some ingenuity was needed.

 

Richard

 

I wonder whether the original was a split race, which had been replaced at some point with a deep groove ball race of the same size? Those can carry quite a lot of thrust. My Gleniffer manual is not to hand at the moment, so I can't see whether that offers any clues.

I seem to remember that a full set of bearings for the box that I had cost me over £250, with a good discount, about 25 years ago.

 

Tim

[luctor et emergo]

That's twice recently you've done that. Stop it please. There is no need to quote the whole of Richard's post with pictures. It makes for a long and unwieldy page. Fine though Richard's exposition is, and it is rather fine, once is adequate.

 

 

Excuse me?

 

 

The reference to pink, is to Starcoaster's gearbox, which I believe has had an unfortunate coming together with a quantity of pink paint.....

[matty40s]

The reference to pink, is to Starcoaster's gearbox, which I believe has had an unfortunate coming together with a quantity of pink paint.....

Quite right, she occasionaly lets MiketheBoilerman have a play on Reginald too.....if he's made the coffee.

[RLWP]

So if I understand that correctly, in forward gear the bevel gear on the input shaft, and the spider carrying the transfer bevels, are both locked to the input shaft. That means that the drive is transmitted from the input bevel, to the two transfer bevels to the output shaft bevel, with the whole assemby rotating, but none of these bevel gears are rotating on their shafts. That means that once forward gear is engaged the drive is transmitted by tooth to tooth contact between particular teeth, rather than being transferred from one tooth to the next as would happen if the gears were rotating (as indeed happens in reverse). I guess this is OK in a marine application, but if used in a vehicle I would have thought there would be scope for impact damage to the load bearing faces of individual teeth, which would then affect the smooth running of the gearbox.

 

David

 

Yes, that's right

 

Richard

 

 

Does being pink compromise the reverse engagement?

 

 

It works much better for being pink.

 

 

And pink is the only colour for a proper gearbox

 

 

Excuse me?

 

To allay any confusion:

 

It's all Starry's fault, she was included in a greasy bloke's conversation by PM and made this fine contribution:

 

Fa la la, kittens, hair ribbons, daisies...

**trips over giant metal engine thingy, paints it pink and skips off**

Fa la la, time to paint my nailsssss...

 

Well, what would you do?

 

 

 

Richard

 

I wonder whether the original was a split race, which had been replaced at some point with a deep groove ball race of the same size? Those can carry quite a lot of thrust. My Gleniffer manual is not to hand at the moment, so I can't see whether that offers any clues.

I seem to remember that a full set of bearings for the box that I had cost me over £250, with a good discount, about 25 years ago.

 

Tim

 

It is possible Tim. All I had to hand was the manual and the parts themselves. With this particular bearing, in a traditional boat this bearing would take the thrust from the propeller. In Reginald, there is a plummer block further back doing that job, so I decided that it was worth fitting the bearing I could get. If it fails, it's very easy to get to, and it should give some good audible signs well in advance of that

 

Richard

[Timleech]

 

It is possible Tim. All I had to hand was the manual and the parts themselves. With this particular bearing, in a traditional boat this bearing would take the thrust from the propeller. In Reginald, there is a plummer block further back doing that job, so I decided that it was worth fitting the bearing I could get. If it fails, it's very easy to get to, and it should give some good audible signs well in advance of that

 

Richard

 

If there's a thrust bearing external to the box, it may be better to have a plain ball race anyway.

Certainly Parsons, who used split race ball thrust bearings in their boxes, were emphatic about the need to replace them with a plain ball bearing if the thrust was taken elsewhere.

 

Tim

[jake_crew]

Great pictures Richard,

 

We started our Gleny DB2 at the weekend, after a LONG rebuild.

 

We have the 'box in pieces at present, but I still struggle to understand the action. With the reduction box on, it must weight 4cwt. I had it resting on my Totector at one point !

 

The manual states that thrust from the prop is NOT requuired, and anyway, the reduction box has back to back gears to prevent thrust being transmitted.

 

In forward gear, nothing is moving apart from input & output shafts and associated components.

 

IMHO the engagement energy is to generated by that sprial.

[RLWP]

Great pictures Richard,

 

We started our Gleny DB2 at the weekend, after a LONG rebuild.

 

We have the 'box in pieces at present, but I still struggle to understand the action. With the reduction box on, it must weight 4cwt. I had it resting on my Totector at one point !

 

The manual states that thrust from the prop is NOT requuired, and anyway, the reduction box has back to back gears to prevent thrust being transmitted.

 

In forward gear, nothing is moving apart from input & output shafts and associated components.

 

IMHO the engagement energy is to generated by that sprial.

 

The thing about "thrust from the prop is NOT required". The gearbox on Alnwick's Kelvin uses the propeller thrust to hold it's cone clutches in engagement.

 

If you want any help with that gearbox...

 

Richard

[Timleech]

The thing about "thrust from the prop is NOT required". The gearbox on Alnwick's Kelvin uses the propeller thrust to hold it's cone clutches in engagement.

 

 

Richard

 

As do the early 'Conic Drive' Gardner gearboxes which work in pretty much the same way as the Kelvin box.

The later Kelvin boxes, however, have much more in common with a PRM, but they came in two sizes. The larger one, RG32, weighs half a ton. The smaller RG12, probably about 6cwt at a guess.

 

Tim

[jake_crew]

If you want any help with that gearbox...

 

Richard

Thanks for your kind offer Richard, we shall most certainly bear it in mind.

 

I did send a copy of the manual to Mike (Reginald) last year.

 

I assume you are in Kenilworth ? We are near Doncaster, however we fully intend visiting Braunston at the end of June (but not by water).

 

Regards

 

Colin