Cue n Review

Dear Matt,

Might be worth a review?

https://model-railroad-hobbyist.com/node/43364?page=2#comment-485733
(check the whole thread too, if you're interested in an Arduino DIY option...)

this might also be relevant?
https://model-railroad-hobbyist.com/node/40516

Happy Modelling,
Aim to Improve,
Prof Klyzlr

Team Digital

Team Digital has some products that may be suitable.  http://www.teamdigital1.com/

@ rail.burd

According to their website they are not selling boards anymore

Quad Servo DCC Decoder

If you are willing to go down the DIY route, you might check out my QSDD which gives you full control over the setup and operation of four servos, details of which were originally published in MRH February and March 2020, with later amendments posted in the forum ( https://forum.mrhmag.com/post/quad-servo-dcc-decoder-%E2%80%93-improved-version-12217939 and  https://forum.mrhmag.com/post/quad-servo-dcc-decoder-%E2%80%93-arduino-software-update-12218580). You can also find complete information on my website at https://www.a-train-systems.co.uk/download.htm#Projects.

Remember MRH advertiser MRCS …

Price comparison

   The controller provides all of the electronics required to control one servo, such as the popular SG90. This controller sells for $13. To control 12 turnouts equals $156. From website of Model Railroad Control Systems.

   Control up to 12 servos around your layout. Individually configurable for turnouts, semaphore signals, gates, barriers, doors etc. This item sells for $90.80. From MegaPoints Controllers website.

   The BlockSignalling Smart ServoSwitch SRV4 is designed to operate up to four servos between their Start and Stop positions, whenever a control input is switched. This unit sells for $34.44. For 12 turnouts price is $103.32. From BlockSignaling website.

    I do not work for, have stock in or in any way connected to the above list vendors other than being a customer of  BlockSignaling.

    I used 12 turnouts as a control number as the Megapoint controller does 12 turnouts. Please note the MegaPoint  and BlockSignaling item prices are adjusted to US dollars as of todays exchange rate with the British pound.

    I have close to 75 turnouts to be operated on my layout. So, pricing is a serious concern. Servo motors will be a big savings compared to other methods available. That being said, for me the Walthers servo turnout controllers are out of my price range after you buy all the controlling extras.

    Hope this might be "food for thought" and open up a door to a couple of other vendors for anyone interested. More info, of course, is on the inter-web for the above listed sellers.

@John Holt

Can you solder?  If price is such a concern, between one and five Arduinos with a total of five PCA9685 servo expanders handling 15 servos each will do what you want.  The trick is, who's writing the software?  Oh, wait, there's a plethora of solutions out there, you just need to be able to evaluate and select one.  But beware, this is truly a roll-your-own environment.  You have to roll up your sleeves.

Where are you?  Somewhere in the US or Canada?  I'm in Canada, and I can buy nanos for $7 ea/5, PCA9685 boards for $8ea/3.  So depending on how you spread out the Nanos, you could have the control side covered for between 55 and 85 CDN.  That's not bad, though you can do much better direct from AliExpress, if you go there.  The servos run around $1.50 ea, depending on source and quality.  So for 75, you might be able to land the whole schmear for under $200, not counting servo cables, 5V power, and a few other little extras.

Doing it this way opens up the opportunity to network the Arduinos via RS485 (adds about $5 per node, plus $10 for the PC end) and use CMRI/JMRI(all free) on top for a really sharp control approach - or you can use the I/O left over on the Nanos to monitor buttons, drive LEDs, etc. at each location; you could even implement a mixup of the two, with a bit of 'elbow grease'.

If you want to discuss a collaborative, reach out to Joe Fugate (see above, since they've blocked the old contact tool, wisely), have them reach out to me with your email address or give you mine, and we can talk.  I've got most of the code written for a 16-servo implementation anyway.

Blair Smith

Thanks for the resources! And a follow up question.

Thanks Prof, for the thread pointers, Terry and Blair for the Arduino ideas, and Joe, Ted and John for the reminders of vendors who are in this space. Once I saw MRCS and Megapoints written, I of course immediately remembered them. The others are new to me. 

Regarding servos, I bought a batch of twelve earlier this year to get a start of bench testing the Tam Valley board, and found two wouldn't work. On manually manipulating them, both seem to have gear issues (probably flash) that would cause the motors to stall. 

Is this to be expected, or is their really a hierarchy of quality with these things?  Are there sellers you've consistently gotten a high pass rate from?

Thanks!

Servos

Like a lot of things, you get what you pay for. If you only pay a few dollars for a servo, you will get a mass-produced, low quality item. They are not made to close tolerances, either mechanically or electrically, so you will almost always get some bad ones in any batch - they either will not work at all (as you found) or will fail within a short time.

I tend to regard them as expendable - and simply replace them with another one. One consequence of this is that you need to pay attention to how they are mounted next to your turnout - such that they can be replaced without a lot of hassle.

If you want high-quality, reliable servos then you will have to pay a lot more for them - servos such as those from Futaba which are made for model aircraft use - and even then you need to be aware that there are a lot of counterfeit servos on the market. As always - caveat emptor . . .

What Terry said.

Yes, they're cheap.  Doesn't work, goes in the bin, unless I decide I want to disassemble it for kicks.  I have two, out of 20, that have gone there.  Before I build any into the layout, I'll definitely set up a test bench to work the "infant mortality" kinks out.  But even if I ended up tossing 50%, that means the rest are $3 or so apiece. Still beats a Tortoise.  Now, be aware, without more work, you don't have any feedback on the position of the turnout, but many don't care.  If you do, you'll want to arrange a microswitch, magnetic sensor and magnet, optical interrupter, or other device to give you feedback, but that should be easily do-able for a buck or two.

With the Arduino solution, it's much more about "I did it myself, because I could", not just "got it done, on to the next challenge", so YMMV.

Blair

Servo Control

A link from the Tam Valley website brought me here to this page.

https://www.n3ix.com/quadln_s/about-us/

Could be similar to the Octo.

Something to ponder.

Digitrax also has a new servo controller board.   

@ Blair Smith et al

I thank you for your thoughts........But

My only intention was to pass on some info I have collected for my own layout. And yes, I solder very well. My thoughts were to present something similar to the Tam Valley item mentioned in the original thread/post, no programming required.

My background is in structural engineering and NOT electrical. At age 73, I really don't want to go through a learning curve on Arduino programming or depend on someone else for my programming needs that may not be available when I encounter a problem.  That seems to be a hobby within a hobby. While there are some different options available for turnout controls, I believe a person needs to go with what they feel fits their needs and capabilities.

For me (maybe not for others) the BlockSignaling items are very nice and user friendly. Very easy to set servo limits and run lights to control panel/signals LEDs and DCC friendly. Great documentation and communications with owner. While a bit more expensive than the MegaPoint controllers, I believe the hook-up will be easier. BlockSignaling is not very well known in the US, but they offer a great product line. I have a CDU and a couple of controllers I purchased from them for twin coil switch machines..... quick and easy hook=up.

As for inexpensive servo motors, I buy mine from HobbyKing, the HXT 900 model. I Have purchased some Tower Pro servos BUT, only buy TowerPro servos that say Tower Pro cast into the plastic top of the servo as there are a lot of knock-offs of that brand of servo. Some YouTube research paid off here. If you see a dozen servo motors for sale for $5.99, trust me, this is not a deal.   I have a servo tester and plan to test the servos I have just recently purchased sometime this week. Hopefully they will all be OK.

Thanks....John

Fair enough, John,

as I said, it's very much a YMMV game.  With an I&C background, I gallop into any electrical/electronic rabbit hole, and generally fare okay, but I realize that part of this vast hobby isn't for everyone.  I think the latest posting before yours holds out some hope for the Tam Valley servo offerings, if the primary concern, Robin, is willing to grow the product line.  There will always be edge concerns, like myself, who prefer DIY even when there are proven, well tested products available, but for the rest of the hobby, I hope Robin goes gangbusters; Tam Valley showed a lot of promise, but they were waylaid by the jester.

Blair

N3IX Item

QuadLN_S Board

N3IX

Robin recently helped us test and identify a fault that had occurred in a QuadLN_S working live remotely from his home in California.  I have the utmost respect for Robin and his great customer service.  And this was not the first time Robin answered support questions about TVD products.

A VERY satisfied customer.

Servo Failure Expectations and Control Goals

Thanks Terry, John and Blair for setting expectations on servo failure. In my case, they weren't completely broken, they just didn't work all the time .  I'll assume my 2 of 12 failure rate is typical and be happy if it's lower than that. 

John, I've seen the term "Tower" in about every Ebay description, but wasn't sure if that was a brand or some common categorization term. It's helpful to know the context around what I'm reading. As far as Futaba goes, I think I'll use the swap-the-cheapies-as-they-break approach, and use better ones for less accessible locations.

Blair, I do want to do this on my own - I'm reasonably comfortable in an IDE and with general coding practices, but have decided I don't want to take the time to build this particular thing at this particular time. One of our local Division members has developed his own servo control solution, so I have that path I can look dow as well (which has the advantage of having local customer service!)

Dave, I had some questions for Robin after scanning the web page - parts of his descriptions made it sound like the Quad LN was built around loconet, and all configuration is via JMRI, which I want clarification on. I just need to formulate my questions. My goal is to have the ability to wire the boards into a non-proprietary network (though not DCC) for future remote control, but until then ensure they can work completely independently via via local buttons/toggles, etc. The Octo III met these druthers.

QuadLN_S

Matt,

On a friend's layout I did use JMRI DecoderPro with LocoNet to setup the QuadLN_S decoders, but once they were set up the push buttons will control the servos with no needed connection to LocoNet or a computer.  I used the Version 2 decoders and Robin has included so many possible configurations that using JMRI makes it a snap. JMRI also helps with the setup because we use LocoNet to allow remote servo control and servo position feedback to the computer.

You can actually set the travel on the servos without LocoNet, just using the Remote Align Board with the supplied cable.  I don't recall what the default push switch settings were.  You don't have to use LocoNet, it just increases access to the features of the boards.

On that particular layout LocoNet is set up as a stand-alone network and there are no Digitrax components in the system which includes an NCE PowerPro DCC controller.  An RRCirKits LocoBufferUSB and SSB-Gateway provide the LocoNet network.  A RaspberryPi running JMRI was used for setup, but because this particular layout is intended to be run with automation, the RocRail software runs during normal operations, not JMRI.  Both software reside on the same RPi without conflict, though you only run one at a time.

In one sense LocoNet is proprietary, but there are several other vendors that make components that utilize LocoNet so that it isn't restricted to Digitrax, and in that sense I don't really consider it strictly proprietary.  As an example, on my personal HO layout I use the Digikeijs DR5000 with LocoNet.  

YMMV

Non-Proprietary LocoNet

Thanks Dave, that nicely clarifies the meaning of the website descriptions and provides real-world context. If I network the layout, I was looking at LCC - which might work with the Locobuffer to translate. Something worth investigating - I don't want two networks on the layout!

LCC vs LocoNet - My Thoughts

Matt,

The following are my thoughts and my personal opinions.  By way of background I started with a Digitrax Zephyr back in 2007, my first entry into model railroads, with the promptings of a young grandchild.  I ended up with Digitrax because that is what the LHS showed me, and I'm glad I did, mostly because of LocoNet.  BTW I do own an NCE Power Cab, a Sprog 3, a PiSprog, a couple of DCC++ command stations, and now a Digikeijs DR5000.

The first layout was designed with 27 Tortoises, a couple of SE8c's, some DS64's and BDL168's.  Track was down and LocoNet components were being added.  Six years ago a new position at work allowed me to move back to St George, where we had raised our family.  The old layout didn't fit the new train room so I started over.  At just that time JMRI had added a connection to DCC++. 

The new layout has 9 sections on two levels including a 2-track helix and I thought it would be ideal to have an Arduino Mega on each section controlling servos, signals and detection.  But the people on the DCC++ thread didn't follow thru with connecting  multiple Mega's, and that was above my pay grade, so I went back to the Zephyr and Tortoises, and Digitrax components from the previous layout.

The plan is to allow automated running for some or all trains, and RailSync is something I desired to use.  I closely followed the development of LCC and then TCS announced their command station and throttles with LCC capability and RailCom, but no LocoNet.  I waited over a year for the TCS command station to be released, thinking that I could use JMRI to connect LocoNet to the TCS CS (two networks).  But it never happened.

So, I went with the Digikeijs DR5000, replaced the BDL168's with DR5088RC detectors so that RailSync would work, and moved forward (at a snails pace) with the layout.  I really considered LCC for many reasons, but in my case, I could do all I need with LocoNet and JMRI on a Raspberry Pi.

With LCC you can setup components (signals, detectors, turnouts) to communicate with each other, and function without a computer.  That is the only major advantage that I have seen going with LCC in my particular situation.  My layout requires a computer to operate and I already have several LocoNet components.  That's the price I pay to not use LCC.

So, I strongly recommend that you start where you are at today, and decide whether you want to use LCC, LocoNet, Arduinos, or any other system, before you buy your next component.  Then move forward in that direction.  If you plan on LCC, buy servo controllers that will (first) work with LCC in the future, and (second) provide the local control that you desire now.

Again, this is my opinion, and my thoughts. YMMV.

Inexpensive servos

I agree Matt, 

The reason I went the servo motor route was the price and for me, ease of set-up. The Futaba servos are NOT cheap but the 9g (inexpensive servo) models are fine.

As far as Futaba goes, I think I'll use the swap-the-cheapies-as-they-break approach, and use better ones for less accessible locations.