Easton, Isle of Portland

[nickwood]

1 hour ago, Stubby47 said:

MERG do simple single servo boards, or a 4-servo board.

 

The servos are cheap from the internet - MERG recommend a couple of alternatives.

 

MERG Stuff can, of course, be used with JMRI and other stuff - @peter220950  (plus many others) could give far more advice.

 

Also : https://www.merg.org.uk/merg_resources/servos.php

 

Mick, I know, that you know, someone who knows how, who could help you with that

 

[Mick Bonwick]

Don't get too excited, there are no photographs in this post.

 

There is a post in this photograph, though.

 

 

I'd be stringing you along if I said that it was a load of old rope.

 

 

[Stubby47]

Not to mention you've posted a web on the web.

[Mick Bonwick]

A recent mention of MERG reminded me of some stuff I bought a while ago. The plan then was to build a small layout for something-or-other, that would keep me motivated with Easton, at a time when there were a lot of distractions.

 

I delved into the boxes that usually live under Easton and found some point motor related items. At the time I even downloaded the software for my PC and built the 4-server board and successfully tested it all with the 4 servos connected. At some point the room was tidied up (again) and they disappeared for a few years.

 

I now have some mounting kits and microswitches that can be used to employ these servos as intended, so it is not beyond the bounds of possibility that they could be used to replace the Cobalt-SS failure.

 

[Mick Bonwick]

The intention to start the build of the control panel was well meant. The rate of progress is minimal.

 

I have a test rig that was created some time ago for checking the operation and electrics for solenoid point motors while building a large train set. The idea was to test each point installation while the boards were on their side for easy access to everything, before the control panel was connected to the under-board wiring.

 

It has been resurrected, with minor modifications, to test all the point motors' operation in the fiddle yard at Easton before wiring up the control panel. It also serves to confirm the logic employed when I came up with the wiring colouring scheme and connectivity.

 

Quite simple, really, in that the wires from the point motor (for LED incicators) and the decoder (solenoid operation by manual switching) are connected to the test rig in turn for each and every point.

 

 

The cables connect to the terminations seen earlier in this thread, and the two DB25 break-out boards will be replicated on the underside of the control panel proper.

[Mick Bonwick]

On 22/07/2020 at 20:36, Mick Bonwick said:

 

If I had any I'd have tried them. I don't so I didn't.

 

Which would you recommend and how would you drive them?

 

I told a lie, didn't I? I did have some but had forgotten about them. I wonder if I've got enough time to go through all my boxes and see what there is, tucked away and forgotten about . . . . . . . . . . . .

[ian]

On 30/07/2020 at 12:58, Mick Bonwick said:

 

 I wonder if I've got enough time to go through all my boxes and see what there is, tucked away and forgotten about . . . . . . . . . . . .

 

No.

 

Next question please.

[Mick Bonwick]

What to do now?

 

Testing of all 36 installed point motors in the fiddle yard has now been completed. There are 19 that work perfectly, in that they report the position of the blades accurately, and 17 that don't. Of the 17 that don't I have repositioned 3 of them so that the integral switch contacts work properly. The reason for the question is, how much more effort do I want to expend in correcting the remainder?

 

The options are:

 

1. Fix the rest.
2. Abandon the idea of a mimic panel and use JMRI.
3. Replace the rest.
4. Change the decoders that power the rest.
5. Replace all 36 with servos or something.
6. Take up hang gliding.

Wot d'yer fink?

[KNP]

Why don't you do what I do in these situations.

Park it.

Get on with something else.

Return to it a few days later refreshed and full of ideas!

[Mick Bonwick]

12 minutes ago, KNP said:

Why don't you do what I do in these situations.

Park it.

Get on with something else.

Return to it a few days later refreshed and full of ideas!

 

It is parked.

 

I am searching boxes to see what has been hiding for so long I'd forgotten about it.

 

It could be parked for some time!

[Tony Teague]

Sorry to hear of this Mick.

Whilst you search your boxes think "solenoids, CDU, push-buttons"!

(runs for cover....)

[Stubby47]

Wire in tube.

 

'nuff said.

[Tim Dubya]

little white mice

[Mick Bonwick]

4 hours ago, Tony Teague said:

Sorry to hear of this Mick.

Whilst you search your boxes think "solenoids, CDU, push-buttons"!

(runs for cover....)

 

That's what I'm trying to do, Tony. The point motors are solenoids (SEEP PM1), the decoders have CDUs built in, and the toggle switches (as opposed to push buttons) all work. The bit that's not working is the LED indication from the PM1s back to the control panel. The contacts built in to the PM1 PCB are not in quite the right position, althought the point blades are switched successfully. All the points work and always have done via JMRI. All the point settings are reported well enough through the JMRI control panel. It's the physical blade position reported by the PM1 itself (which is what I'd rather rely on) that's not working.

[Mick Bonwick]

1 hour ago, Tim Dubya said:

little white mice

 

My brain isn't clever enough to figure that one out, Mr W. You'll have to explain it in words of one syllable.

[Tim Dubya]

3 minutes ago, Mick Bonwick said:

 

My brain isn't clever enough to figure that one out, Mr W. You'll have to explain it in words of one syllable.

 

https://www.bbc.co.uk/programmes/profiles/1MmQjXrkwp98jNdc17j3tr3/mice

 

[Mick Bonwick]

6 minutes ago, Tim Dubya said:

 

https://www.bbc.co.uk/programmes/profiles/1MmQjXrkwp98jNdc17j3tr3/mice

 

 

Ah, yes. That one passed me by while I was distracted. Often meant to see what the fuss was all about, but never did.

[figworthy]

6 hours ago, Mick Bonwick said:

What to do now?

 

Testing of all 36 installed point motors in the fiddle yard has now been completed. There are 19 that work perfectly, in that they report the position of the blades accurately, and 17 that don't. Of the 17 that don't I have repositioned 3 of them so that the integral switch contacts work properly. The reason for the question is, how much more effort do I want to expend in correcting the remainder?

 

The options are:

 

1. Fix the rest.
2. Abandon the idea of a mimic panel and use JMRI.
3. Replace the rest.
4. Change the decoders that power the rest.
5. Replace all 36 with servos or something.
6. Take up hang gliding.

Wot d'yer fink?

 

Not the last one, we'll miss your erudite postings.

 

Adrian

[Tony Teague]

12 hours ago, Mick Bonwick said:

 

That's what I'm trying to do, Tony. The point motors are solenoids (SEEP PM1), the decoders have CDUs built in, and the toggle switches (as opposed to push buttons) all work. The bit that's not working is the LED indication from the PM1s back to the control panel. The contacts built in to the PM1 PCB are not in quite the right position, althought the point blades are switched successfully. All the points work and always have done via JMRI. All the point settings are reported well enough through the JMRI control panel. It's the physical blade position reported by the PM1 itself (which is what I'd rather rely on) that's not working.

 

Apologies, I thought you were using tortoise type motors.

As you know, I have LED's on my panel that show how points, crossovers, etc, are set, however, they are switched by relays wired in parallel to the point motors, so they are actually reporting on what button I last pressed, rather than the position of the switchblades.

There is a risk in this but despite having "several" points on the layout, the number of 'mis-fires' and therefore erroneous LED's showing is miniscule - after all, solenoids are pretty reliable once set up properly.

What you are attempting is far more sophisticated, but I took the view that if there was an error, no-one was going to die! (well probably not).

Good luck in solving the problem.

Tony

[Mick Bonwick]

The Tortoise/Cobalt types are installed on the scenic side. I suppose at the moment it should be called 'the other side' as there's no scenery yet. The switch contacts on those will be used to report 'blade' positions when the time comes to connect a mimic panel to that side of the layout.

 

Like you, I have had very few failures, even with my cheap solenoids, but want to reduce the risk as far as possible. There's quite often a difference between what we want and what we can reasonably achieve. Compromise is where we end up. I'm getting there fast!

 

I'm off to Easton to solve a problem . . . . . . . . . . .

[Tony Cane]

As the problem is in teh fiddle yard if you want proof of tie bar position may i suggest  using a microswitch as in the attached picture.

The PM1 switches are not a good design and you may have problems latter with ones that are working.

If you are interested in this solution I can dig  out the source of the switches I used, obtained via Ebay.

[Mick Bonwick]

33 minutes ago, Tony Cane said:

The PM1 switches are not a good design and you may have problems latter with ones that are working.

 

The failings of the PM1 are known, and when I started down this path I bought spares for easy(ish) replacement in the inevitable event. Fortunately I haven't yet had a failure. It is the poor positioning of the motor at the time of installation that has brought me here, so I can't really blame the design.

 

If I had used Peco PL10s or PL10Es I could have made use of their adjustment facility and also added a microswitch to each one. That would have increased the cost for the 38 (at the time) that I needed to way beyond my budget.

[Tony Teague]

40 minutes ago, Tony Cane said:

As the problem is in teh fiddle yard if you want proof of tie bar position may i suggest  using a microswitch as in the attached picture.

The PM1 switches are not a good design and you may have problems latter with ones that are working.

If you are interested in this solution I can dig  out the source of the switches I used, obtained via Ebay.

 

 

I have also used this solution in a couple of places, as well as using it for polarity switching on 3-way points within the fiddle yard, however, I have used a different type of microswitch with a long metal lever touching the point switching bar and this has proved pretty reliable.

Tony

[Mick Bonwick]

Ah, yes. Microswitches. I have some of those! Right in the middle of this photograph:

 

 

There is usually a gap between me doing something and me reporting that doing of something  on here.

 

Three or so years ago I joined the Leamington and Warwick Model Railway Society and became involved in their O Gauge exhibition layout, Kimble. Most of the work I did to begin with was helping with the rewiring of the layout, coupled with the helping with a change from ZTC controls to MERG CBus and DCC controls. One of the solutions to accurate point blade reporting on a control panel was . . . . . . . . . . . . . . . . . . . . . . . . . . . . . the use of microswitches.

 

I also used microswitches a little while ago when installing a lifting flap. If you want to know about that then you'll have to find it yourself.

 

All of which is a long-winded way of saying, "I've solved the problem by using microswitches". The wiring from the point motor connections under the baseboard have been brought up through the baseboard and connected to the microswitch, positioned right beside the tiebar.

 

 

 

 

Now the correct indication appears on my control panel test rig when I operate the switch. Magic!

 

 

 

 

Edited August 2, 2020 by Mick Bonwick

[Mick Bonwick]

19 hours ago, KNP said:

 

Park it.

 

 

Not only did I park it, I remembered where it was when I came back!