Turntable troubles

User avatar
Andrew Bluett-Duncan
Posts: 21
Joined: Tue Mar 01, 2016 7:47 pm

Turntable troubles

Postby Andrew Bluett-Duncan » Sat May 20, 2017 10:53 pm

Hello Everyone
Ive been installing a Chatham turntable mechanism under a 50' London Road Models turntable on my embryonic layout of Yeovil Pen Mill (set in 1922/3). Its not been a simple project exactly but I have got it working ok without an occupant. However as soon as I put a loco on the turntable it goes out of register in that it no longer lines up correctly on the two exit roads. Ive attached a couple of pictures to show what I mean.

In the first you can see the overall setting with the turntable aligned ok.
Yeovil Turntable  no load..jpg
Yeovil Turntable no load..jpg (157.49 KiB) Viewed 6106 times

In the second with a loco on board and after half a revolution the alignment has gone out by 3mm or 4mm!
Turntable misalignment.jpg
Turntable misalignment.jpg (109.94 KiB) Viewed 6106 times

Has anyone any ideas what maybe going wrong when the weight of a loco is on the turntable? For the moment this has me stumped!

Kind regards
Andrew

User avatar
Paul Townsend
Posts: 775
Joined: Mon Sep 14, 2009 6:09 pm

Re: Turntable troubles

Postby Paul Townsend » Sun May 21, 2017 5:23 am

I had a similar issue with the Merg tt mechanism . All my roads approached the turntable on a slight curve and the resulting sidethrust took up the slack in the gear chain.
I tinkered with mechanical locks but found it too much faff.
I abandoned the gear mechanism for toothed belts and thus needed a microstepping motor drive so abandoned the merg electrics too.

Do you get a difference according to the loco approaching on the straight or curved route?

User avatar
grovenor-2685
Posts: 3167
Joined: Sun Jun 29, 2008 8:02 pm

Re: Turntable troubles

Postby grovenor-2685 » Sun May 21, 2017 8:29 am

There are two separate problems here,
1. The turntable stopping in the wrong place after turning, which is what I understood andrew to be reporting
2. The turntable moving when a loco is driving on or off due to backlash in the geartrain and no lock.

To understand 1 we need some idea of how the drive, and particularly the detection of position, is arranged.
Regards

User avatar
Guy Rixon
Posts: 602
Joined: Sun Mar 27, 2011 6:40 pm

Re: Turntable troubles

Postby Guy Rixon » Sun May 21, 2017 9:25 am

If it's the mechanism in this video:



then I can see a problem. Look at the detent assembly down at the bottom: it's flexing and rattling all over the place! From the description on the MRD site, I gather that it depends on a tapered, sprung detent to align and lock the table in alignment when stopped. This is good, but only if the detent spring is powerful enough to force the right alignment and if it's rigidly mounted. Perhaps the loaded table has enough friction to stop the detent from working?

garethashenden
Posts: 281
Joined: Tue Apr 07, 2015 9:41 pm

Re: Turntable troubles

Postby garethashenden » Sun May 21, 2017 4:11 pm

Guy Rixon wrote:If it's the mechanism in this video:



then I can see a problem. Look at the detent assembly down at the bottom: it's flexing and rattling all over the place! From the description on the MRD site, I gather that it depends on a tapered, sprung detent to align and lock the table in alignment when stopped. This is good, but only if the detent spring is powerful enough to force the right alignment and if it's rigidly mounted. Perhaps the loaded table has enough friction to stop the detent from working?


Can anyone identify the locomotive? It looks German from the paint scheme but that's as far as I've gotten. I really like the pointy smoke box door.

User avatar
Noel
Posts: 1130
Joined: Wed Jun 23, 2010 1:04 pm

Re: Turntable troubles

Postby Noel » Sun May 21, 2017 5:13 pm

I think its a Marklin HO German DB BR 18 478. The original was one of one variety of Bavarian S 3/6, later the German BR 18 4-5. The original is preserved. https://en.wikipedia.org/wiki/Bavarian_S_3/6
Regards
Noel

dal-t
Posts: 621
Joined: Mon Jan 11, 2010 8:06 pm

Re: Turntable troubles

Postby dal-t » Sun May 21, 2017 5:20 pm

Class BR 18 Rheingold? Just guessing ... (too slowly, apparently)!
David L-T

User avatar
Tim V
Posts: 2247
Joined: Tue Jul 29, 2008 4:40 pm

Re: Turntable troubles

Postby Tim V » Sun May 21, 2017 5:43 pm

I remember Mike Sharman commenting on turntables, if he had a pound for everyone he'd built but couldn't get to work....

What hope is there for the rest of us?
Tim V
Scalefour News Editor

User avatar
Guy Rixon
Posts: 602
Joined: Sun Mar 27, 2011 6:40 pm

Re: Turntable troubles

Postby Guy Rixon » Mon May 22, 2017 9:54 am

It's feasible if one uses appropriate engineering. Make the detent pin engage with the turntable carriage, not the indexing disc, and all the lost motion goes away. For better alignment, use a ball-ended pin in a conical notch, so that it aligns the carriage end both vertically and horizontally. As I described before: https://www.scalefour.org/forum/viewtopic.php?f=5&t=4348&p=39505&hilit=conical+detent#p39505.

Of course, this approach requires the carriage structure, which include the detent notch/cone to be part of the working mechanism; it can't be purely cosmetic. However, I see no reason why a cosmetic "body" can't be arranged over a rigid, working carriage.

User avatar
Guy Rixon
Posts: 602
Joined: Sun Mar 27, 2011 6:40 pm

Re: Turntable troubles

Postby Guy Rixon » Mon May 22, 2017 10:31 am

Another thought about Andrew's specific problem...

Suppose that the detent assembly in Andrew's turntable isn't loose. I would then suspect the alignment of the microswitch on the detent arm. This is supposed to cut the motor power when the detent pin reaches the target notch and to allow the detent spring to make the fine alignment. If, for some reason, the switch is opening too early or too late, then the pin won't go into the notch and the table will be loose and significantly misaligned.

User avatar
John McAleely
Posts: 1114
Joined: Wed Oct 01, 2008 5:08 pm

Re: Turntable troubles

Postby John McAleely » Mon May 22, 2017 12:24 pm

Does it behave the same way in both turning directions?

User avatar
Andrew Bluett-Duncan
Posts: 21
Joined: Tue Mar 01, 2016 7:47 pm

Re: Turntable troubles

Postby Andrew Bluett-Duncan » Mon May 22, 2017 10:13 pm

Paul Townsend wrote:I had a similar issue with the Merg tt mechanism . All my roads approached the turntable on a slight curve and the resulting sidethrust took up the slack in the gear chain.
I tinkered with mechanical locks but found it too much faff.
I abandoned the gear mechanism for toothed belts and thus needed a microstepping motor drive so abandoned the merg electrics too.

Do you get a difference according to the loco approaching on the straight or curved route?


Hello Paul
Thanks for your thought but I don't yet have power to the track so I've just been placing the loco on the turntable and switching on. This has been enough to give me the problem I outlined. I haven't even begun to consider the effect of curves on the problem...! I'm going to reply to the others separately, I'm unused to this forums workings and hadn't appreciated that anyone had answered until I checked tonight and found a bevy of replies.
Kind regards
Andrew

User avatar
Andrew Bluett-Duncan
Posts: 21
Joined: Tue Mar 01, 2016 7:47 pm

Re: Turntable troubles

Postby Andrew Bluett-Duncan » Tue May 23, 2017 12:06 am

Guy Rixon wrote:If it's the mechanism in this video:

then I can see a problem. Look at the detent assembly down at the bottom: it's flexing and rattling all over the place! From the description on the MRD site, I gather that it depends on a tapered, sprung detent to align and lock the table in alignment when stopped. This is good, but only if the detent spring is powerful enough to force the right alignment and if it's rigidly mounted. Perhaps the loaded table has enough friction to stop the detent from working?


Hello Guy
Thanks for your thoughts on the locking mechanism. The detent /catch which engages in the control disc was indeed like the one in the video you put up. And I found it gave too much slack as you thought, so I shimmed it out which has drastically reduced unwanted movement. But it didn't stop the problem of misalignment occurring. But the control disc being smaller than the diameter of the turntable exaggerates any error by a significant factor so that doesn't help. Having said that I don't think that s the problem.

I'm beginning to wonder if the weight of the loco is sufficient to make the connection between the turntable and the mechanism slip. At present it's fixed with a single 8BA bolt through a collar onto a flat on the drive shaft. And I'm using a brass bolt and am wondering if that's the problem should I be using steel as it's harder? Or two bolts as I believe Mark Tatlow did on his installation? The othe area that a similar problem could occur is that the control disc also attaches to the drive shaft with two steel bolts opposit one another bearing directly on the shaft as well.

So a little more investigation is needed I think. I'll report back.
Kind regards
Andrew

User avatar
Andrew Bluett-Duncan
Posts: 21
Joined: Tue Mar 01, 2016 7:47 pm

Re: Turntable troubles

Postby Andrew Bluett-Duncan » Tue May 23, 2017 12:24 am

John McAleely wrote:Does it behave the same way in both turning directions?


Hello John
No it doesn't. At present I've only managed to get a reliable performance out of it in one direction only and without the weight of a loco on it. If I change direction it goes out of alignment straight away

So continuing the thought that the loco weight is upsetting it somehow, it's just occurred to me that when the mechanism stops there's quite a bit of weight/momentum flying round that has to be stops in an instance. I think it's looking more and more like a lack of grip on the drive shaft.

Although having said that it doesn't answer why it goes out of register when reversed.

First things first I'll see what I can do to increase the grip on the shaft

Thanks for your question
Andrew

User avatar
Mark Tatlow
Posts: 706
Joined: Tue Dec 02, 2008 11:24 pm

Re: Turntable troubles

Postby Mark Tatlow » Tue May 23, 2017 1:04 pm

Andrew,

I have been struggling with the same drive (https://highlandmiscellany.com/tag/turntable/) and although there are a number of aspects I like about it, there are some fundamental flaws too. Chief of these is that the arrangement of the locking plunger (what is refered to as a detent above) into the disk has a degree of slackness to it such that in normal circumstances it engages slightly to one side of the gap in the disk depending upon which way the gap is approached (ie which direction the turntable operates in).

This means that if the turntable is always operated in the same direction, this problem will not manifest itself as it will always operate a touch early and you merely align the rails to suit. It is where there is the need to operate in both directions (more or less inherent in all but a straight in-straight off arrangements) that this is going to manifest itself.

I do also wonder whether any mechanism that drives through a gear mechanism attached to the pin is going to be an issue. I think there is too much gear lash and you need to drive on either the ends of the deck or on a large disk on the underside like the Midland Railway Centre version http://www.midrailcentre.com/turntable-motorising-units.

At the moment I am still battling with my turntable (although not presently!) and I am resigned to coming up with a plunger mechanism that engages a pin into a socket at the ends of the turntable deck to confirm the alignment. These will probably be operated by servos and may well also seek to pass power to the rails. It is galling to add this to the drive mechanism but if it works then I will live with it and if not then it will be back to the drawing board.

However, what I would say is that if the problem only occurs after half a revolution it may be that the pin is not quite central on the the deck. Alternatively, if it works well without a loco (on the half and full revolution) but is a problem with then it may be the weight of the deck making it a bit slower going around such that the locking plunger operates a tad quicker than it might otherwise?
Mark Tatlow

Armchair Modeller
Posts: 1131
Joined: Thu Jan 06, 2011 7:34 pm

Re: Turntable troubles

Postby Armchair Modeller » Tue May 23, 2017 1:25 pm

Having thought long and hard about mine, I came to the same conclusion - that a mechanism locking the end of the turntable deck immediately below rail level is the only way of getting good, reliable alignment.

User avatar
Andrew Bluett-Duncan
Posts: 21
Joined: Tue Mar 01, 2016 7:47 pm

Re: Turntable troubles

Postby Andrew Bluett-Duncan » Tue May 23, 2017 10:26 pm

Mark Tatlow wrote:Andrew,

I have been struggling with the same drive (https://highlandmiscellany.com/tag/turntable/) and although there are a number of aspects I like about it, there are some fundamental flaws too. Chief of these is that the arrangement of the locking plunger (what is refered to as a detent above) into the disk has a degree of slackness to it such that in normal circumstances it engages slightly to one side of the gap in the disk depending upon which way the gap is approached (ie which direction the turntable operates in).


Hello Mark
Thanks for your various thoughts. I'm interested in your observation about the locking plunger stopping in subtly different places dependent on the direction of travel.
Quite by chance I removed some of the wrong fingers when I was trying to get the mechanism to stop "exactly " 180 degrees opposite the point that it had started. This proved tricky and in the process required me to put right what I'd done incorrectly. To fill the gap where id removed the fingers in error, I cut a piece of scrap brass to fill it and soldered it in position. Then instead of cutting a U shape in the disc to restore the stop point, I cut a V shape roughly the same shape as the locking plunger. This seems to have the benefit of not really being able to stop in more that one place assuming the plunger goes fully home. I'm wondering therefore should I do the same thing on the opposite side of the control disc?

So thank you for that observation, I think I may give it a try and see what changes occur....
I'll report back on my experiences

Kind regards
Andrew

junctionmad
Posts: 247
Joined: Tue Dec 08, 2015 3:38 pm

Re: Turntable troubles

Postby junctionmad » Wed May 24, 2017 10:18 am

After several tests , I'm also in the camp of the solenoid driven conical locking pin idea ( not a ball detent by the way). In my case an N20 gear head motor driving through a rubber timing belt , with an arm underneath carrying the solenoid and the two sensors ( approaching the locking point from either direction )

Once the stop signal is detected , a timer brings the table to a stop within the engagement radius of the centring pin . The soloenid engages and there is enough gearbox slack to allow the table to be centred by the pin ( the next iteration will use a linear servo, as the solenoid is too loud and abrupt )

The table is the locked and centered both horizontally and vertically

The next iteration will have the solenoid mounted on the baseboard and the cone on the table as this turntable has only two positions and it's simplifies wiring to the table. I also intend to mount the solenoid/linear servo on an adjustable mount then allow me to fine position horizontally the stopping position.

This is after tests with steppers and indexers and attempts at fine sensor positioning

The locking systen is the only way to guarantee repeatable positioning that doesn't move as the locos move etc ( or someone bumps the baseboard !! )

Lindsay G
Posts: 130
Joined: Thu Aug 07, 2008 2:16 am

Re: Turntable troubles

Postby Lindsay G » Fri May 26, 2017 10:39 am

I'm also presently deliberating over a means of turntable operation for that within Burntisland's roundhouse. The means of stopping it at the correct location is now onto its third iteration.

Overall control is via an Arduino or Arduinos which control many other aspects as well. As far as stopping is concerned, current thinking is that is that when a wheel on the turntable completes a circuit the motor will be stopped. That aspect has yet to be proven!

This does avoid backlash present in the gearing, but that backlash does allow some to and fro movement in the turntable. I'm not sure if this could result in the turntable being moved by a loco being driven on or off the turntable causing derailment. If that were to prove to be the case, I was thinking of introducing a pointed pin being raised by, say, a servo into a V grooved area on the turntable which both ensures it being located correctly and holding it in place. This doesn't sound too adrift from ideas put forward in previous threads.

All to be proven or otherwise in due course!

Lindsay

User avatar
Russ Elliott
Posts: 930
Joined: Thu Jun 02, 2011 6:38 pm

Re: Turntable troubles

Postby Russ Elliott » Fri May 26, 2017 11:31 am

In a sector plate design, I came to the conclusion that using a (horizontal, in my case) tapered locking pin could use the backlash in the sector plate traverse to advantage. The pin movement, driven by a 4:1 arm from a screw mechanism, will be approx 40mm, and will be required to nudge and lock the 2.5kg sector plate into final position.

sector-lhs.gif

Engine turntables would not need this degree of locking strength.

Hubert Carr
Posts: 2
Joined: Tue May 23, 2017 1:02 pm

Re: Turntable troubles

Postby Hubert Carr » Tue May 30, 2017 12:27 pm

I have been steered to this site by a fellow member of SLAG, as I was the one who found the gentleman from America who devised this mechanism. I was impressed with his sample, however It was not set up to any actual tracks, so the backlash problems were not apparent. We set a mechanism up for the SLAG project, and found we had the same problems as Andrew and Mark experienced. Our supplier came up with a bodge solution of fitting side 'wings' onto the slide which carries the 'interupter' to the index wheel. This improved things but was not a proper answer. There were some other potential problems with the mechanism, and I stopped selling them until I had sorted it. However for those of you who bought the mechanism, I have come up with a solution to taking the 'slop' out of the 'slide'. So I shall be making up a replacement mechanism, which I will distribute to all those who require one (at no cost). But I do not have a record of those who bought the original mechanism from an exhibition. Nor is it easy to locate others who bought it by post. So if you are one of those, please get in touch. When I have completed the adaptation, I will post a photograph on the Forum. My deadline is the Society AGM.

User avatar
Andrew Bluett-Duncan
Posts: 21
Joined: Tue Mar 01, 2016 7:47 pm

Re: Turntable troubles

Postby Andrew Bluett-Duncan » Tue May 30, 2017 11:16 pm

Hello Hubert
Thanks very much for your reply and your helpful attitude in trying to resolve the issues that we've experienced. I have managed to shim out 90% + of the movement in the slid control that was one of the problems, but I could well imagine that some form of wall / keeper that stopped the slider form wobbling back and forth would be the ultimate in stopping any wobble dead in its tracks!

Another problem that Ive also experienced is finding the finger at exactly 180 degrees away from any given point on the indexing disc to achieve an exact 180 degree turn? Having said that I just realised that I've not tried counting the number of fingers to see if there's an obvious solution to this question...perhaps you know off the cuff if there is?

Anyway I look forward to hearing more from you as and when you have the mod ready use.

As an aside is there a possibility of you producing with Peter a better mechanism based on the current design (which has some great qualities) that might over come some of the weaknesses of the current one. Chief among these is the small size of the indexing disc that exaggerates any problem significantly, the bigger the turntable the worse the problem gets. Judging by the response to this post there seems to be a demand for a good well designed reasonably priced indexing turntable mechanism.

Kind regards
Andrew

dal-t
Posts: 621
Joined: Mon Jan 11, 2010 8:06 pm

Re: Turntable troubles

Postby dal-t » Wed May 31, 2017 7:27 am

Here's an off-the-wall thought on the indexing issue. My Facebook feed keeps getting filled by adverts for a toy vehicle system that will follow a randomly drawn black line (apparently anywhere). I haven't succumbed to buying one yet (partly because I'm pretty sure how the domestic authorities will react to my applying a squiggle of black felt-tipped pen all over the living room tiles), but when I do it's inevitable I'll tear the mechanism apart to see what's going on inside. On the face of it there must be an optical sensor and related steering circuitry, all at a knock-down price. Translate that to an optical indexing disc and you should have the ability to stop anywhere you choose to put a mark, without the need for first taking out a second/third/fourth mortgage. There may be a need for a bit of fancy switching to get the thing to start again or reverse once it's found an index mark, and it would make sense to have a good hoover before each operating session, to make sure dust/dead flies don't cause any unfortunate incidents, but I like the idea because it sounds like a good 21st century solution to a 19th century problem.
David L-T

User avatar
Guy Rixon
Posts: 602
Joined: Sun Mar 27, 2011 6:40 pm

Re: Turntable troubles

Postby Guy Rixon » Wed May 31, 2017 1:45 pm

dal-t wrote:Here's an off-the-wall thought on the indexing issue. My Facebook feed keeps getting filled by adverts for a toy vehicle system that will follow a randomly drawn black line (apparently anywhere). I haven't succumbed to buying one yet (partly because I'm pretty sure how the domestic authorities will react to my applying a squiggle of black felt-tipped pen all over the living room tiles), but when I do it's inevitable I'll tear the mechanism apart to see what's going on inside. On the face of it there must be an optical sensor and related steering circuitry, all at a knock-down price. Translate that to an optical indexing disc and you should have the ability to stop anywhere you choose to put a mark, without the need for first taking out a second/third/fourth mortgage. There may be a need for a bit of fancy switching to get the thing to start again or reverse once it's found an index mark, and it would make sense to have a good hoover before each operating session, to make sure dust/dead flies don't cause any unfortunate incidents, but I like the idea because it sounds like a good 21st century solution to a 19th century problem.


It's a possibility. You still need something to keep the table in position once it's got there. If you don't have a locking pin, then you either have the drive continually servoing, or you make the drive train stiff enough (and the motor powerful enough) that static friction holds the table in place.

Also, how good is the position encoder in a toy car? If I was designing one of those, I'd consider "follows the line to ±5mm" to be plenty accurate enough and anything better to be a waste of money. You'd need a seriously big index-disc to overcome that kind of uncertainty.

User avatar
Guy Rixon
Posts: 602
Joined: Sun Mar 27, 2011 6:40 pm

Re: Turntable troubles

Postby Guy Rixon » Wed May 31, 2017 2:43 pm

Pushing a bit on encoders, the problem with a railway turntable is that we need a high accuracy in angular position. If the radius of the turntable is r, the tolerance on tangental misalignment is t and the resolution of the encoder is q (in increments per rev, and making the optimistic assumption that one can stop the table to ±1 increment), then (I think) q = 2*Pi*r/t. Putting r=140mm (i.e. 70ft table, 4mm scale) and t=0.5mm gives q=1,760. If you want ±0.25mm alignment then q=3,519 or better. This is quite a high resolution!

For reference, cheap shaft encoders exist, but not very good ones. The best I could find by Googling was http://uk.farnell.com/broadcom-limited/aedb-9140-a13/encoder-3channel-500cpr-8mm/dp/1161087?mckv=YO18ZYDl_dc%7Cpcrid%7C78108376509%7C&gross_price=true&CATCI=pla-41477300408&CAAGID=14406255429&CMP=KNC-GUK-GEN-SHOPPING&CAGPSPN=pla&gclid=CIrFjcOkmtQCFc-6GwodnpYKFw&DM_PersistentCookieCreated=true&CAWELAID=120173390000244931 which gives q=500 for about £25.

Conversely, here's a shaft encoder with (apparently) better resolution, but it costs much more: http://uk.rs-online.com/web/p/rotary-encoders/2914349/.

Note that these encoders are not complete systems. One would have to devise and build the electronics that use them (therefore add cost of parts + black chickens for sacrifice, possibly also psychiatrist's fees during subsequent recovery).

A better system for setting a large-radius wheel to high accuracy might be a linear encoder wrapped around the rim, e.g. a magnetic encoder reading a tape fixed to the index wheel. These things exist, but I haven't found a reading head for under £100. However, since we only need, say, ±100 um accuracy which is very low, the cheapest heads would probably do.


Return to “Electrics”

Who is online

Users browsing this forum: No registered users and 1 guest