LNWR Problem 2-2-2 CSBs

What individual members are up to.
User avatar
Winander
Posts: 842
Joined: Thu Mar 20, 2014 12:19 pm

LNWR Problem 2-2-2 CSBs

Postby Winander » Tue Feb 09, 2021 11:59 pm

This topic https://www.scalefour.org/forum/viewtopic.php?f=19&t=7073#p77876 discussed CSBs for a 2-2-2.

I recently purchased from London Road Models a kit for a LNWR Problem class locomotive (designated by the LNWR as a 7' 6" Single) together with wheels originally from Mike Sharman. The carrying wheels are 3' 9" and more readily available, but the drivers at 7' 6" are proverbial hen's teeth.

https://www.lnwrs.org.uk/PassClassLocos/pass_class_menu.php?display_class_details=rpro

The plan is tender drive using Ultrascale universal joints, High Level gearboxes, Mashima motor and continuous springy beams.

The first challenge to be overcome is the rather diminutive nature of the frames for this engine first built in 1859. The advantages are, no loco brakes, and no connecting rods to worry about.

Below is the plot from Will's CSB spreadsheet. I was running it in MacOS Numbers which warned that the conditional cell processing was ignored which I take as an indication that the red cell warnings weren't working. Running it in MS Excel may look different.

I have little to play with in the length of the chassis, front axle to end of the frame is 12.5 mm, back axle 10mm. I have allowed about 1mm for the CSB wire to extend. The wheelbase at the front is 7' 7" and the back 7' 10".

The screenshot below is from the auto calculate option. The figures shown entered for the NOT calculated option give results that are not that different but provide more wiggle room at the ends of the CSB.

Edit: Fixed link to correct thread!
You do not have the required permissions to view the files attached to this post.
Last edited by Winander on Thu Feb 11, 2021 1:54 pm, edited 1 time in total.
Richard Hodgson
Organiser Scalefour Virtual Group. Our meeting invitation is here.

User avatar
Will L
Posts: 2516
Joined: Sun Jul 20, 2008 3:54 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Will L » Wed Feb 10, 2021 11:36 am

Richard
This was good to see. I have a couple of, hopefully, helpful comments but no time to give them at the moment. I will post again, perhaps tonight if not tomorrow.

Winander wrote: I was running it in MacOS Numbers which warned that the conditional cell processing was ignored which I take as an indication that the red cell warnings weren't working. Running it in MS Excel may look different.

I think that's probably right. I'll have to have a think about what I can do about it.

User avatar
Winander
Posts: 842
Joined: Thu Mar 20, 2014 12:19 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Winander » Thu Feb 11, 2021 4:21 am

I must confess I had not taken any account of the suggested weight distribution of 30/40/30 in this post https://www.scalefour.org/forum/viewtopic.php?f=19&t=7073#p77876 when entering data into the spreadsheet.

I have drawn a first draft of the frames. The red line is the position of the CSB wire at position C in the carrier tag which isn't high enough to permit placement of the fulcrums. I have measured the distance between the axle centre and position C from the High Level jig at 4.5mm.

Frames First Draft.png

The drawing was made from the etch supplied by LRM the design of which was copyrighted by Jol Wilkinson.

Edit: removed reference to "static" position of the CSB wire so as not to confuse it with the static deflection
You do not have the required permissions to view the files attached to this post.
Last edited by Winander on Thu Feb 11, 2021 1:12 pm, edited 1 time in total.
Richard Hodgson
Organiser Scalefour Virtual Group. Our meeting invitation is here.

User avatar
Jol Wilkinson
Posts: 1111
Joined: Mon Jul 21, 2008 7:39 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Jol Wilkinson » Thu Feb 11, 2021 9:26 am

Richard,

whose hornblocks are you using and how are you "attaching" the CSBs to them? I may be able to suggest a way of dealing with the CSB mountings on the frames.

Jol

User avatar
Winander
Posts: 842
Joined: Thu Mar 20, 2014 12:19 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Winander » Thu Feb 11, 2021 11:42 am

Jol,

I was going to use High Level with their carrier tags, but don't have a preference. I'm all ears!

thanks
Richard Hodgson
Organiser Scalefour Virtual Group. Our meeting invitation is here.

User avatar
Will L
Posts: 2516
Joined: Sun Jul 20, 2008 3:54 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Will L » Thu Feb 11, 2021 12:49 pm

Winander wrote:I must confess I had not taken any account of the suggested weight distribution of 30/40/30 in this post https://www.scalefour.org/forum/viewtopic.php?f=19&t=7073#p77876 when entering data into the spreadsheet.

I have drawn a first draft of the frames. The red line is the static position of the CSB wire at position C in the carrier tag which isn't high enough to permit placement of the fulcrums. I have measured the distance between the axle centre and position C from the High Level jig at 4.5mm.

Image
The drawing was made from the etch supplied by LRM the design of which was copyrighted by Jol Wilkinson.


Richard

Yes it was on the whole topic of weight distribution that was wanted to have a think about and I have had a chance to run the numbers this morning. Of course it up to you what weight distribution you go for, and the auto calculated result was fair enough, but given a singe driver will benefit from all the weight it can get on the driving axle, this plot appears to show a bit more bias way from the centre axle (the 12% figure) than was perhaps desirable.
N.B.the 12% refers to difference in the Y1.Y2 and Y2 figures NOT to the weight distribution, and magnifies the effect significantly which is a good thing when considering the result.

When I checked it out it the weight distribution works out as 34 / 32 / 34 %, which shows less weight distribution difference than I expected. If you want to continue with an equal weight distribution a slightly better plot would bring the rear most two fixed fulcrums in a bit as follows.
LNWR Problem 2-2-2 CSBs.jpg

This shows you how an auto calculated result can be modified by applying adjustments, and that the spreadsheet is complaining that the centre axle is very slightly stiffer than the outer two.

If you decided you would like to go for a 30 / 40 / 30 %, this plot shows how its done
LNWR Problem 2-2-2 CSBs 30-40-30.jpg


You turn off the auto calculate, turn on the User entered distribution, enter the distribution percentages (30/40/30) and enter your first guess as to the fulcrum points and modifying these until you get Y1, Y2 and Y3 as close to equal as you can get (less than 0.1 different). Finding numbers that work is where experience comes in.

As to your diagram I think I would be tempted to put the Fulcrum centre line 0.5mm higher so you don't get the centre 2 fixed fulcrums are not right on the edge of the frames.

I will reply to your email shortly
You do not have the required permissions to view the files attached to this post.

User avatar
Jol Wilkinson
Posts: 1111
Joined: Mon Jul 21, 2008 7:39 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Jol Wilkinson » Thu Feb 11, 2021 1:21 pm

Richard,

I designed a set of etched CSB adaptors for John at LRM, to fit his horn block bearings but I think they will also fit the HL ones. The design diameter of the hole to fit over the horn block is 3.8mm. The etch includes some frame fittings which may enable you to bring the CSB down to the level you want.

I've attached the instructions and a scale copy of the etch to show what is provided. I've also attached a photo of the test build of the Webb modified Problem to show others what the model looks like and that any minor protuberance below the bottom edge of the frame may not be visible.

Hornblock spring adaptor instructions.pdf


HBA PR dwg.pdf


Hblox CSB adaptor.pdf


LRM LNWR Problem.JPG
You do not have the required permissions to view the files attached to this post.

User avatar
Winander
Posts: 842
Joined: Thu Mar 20, 2014 12:19 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Winander » Thu Feb 11, 2021 2:46 pm

Will,

Thanks very much - very useful to know more about how to use the spreadsheet.

I am inclined to see what effect the 30/40/30 weight distribution has on the pulling power and inclination to porpoise - as discussed in the CSB for a single driver thread (now correctly linked in my first post).

Jol,

Your etch looks useful so thanks for highlighting it. I agree that given its position along the edge of the frames, it is probably not going to be obtrusive but until your suggestion was struggling to see how it could be fastened. I have measured the boss on the HL bearings at 3.68 so with a little help from some etch cusp I may get a decent fit. Above all, I do not want to have to manufacture carrier tags as they need to be accurate and consistent.

What my drawing does not show was that the ashpan takes up most of the space between the centre and rear axles which is the most visible part of the frames. I like the idea of the surface mounted fulcrums and the range of possibilities they give. As this is very much an experiment in assessing tractive effort, it is possible the fulcrums may be moved.

Once again, thanks to you both.
Richard Hodgson
Organiser Scalefour Virtual Group. Our meeting invitation is here.

User avatar
Winander
Posts: 842
Joined: Thu Mar 20, 2014 12:19 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Winander » Thu Feb 11, 2021 6:44 pm

I needed to know the distance between the axle centre line and the various holes for the CSB wire (A, B &C) on the carrier etch - the one soldered to the bearing. I thought that measuring the High Level jig would give me that figure as it was a lot easier to measure than the carrier etch simply because the centre line of the bearing is marked on the jig.

It did not. The jig sets out the horizontal level of the CSB fulcrums (A, B or C) 0.5 mm below the carrier tags' holes. That allowance is the static deflection of the beam. Explained here by Chris Gibbons http://highlevelkits.co.uk/jigtutorial5.html and http://www.clag.org.uk/pics/beams/high-level-CSB-jig-and-hornblock-system.pdf and, in more depth, here by Ted Scannell http://www.clag.org.uk/beam-annex4.html.

I am glad I paid attention to that nagging feeling. Was I using the HL components in a conventional manner, all the adjustments are built into the jig and measuring is not necessary - "If the jig is used in conjunction with our own Hornblocks and CSB Tags, these anchor points will automatically line up with holes in the tags, allowing for 0.5mm deflection of the wire under the weight of the loco."

I am also mindful of the implications regarding the static deflection when the beam is under the axle, that is to add the static deflection 0.5mm to the distance between the axle centre line and the carrier tag hole.(Edit: WRONG see Keith's and Will's posts immediately below).
Last edited by Winander on Fri Mar 05, 2021 4:12 am, edited 1 time in total.
Richard Hodgson
Organiser Scalefour Virtual Group. Our meeting invitation is here.

User avatar
grovenor-2685
Forum Team
Posts: 3917
Joined: Sun Jun 29, 2008 8:02 pm

Re: LNWR Problem 2-2-2 CSBs

Postby grovenor-2685 » Thu Feb 11, 2021 7:06 pm

Winander wrote:I am also mindful of the implications regarding the static deflection when the beam is under the axle, that is to add the static deflection 0.5mm to the distance between the axle centre line and the carrier tag hole.

The static deflection at the axle is upwards, so as I see it the 0.5 mm would be added for the carrying wheels and subtracted for the drivers. Then the wire at the axle location will be 0.5 mm above the chassis anchor points.
Regards
Keith
Grovenor Sidings

User avatar
Will L
Posts: 2516
Joined: Sun Jul 20, 2008 3:54 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Will L » Thu Feb 11, 2021 7:56 pm

grovenor-2685 wrote:
Winander wrote:I am also mindful of the implications regarding the static deflection when the beam is under the axle, that is to add the static deflection 0.5mm to the distance between the axle centre line and the carrier tag hole.

The static deflection at the axle is upwards, so as I see it the 0.5 mm would be added for the carrying wheels and subtracted for the drivers. Then the wire at the axle location will be 0.5 mm above the chassis anchor points.


Keith is correct. With the fixed fulcrum line as on the original diagram, the fulcrum point tags on the carry axles should be 5mm above the axles centre and on the driving axle it would be 3mm below.

As the high level tags have holes set at 3, 4 and 5 mm from the axle centre line you should be able to use them as they come.

Edited to correct misinformation
Last edited by Will L on Fri Feb 12, 2021 5:36 pm, edited 1 time in total.

User avatar
Winander
Posts: 842
Joined: Thu Mar 20, 2014 12:19 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Winander » Fri Feb 12, 2021 5:03 pm

grovenor-2685 wrote:The static deflection at the axle is upwards, so as I see it the 0.5 mm would be added for the carrying wheels and subtracted for the drivers. Then the wire at the axle location will be 0.5 mm above the chassis anchor points.

Yes, thanks, I didn't take the time to understand the salient part of Ted Scannell's article and rushed to the wrong conclusion.

I also made another mistake. I needed to know the distance between the holes in the carrier tag and the axle centre line and, as the carrier tag does not mark that centre line and the jig does, I measured the latter. I overlooked the fact that the carrier tag enables the static deflection and the distance between the axle centre line and holes is 0.5mm longer. This illustration shows the difference. A top hat bearing is installed in the jig and a carrier tag placed over it. You can see that the half etches and holes in the carrier tag are further from the axle than the holes on the jig used to mark the fulcrum positions.

HL CSB Jig.jpg


I have always liked the idea of CSBs but never taken the time to fully understand the geometry, anticipating I could rely on the functioning components of the High Level jig system which work perfectly. I don't want my errors to discourage others from using them, as, in the conventional circumstances of a beam above the axles, a ruler is redundant because the jig sets everything out. My circumstances are not conventional as I have two axles below the beam and one above, so I need to employ a ruler.
You do not have the required permissions to view the files attached to this post.
Richard Hodgson
Organiser Scalefour Virtual Group. Our meeting invitation is here.

User avatar
Will L
Posts: 2516
Joined: Sun Jul 20, 2008 3:54 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Will L » Fri Feb 12, 2021 5:38 pm

Will L wrote:
grovenor-2685 wrote:
Winander wrote:I am also mindful of the implications regarding the static deflection when the beam is under the axle, that is to add the static deflection 0.5mm to the distance between the axle centre line and the carrier tag hole.

The static deflection at the axle is upwards, so as I see it the 0.5 mm would be added for the carrying wheels and subtracted for the drivers. Then the wire at the axle location will be 0.5 mm above the chassis anchor points.


Keith is correct. With the fixed fulcrum line as on the original diagram, the fulcrum point tags on the carry axles should be 5mm above the axles centre and on the driving axle it would be 3mm below.

As the high level tags have holes set at 3, 4 and 5 mm from the axle centre line you should be able to use them as they come.

Edited to correct misinformation

I'm glad your awake Richard because I clearly wasn't. My previous post has been corrected as above.

User avatar
Winander
Posts: 842
Joined: Thu Mar 20, 2014 12:19 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Winander » Fri Feb 12, 2021 9:30 pm

Will L wrote:I'm glad your awake Richard because I clearly wasn't. My previous post has been corrected as above.


Not me, Keith's keeping me on the straight and narrow ;)
Richard Hodgson
Organiser Scalefour Virtual Group. Our meeting invitation is here.

User avatar
Winander
Posts: 842
Joined: Thu Mar 20, 2014 12:19 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Winander » Thu Mar 04, 2021 10:35 pm

I have been considering the design of the tender drive. I want and think I can get the cardan shaft below the footplate. This is important because the as first built by Ramsbottom they had open cabs - just a front sheet. Jol's picture above has a full cab for the namby pambies :).

The cardan shaft is from Ultrascale attaching a High Level TendeRiser to a HL gearbox with a Drivestretcher extension to get things low enough. I have no idea where to put the two universal joints. Is there an optimum position? The only thing I think I saw was to keep the shaft on the motor the same length as the one on the gearbox i.e. the outer two of the three part drivetrain.

thanks in advance
Richard Hodgson
Organiser Scalefour Virtual Group. Our meeting invitation is here.

davebradwell
Posts: 1173
Joined: Fri Jan 18, 2019 3:48 pm

Re: LNWR Problem 2-2-2 CSBs

Postby davebradwell » Fri Mar 05, 2021 9:24 am

Don't bother getting too worried about where the universals end up - yes, you want to try and get equal angles but straying a little isn't the end of the world and practical issues of space will keep you confined. The couplings should be close to the gearbox bearings to ninimise overhang and if you end up with a long shaft, the chord it takes on bends might take up too much space. The shaft should be pretty straight on straight track. Don't bolt the motor down - mount it on something soft like double sided tape or bath sealant.

A bigger trap is restraining the gearbox torque re-action but let's tackle one thing at a time.

DaveB

User avatar
Winander
Posts: 842
Joined: Thu Mar 20, 2014 12:19 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Winander » Tue May 25, 2021 2:08 pm

Is there any merit in reducing the revs at the motor when using tender drive?

I read in the EMGS virtual exhibition demonstration on 'Scratchbuilding on the Computer' that the author found the brass bearings in the driveshaft 'constantly required lubrication' and he substituted small roller bearings. I'd appreciate the collective wisdom - would phosphor bronze bearings be better. IIRC Chris Gibbons recommends sintered bronze.
Richard Hodgson
Organiser Scalefour Virtual Group. Our meeting invitation is here.

User avatar
Will L
Posts: 2516
Joined: Sun Jul 20, 2008 3:54 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Will L » Tue May 25, 2021 4:53 pm

Winander wrote:Is there any merit in reducing the revs at the motor when using tender drive?

I read in the EMGS virtual exhibition demonstration on 'Scratchbuilding on the Computer' that the author found the brass bearings in the driveshaft 'constantly required lubrication' and he substituted small roller bearings. I'd appreciate the collective wisdom - would phosphor bronze bearings be better. IIRC Chris Gibbons recommends sintered bronze.

Surly you want the motor to spin at it designed speed, and vary the gearing so suit the size of the driving wheel and what top speed you want the loco to go at. Highlevel have a tool on their website to work out what gearing ratio you need.

How much you need to lubricate your bearings will depend a lot on what sort of running you want your loco to do. If its going to run all day and every day, then ball bearings may be needed, but the average 4mm loco which stands still a lot more than it runs is unlikely to ware out brass ones in its lifetime.

In 15 years of exhibition running (works out at about 8 days intensive running a year) I very rarely oiled anything. I made the mistake of a thorough oiling in the early days and spent much of the remaining 15 years having to clean the residue off the wheels and rails. In the latter years oil was applied extremely sparingly, the locos continue to run sweetly and wheel cleaning became much less of a chore.

davebradwell
Posts: 1173
Joined: Fri Jan 18, 2019 3:48 pm

Re: LNWR Problem 2-2-2 CSBs

Postby davebradwell » Tue May 25, 2021 5:43 pm

Ordinary brass isn't a great bearing material but it's still worrying if it's wearing in our applications - this suggests there's something wrong. I've loads of Chris's gearboxes running with simple brass bearings and they get very little lubrication, as Will says. There is a possibility that the wear is due to the endthrust of the worm being handled inappropriately and I always fit the bearings with flanges towards the worm and pack the space between with washers and spacers so there is only a few thou' backlash. The thrust should be taken by the whole diameter of the bearing flanges. These are the only bearings which might, beneficially be replaced by ball type but they should really be deep groove type. They'll need careful mounting or you'll be worse off than with plain ones.

The given speed of a motor is usually the no-load speed at which it develops zero power - max power is at around half that speed and at which it will probably still make an irritating whining noise. It is perfectly normal to use a much lower and quieter speed range than this if it is controlled properly with a feedback control device - most conveniently a quality decoder.

DaveB

User avatar
Winander
Posts: 842
Joined: Thu Mar 20, 2014 12:19 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Winander » Tue May 25, 2021 7:44 pm

Will L wrote:Surly you want the motor to spin at it designed speed, and vary the gearing so suit the size of the driving wheel and what top speed you want the loco to go at. Highlevel have a tool on their website to work out what gearing ratio you need.


Apologies, I didn't express myself at all well - I meant, is there any benefit to locating the gearing on the motor in the tender to reduce the revs of the cardan shaft, thereby leaving just the worm drive in the loco.

With a 7ft 9in driver I'll have to choose my motor/gearbox assembly carefully.
Richard Hodgson
Organiser Scalefour Virtual Group. Our meeting invitation is here.

Crepello
Posts: 158
Joined: Thu Jan 31, 2013 11:32 am

Re: LNWR Problem 2-2-2 CSBs

Postby Crepello » Tue May 25, 2021 8:49 pm

Winander wrote:
Will L wrote:Apologies, I didn't express myself at all well - I meant, is there any benefit to locating the gearing on the motor in the tender to reduce the revs of the cardan shaft, thereby leaving just the worm drive in the loco.

Unless you choose exotic reduction gears as the first stage, you will introduce gear whine.
Additionally, shaft power = constant x rpm x torque, so there is benefit in keeping the shaft in the high rpm part of the drive drain, so as to reduce the torque reaction between the loco and tender.

User avatar
Will L
Posts: 2516
Joined: Sun Jul 20, 2008 3:54 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Will L » Wed May 26, 2021 9:57 am

Crepello wrote:Additionally, shaft power = constant x rpm x torque, so there is benefit in keeping the shaft in the high rpm part of the drive drain, so as to reduce the torque reaction between the loco and tender.
As Crepello, and theory, says you should not reduce the shaft speed until it gets to the loco end. The torque reaction can result in the tender leaning visible, or in extreme cases, turning the tender on its side. One should point out this will only happen if the loco chassis is seeing a lot of resistance to turning its driving wheels, usually internal resistance but just possibly in terms of loads hauled. I practice a single driver is rather less likely to be troubled by such events than others with more driving wheels.

John Lewsey
Posts: 23
Joined: Sat Oct 22, 2022 1:09 pm

Re: LNWR Problem 2-2-2 CSBs

Postby John Lewsey » Mon Oct 24, 2022 1:56 pm

Hi, are there any updates on this engine
John

User avatar
Winander
Posts: 842
Joined: Thu Mar 20, 2014 12:19 pm

Re: LNWR Problem 2-2-2 CSBs

Postby Winander » Mon Oct 24, 2022 10:10 pm

It was purchased and is "maturing" in my kit stash of two items. Next on the list but I'm afraid I don't know when I'll start it.
Richard Hodgson
Organiser Scalefour Virtual Group. Our meeting invitation is here.


Return to “On My Workbench”

Who is online

Users browsing this forum: ClaudeBot and 1 guest