Tortoise Auxiliary Switches

Discuss the prototype and how to model it.
andrew jukes

Tortoise Auxiliary Switches

Postby andrew jukes » Fri May 15, 2009 5:10 pm

From time to time I've heard comments about the weakness/poor design of the Tortoise auxiliary switches but have pressed on regardless believing they couldn't be that bad as the Tortoise has become such a standard piece of equipment in the US. My plan has been to use them to switch crossing polarity.

Fairly obviously, a pcb based slider switch - which is what the two changeover switches are - is never going to be suitable for very high currents or a very long life. I convinced myself that high currents (and in particular, switching under load) could largely be avoided on a fully interlocked system with carefully designed overlap track circuits. The number of turnouts that will get used intensively enough to wear out is also small and so not a worry.

I have now installed several Tortoises and wired the crossing switches, testing the electrical resistance as I go. I have been surprised at the resistance the switches seem to add and how it can fluctuate - typically from 0.2 up to over 1.0 ohm. 'Running the switches in' sometimes seemed to help, but not predictably.

I've now taken one apart, cleaned the pcb with a fibre glass brush and bent the phosphor bronze wipers a little to increase the contact pressure, and the resistance is now where it should be - effectively zero. So far so good. Apart from smoother normal running (no voltage drop at the crossings) the modified Tortoise switches should be less vulnerable to high currents.

But - there had to be a but! - I am amazed at how badly aligned the wipers are with the pcb contacts. Built in to the moulding or pcb is a mismatch of over 1mm. One of the switches (the one using tracks 5, 6 and 7) seems in danger of cross-connecting the two connections that are being changed over.

I wonder if others have been down this path and what their conclusions were. At the moment mine are that the tracks 2, 3 and 4 switch is useable, provided it is cleaned and the wipers bent but that the other one is too risky, and that for complex formations where two changeover switches are needed, a 12V relay connected in parallel with the Tortoise is essential.

Regards

Andrew Jukes

User avatar
grovenor-2685
Forum Team
Posts: 3918
Joined: Sun Jun 29, 2008 8:02 pm

Re: Tortoise Auxiliary Switches

Postby grovenor-2685 » Fri May 15, 2009 6:26 pm

I have three tortoises in use, I have not had any problems with the switch contacts, but then I have never bothered to measure the resistance, I have never noticed any effect on train running.
But that said most of my locos will span the crossings anyway and work without any switch, the 08 shunter does need the feed to the crossing and has not shown any problems.
On the other hand I have had two mechanical failures of the tortoises, possibly provoked by my non-standard mechanical connection, but on dismantling its obvious that the bearings for the final drive gear are inadequate for the torque that the motor can produce, I had to restore them with brass bearings.
On any new work or repairs now I will use Servos.

Regards
Regards
Keith
Grovenor Sidings

User avatar
Tim V
Posts: 2867
Joined: Tue Jul 29, 2008 4:40 pm

Re: Tortoise Auxiliary Switches

Postby Tim V » Fri May 15, 2009 7:36 pm

I have been using Tortoises for a number of years without problems. As opposed to the problems I had with Lemaco/Fulgarex.

What is all this worry about a couple of ohms, or are you using DC control, where it just might make a slight difference?
Tim V
(Not all railways in Somerset went to Dorset)

andrew jukes

Re: Tortoise Auxiliary Switches

Postby andrew jukes » Fri May 15, 2009 9:34 pm

What is all this worry about a couple of ohms, or are you using DC control, where it just might make a slight difference?


Tim

From the mutterings I have heard, the alleged failures were under fault conditions. If there is next to no resistance anywhere else in the circuit, 1 ohm could be enough to create problems at the contact point. If this is a real issue, it would be a bigger one under DCC.

Additionally, there are track layouts where you end up wanting to use two switches in series. Having that much (unpredictable) contact resistance around doesn't feel like the way to build a reliable railway.

In the end, though, I was more concerned about the misalignment than the contact resistance - which I can deal with. I guess I can take some comfort on that from your and Keith's lack of problems.

Regards

Andrew

User avatar
grovenor-2685
Forum Team
Posts: 3918
Joined: Sun Jun 29, 2008 8:02 pm

Re: Tortoise Auxiliary Switches

Postby grovenor-2685 » Fri May 15, 2009 9:57 pm

From the mutterings I have heard, the alleged failures were under fault conditions. If there is next to no resistance anywhere else in the circuit, 1 ohm could be enough to create problems at the contact point. If this is a real issue, it would be a bigger one under DCC.


Generally this is not correct, DC or DCC if you are just switching crossings the loading is never more than one loco, and that usually for a very short time, and if the loco stops on the crossing the current is even less. Fault currents are usually much better managed on DCC boosters/power stations than with DC controllers. If you drive onto a wrongly set point the current will be cut off very quickly and stay off until you resolve the problem, I have done it often enough and nothing has been damaged yet.

The danger comes when the users deliberately disable the protective cut out. eg. by using a car lamp in series with the supply. This stops the cut out operating by limiting the current to a level below that where the cut out operates, but still enough to cause damage if you don't react very quickly. When you see car lamps being recommended for DCC ignore the advice.
Regards
Regards
Keith
Grovenor Sidings

andrew jukes

Re: Tortoise Auxiliary Switches

Postby andrew jukes » Fri May 15, 2009 10:31 pm

Thanks for the comments.

Trawling back through the email lists, the suggestion of problems was principally from Martin Wynne and it did relate to damage under fault conditions. In the absence of any other bad experience, it sounds like I should just keep on installing Tortoises straight from the box and stop worrying!

Andrew

DavidM
Posts: 74
Joined: Sun Jul 20, 2008 7:32 pm

Re: Tortoise Auxiliary Switches

Postby DavidM » Sat May 16, 2009 8:01 am

Andrew,

I hate to pour further cold water on the idea of using Tortoises as, like you, I have assumed they are the way to go. I have recently repaired two units (out of seven currently installed on a small layout) when the internal switch contacts literally came adrift from the operating arm - it appears as if the small moulded pips holding the copper switching contacts just sheared (or melted!) off. I can't find a particular cause for this - and I don't think it's related to high current during a fault situation, as the controller in use was a Pentroller. Anything's possible of course, but it was (in both cases) the contact being used to do the switching, the non-switching contact remained undamaged. It may be a batch problem - I have several from a later batch which I will substitute if and when it recurs. The repair was easily effected using a couple of very small self-tapping screws (1mm x 3mm if I recall correctly).

From my experiences so far, which I accept are not typical of other users, I think Martin's suggestion of using a relay might have some merit - at least for heavily used units.

David Murrell

Andy G
Posts: 69
Joined: Wed Mar 04, 2009 2:25 pm

Re: Tortoise Auxiliary Switches

Postby Andy G » Sat May 16, 2009 11:31 am

On Slattock's only outing so far we noticed potential issues with the switching and on examination back at Dean Hall found a build up of dirt. The decision was taken to irradicate them and all motors have now had the copper wipers removed and microswitches screwed to the outside of the motor in alignment with the motor arm. Drastic action but we felt it was better to fit better quality switches before the layout developed too far than to try and retrospectively fit them when it eventually made it out onto the exhibition circuit. We've subsequently found that the Wirral lads run their switches in parallel with no issues, thus halving the current through each contact.

Andy
Cheers
Andy
EBMA Hobby & Craft
https://www.ebmahobby.co.uk

andrew jukes

Re: Tortoise Auxiliary Switches

Postby andrew jukes » Sun May 17, 2009 2:33 pm

I've been looking at the detail of the design more closely and have come to what might seem a rather extreme conclusion.

If you look at the marks made by the wipers on the pcb tracks, it is clear that for three of the four wiper pairs, one of the pair is running right on the edge of the track. If pressure on the wiper pairs is uneven (likely, I would have thought) and the more heavily loaded is the one on the edge of the track, problems seem inevitable. If I were just fitting half a dozen Tortoises, the risk might be worth taking but the thought of installing 50+ and then having to revisit the switching arrangements is just not bearable.

So, what to do? I could forget the switches and use relays or microswitches instead - but that is likely to involve quite a lot of extra work/time, and with relays some would be 'under the baseboard' time. Alternatively, I could try to make the switches work properly - taking the view that there is no reason why a well-executed enclosed pcb/wiper switch shouldn't be fine for this application.

Moving the wipers along the actuating arm might help but would be fiddly (moving them 1.0mm), and the pcb design is so poor that their position would still be a compromise. So that leads to the pcb. Swapping the pcb for a better design simply involves unsoldering the two motor connections, removing the existing pcb, fitting a new one and putting the Tortoise back together. To do a complete re-engineering job, a 12BA screw and nut to clamp each wiper to the actuating arm might be worth adding.

Attached is the pcb design I've arrived at. It involves a different arrangement of connections but that seems a small price to pay for full-width tracks placed accurately under the pairs of wipers. I've also ignored the edge-connector capability of the existing pcb and reduced the overall height of the Tortoise by about 2mm (every little helps!).

Incidentally, Martin Wynne's comments on Tortoises failing if used in a switched (as opposed to stall) mode is the sort of thing that, with a supplier's hat on, makes me despair. Products shouldn't get a bad name because they don't do what they're not designed to do.

Andrew Jukes
You do not have the required permissions to view the files attached to this post.

User avatar
grovenor-2685
Forum Team
Posts: 3918
Joined: Sun Jun 29, 2008 8:02 pm

Re: Tortoise Auxiliary Switches

Postby grovenor-2685 » Sun May 17, 2009 3:10 pm

Martin Wynne's comments on Tortoises failing if used in a switched (as opposed to stall) mode is the sort of thing that, with a supplier's hat on, makes me despair. Products shouldn't get a bad name because they don't do what they're not designed to do.


I too think that one unfair, Tortoises are not designed to operate in a switched mode, they are designed to operate in the stall mode.

If you are going to make new circuit boards I suggest making the connections all in line to take a standard PCB header allowing the connections to be made via crimp connectors in an 8 way shell, much better than soldering direct.

But it does seem to be a lot of effort, an external relay would surely be less work and more reliable at the end of the day.
Regards
Regards
Keith
Grovenor Sidings

DavidM
Posts: 74
Joined: Sun Jul 20, 2008 7:32 pm

Re: Tortoise Auxiliary Switches

Postby DavidM » Sun May 17, 2009 9:29 pm

I would agree with Keith's suggestion about using a standard PCB header / 8 way shell as a robust solution - we must assume ultimately that any turnout drive system may fail, however rarely. They sometimes have to be replaced in a hurry and anything that reduces inconvenience and time under the baseboard is worthwhile. :evil:

I note that Circuitron provide a "unprecedented 9 year warranty" on the Tortoise - see their website http://www.circuitron.com/index_files/Tortoise.htm

Is this the manufaturer's way of acknowledging a flawed design in the switching mechanism? In this case, maybe we should just return the failed units to Circuitron - the onus should be on them to redesign the flawed PCB. If enough people return them, we might end up with a product that does the job it was designed for. :shock:

David

andrew jukes

Re: Tortoise Auxiliary Switches

Postby andrew jukes » Sun May 17, 2009 10:33 pm

It's interesting how the type of layout influences the trade-offs made. I'm obsessed with how long it takes to install each of a large number of turnouts and finding things that minimise the time taken and the risk of having to go back and rework anything. From that point of view, PCB headers and 8-way shells are more bits and pieces and more time which I think would be better spent on making the installation itself as robust and reliable as possible.

The same applies to Keith's preference for relays. The mods I'm contemplating will take less than 10 mins per Tortoise and I think deal with the fundamental problems and give the level of reliability I'm after. The snag is cost, but I think the mods will be value for money in terms of time saved. I hope I'm not eventually proved wrong!

As for pursuing Circuitron, I think that's a certain way of spending more time (and money). What is most depressing is that, having voided the warranty and exposed the innards, one finds that the pcb is marked 'REV. B' - so that probably means that it's their third attempt. I think they must have regarded the layout of the connections as more important than the quality of the switches........

Andrew

User avatar
Dave K
Posts: 247
Joined: Sun Jul 20, 2008 6:11 pm

Re: Tortoise Auxiliary Switches

Postby Dave K » Mon May 18, 2009 7:20 am

andrew jukes wrote:
If this is a real issue, it would be a bigger one under DCC.

I was talking to the chap on the ZTC stand at Wells some time ago and he said that if you are using the internal switching on the Tortoise the higher current used on DCC would over time cause the contacts on the PCB board to 'smear' (his word) and you would loose the switching capability. He therefore recommended using a stand alone relay for the switching current to the ‘Vee’.

Therefore, taking his advise I have used a relay/Tortoise combinationon on all the points on Hallatrow.

andrew jukes

Re: Tortoise Auxiliary Switches

Postby andrew jukes » Mon May 18, 2009 8:30 am

I was talking to the chap on the ZTC stand at Wells some time ago and he said that if you are using the internal switching on the Tortoise the higher current used on DCC would over time cause the contacts on the PCB board to 'smear' (his word) and you would loose the switching capability. He therefore recommended using a stand alone relay for the switching current to the ‘Vee’.


This sounds like really muddled thinking.

Surely, because roughly the same power is used at a higher average voltage, the currents drawn by a single power unit in DCC are on average lower than for DC? Fault currents may be higher but, as Keith points out, overload protection is usually better.

Used properly, whatever the currents are, the switches should not be switching the flowing current. So whether DCC or DC, any 'smearing' should be a purely mechanical question of wear, unaffected by current levels. I can see ways the existing design is vulnerable to a build-up of conductive debris, but that would be common to all control modes.

If the switches are used to switch current on (most likely when switching onto an already shorted pair of rails), it's hard to picture how 'smearing' would happen. It would be a different matter if significant currents were frequently switched off, producing a nice spark at the edge of the pcb track.

Andrew

andrew jukes

Re: Tortoise Auxiliary Switches

Postby andrew jukes » Sun May 24, 2009 11:13 pm

Just thought it might be helpful to outline where I think this is going.

I've ordered a test pcb of (almost) the design I posted earlier and will use that to make final adjustments in the fit of the pcb in the Tortoise housing and the alignment of the wipers on the pcb tracks. I then expect to order a largish batch to use myself, using that design (modified if necessary).

When I have tried out the test pcb, I will post a scan showing the original and replacement pcbs, hopefully with wiper tracks visible.

I have also drawn up an amended design with modified connections to take, as Keith suggested, a standard pcb header (see attached). If there is any interest in this, we could get a batch made - though it does work out at around £2.50 per pcb even for a batch of 100. (Note that this is nothing to do with Exactoscale - I would just supply the gerber code which enables the pcb to be produced).
Tortoise rep pcb con.gif

Andrew Jukes

[File unzipped and shrunk - KN]
You do not have the required permissions to view the files attached to this post.

User avatar
jim s-w
Posts: 2186
Joined: Wed Jul 30, 2008 5:56 pm

Re: Tortoise Auxiliary Switches

Postby jim s-w » Mon May 25, 2009 10:36 am

Hi All

This may sound a bit dumb in light of comments on here but after all the shows with Calcutta Sidings we never had a problem with them. This despite running pairs of Heljan class 47's whos awful motors placed enough draw to trip out the contollers on a regular basis. (solved that by replacing the lot of them with Mashima's).

We have no reservations with using them as is on the new layouts and it might be dumb but with 80 ish in the fiddleyard and not far off that on the front of New Street at least we could report an accurate percentage failure rate if we have any.

Cheers

Jim
Jim Smith-Wright

http://www.p4newstreet.com

Over thinking often leads to under doing!

andrew jukes

Re: Tortoise Auxiliary Switches

Postby andrew jukes » Tue May 26, 2009 11:47 am

The situation with these switches is more complicated. I have now pulled apart another five Tortoises and it turns out that the most recent ones I've purchased have a revised pcb, REV C rather than REV B. From the outside, the most obvious differences are that the label on the REV C ones is yellower than on REV B and the ends of the pcb tracks are wider, such that there is only room for the number 1 (REV B has 1, 3, 5 and 7).

REV C is definitely better in the area that most bothered me, with all wipers actually on the tracks and less chance of shorting the two poles of one of the switches - so progress! Of the 7 Tortoises I've installed, 4 are REV B and these are the ones that gave the most worrying resistance readings.

Just to give some idea of how bad REV B is, it is clear that Circuitron had decided that the most problematic wiper had to be bent out of the way to avoid shorting. The attached photo shows this, exactly as it came out of the Tortoise.

I now have to decide whether I'm happy with REV C, in which case I have only a handful of REV Bs to sort out or even use them on crossovers, paired with a REV C so avoiding needing to use their switches.

Andrew
You do not have the required permissions to view the files attached to this post.

DavidM
Posts: 74
Joined: Sun Jul 20, 2008 7:32 pm

Re: Tortoise Auxiliary Switches

Postby DavidM » Tue May 26, 2009 9:04 pm

Very interesting Andrew.

Do you notice if there is any difference between the newer machines and the previous version as far as the locating pips for the wipers go - they look a little more secure in your photo the ones I have - remember, some of mine fell off!

David

andrew jukes

Re: Tortoise Auxiliary Switches

Postby andrew jukes » Wed May 27, 2009 7:57 am

David

I think the picture must be deceptive because if anything it's the other way round. With both this and the other REV B one I've had apart, it's possible to see under the wiper bases whereas with the only REV C one I've looked at this closely on they were properly seated. The pips looked much the same and I couldn't see anything that indicated the moulding had been altered at all. Incidentally, both REV Bs had the bent wiper - hence my view that it was deliberate.

You don't actually say that yours are REV C but if they are it doesn't help my confidence. I would like to be able to approach this just like Jim - just fit them, wire them up and get on with life. With REV B, I won't do that but with REV C.........

Andrew

User avatar
Rod Cameron
Posts: 850
Joined: Fri Jul 25, 2008 12:01 pm

Re: Tortoise Auxiliary Switches

Postby Rod Cameron » Wed May 27, 2009 10:33 am

Andrew

Are there any distinguishing features on the outside to enable you to tell which version you have (or might buy) without taking it apart?
Rod

andrew jukes

Re: Tortoise Auxiliary Switches

Postby andrew jukes » Wed May 27, 2009 10:54 am

Rod

Yes - see my post yesterday!

Andrew

User avatar
Rod Cameron
Posts: 850
Joined: Fri Jul 25, 2008 12:01 pm

Re: Tortoise Auxiliary Switches

Postby Rod Cameron » Wed May 27, 2009 11:40 am

Doh! Thanks Andrew!
Rod

User avatar
Rod Cameron
Posts: 850
Joined: Fri Jul 25, 2008 12:01 pm

Re: Tortoise Auxiliary Switches

Postby Rod Cameron » Wed May 27, 2009 12:02 pm

Just checked mine (all waiting to be installed) and there's only one Rev B 8-) so that can be used for one of the signals.

Thanks for the reports Andrew :D
Rod

DavidM
Posts: 74
Joined: Sun Jul 20, 2008 7:32 pm

Re: Tortoise Auxiliary Switches

Postby DavidM » Wed May 27, 2009 2:22 pm

Andrew,

As far as I can tell (without removing them from below the baseboard) mine are all Rev B - unfortunately I can't see the terminal strips because they face the baseboard - but I'm fairly sure the PCBs were labelled that way. The labels are bright green as well.

I'd be inclined to take my chances with Rev C but modify Rev B in the way suggested or just use them for non-switching applications. If you have any failures or even doubts about wiper mounting then I'd recommend the simple repair I mentioned earlier.

I couldn't agree more though - you need to be confident they aren't going to misbehave!

David

beast66606

Re: Tortoise Auxiliary Switches

Postby beast66606 » Thu May 28, 2009 10:43 pm

We use well over 150 of them and have only had a couple of mechnical failures, just go for it and sort problems out if they happen :?


Return to “Track and Turnouts”

Who is online

Users browsing this forum: No registered users and 0 guests