If you are not able to build your own board, here's a solution

I've just read Terry Chamberlain’s 'frustrations with commercial decoders led to building his own highly configurable quad servo DCC decoder' article.

He writes: "I wanted to control the throw of each servo easily and accurately by hand, prior to their installation on the layout, and then fine- tune the servo movements after they were in place."

While his article is excellent, it may well be that fellow modellers don't have the skills to build his board.

In which case, I suggest that they consider MegaPoints' board, which does exactly this.

I don't have any connection with MegaPoints, other than being a happy customer.

The owner, Dave Fenton, is very helpful too and his instructional videos are very helpful if you're not an electronics whizz!

This one shows how you can easily set a servo to exactly the throw you want (in this case to control a signal)

Rectifier speed?

Not wanting to be a pedant, but it seems that the rectifier bridge in this circuit is designed for use at power-line frequency of 50 or 60 Hertz. I looked at the spec sheet and they don't list a reverse recovery time. (When the voltage across a diode reverses, the diode actually conducts backward very briefly. Some diodes, always for the Schottky type and sometimes for others, have a "fast recovery" characteristic which prevents this, but unless it's promised, I wouldn't assume it.) Not knowing the recovery time for this part, I looked it up for the ubiquitous 1N4002, and there it's 2usec. Given that reversals of the DCC line can come at a rate of 60usec or so, that could be significant. I'm not sure what the result would be, but I think there would be extra loading of the DCC bus and a tendency for the power supply capacitor--I think that's C2--not to hold as much charge as it should. Obviously the item has been built and does work, but I wonder if this is a problem.

Hah!

I can't imagine that being interpreted as pedantry! I really, really wish I understood anything you just said, John!

Not a Problem

Thank you for the comments on my article - glad to see that somebody has read it !

This isn't really the place to delve into electronics theory, but the question of reverse recovery times does not really arise in the case of a bridge rectifier - when the DCC input changes polarity, the opposite pair of diodes begins to conduct in any case, and the reservoir capacitor absorbs any momentary transients, so these are not seen by the DCC system. The whole rectifier system is actually much more efficient when operating at DCC frequencies (around 6 kHz) rather than at normal power frequencies, and any ripple in the DC output supplied to the electronics and servos is much reduced.

The quiescent current taken by the Quad Servo DCC Decoder is around 55mA (0.055 amps) rising briefly to around 200mA when a servo is moving. This is generally much less than a normal HO locomotive, so will not impose an undue load on your DCC system.

Quad servo DCC decoder

I guess I need to stop skimming over an article.    I was a little harsh in my original comments and I have retracted those comments as per logger01.  Yes sir, you are correct, I do need to slow down.  Thanks.  However, for a stationary decoder, having a dual power connection would be nice.   My local club is in the process of rewiring the layout and it is a bit of a mess.  There are issues regarding power and DCC signal distribution all over the place as it was a DC layout moving through several generations of DCC.   But I do like the idea of using the small inexpensive servos for switch machines.

Better reading retention would help

rgrabau Please try reading the article again while working on you retention skills.

Following the link in the article the software can be downloaded at http://www.a-train-systems.co.uk/arduinodownload">http://www.a-train-systems.co.uk/arduinodownload. For those of us who know how to read schematics, the schematics for this very simple board may be of help. But for those who do not read schematic- So What. The instructions adequately show and explain how to build the board and to load and run the software, so for the novice definitely not "mostly worthless".

If you ask politely I am fairly sure that Terry would provide the schematics.

Diode bridge

Regarding the rectifier bridge, I just thought it was worth mentioning. If the designer is happy with it, I won't argue.

Schematics

Thank you Ken K for your helpful comments, and for pointing out that the full software listing is currently available from my website by following the link in the article.

For those who can contain their impatience, the schematics for the QSDD are included in Part 2 of the article which, if all goes to plan, will be published in next month's magazine.

That’s the plan!

Yep, that’s the plan!

Meanwhile, You’ve got 4 weeks to determine how many decoders / keypads you need and get the boards ordered, and to order the parts. (Wink and grin)

Interesting approach

Mobile Accessory Decoder

I hadn't considered using the QSDD as a mobile decoder but you should have enough space within a larger scale model to accommodate one or two decoders, and you can certainly set the servos throws to cover the ranges required - generally wider than for simple turnouts.

The only problem which may arise is a brief interruption of power to the QSDD from the track.where it gets dirty. Such a break in DCC power will reset the QSDD and, as currently implemented, initialisation of the software takes around a full second before the decoder is again operational - which is likely to disrupt any operations in progress at the time.

It should be possible to rewrite the Arduino code to achieve a more rapid recovery from power breaks, but I think you should also consider fitting a "keep-alive" unit to your snow plow as a belts-and-braces insurance.

@ Terry

Terry,

To add complications to the mix, I'm running dead rail.  But I have been beta testing a DCC controller from Blue Ridge Engineering to run my Protothrottle via dead rail. Since that system 'talks' via a Ardurio code as well, it might be possible to combine the two. I'm a long ways off from even experimenting with this part, as I'm still in the design phase of the photo etch for the build. Martin has designed his system to run 3 servo motors, but as I said, I need up to 6.

Would it be possible to increase the code and board to have 6 servos? Although having two separate decoders could represent the two different sides of the plow (fireman vs engineers side). 

http://blueridgeengineering.net/index.php/wiki/protothrottle-receiver/

Anyways, with the pace that I'm going with this model it will be another 3 or 4 years before I even get to the software side of things. 

Craig

Complications, certainly

Craig,

Operating dead rail is actually good news, since it removes the potential problem of momentary power interruptions. However, your proposed use of a Protothrottle might be a bit more problematic since, as far as I can see, the throttle has no provision to issue DCC accessory commands (a different format from loco commands). As the Protothrottle is an open system it's likely that you could rewrite its software to implement accessory control but that would certainly be adding further complications to your project.

Expanding the Quad Servo Decoder to handle 6 servos would, in theory, be possible since the Arduino Nano does have enough I/O pins (just) - although this would require new board designs to accommodate the extra two LEDs on the keypad and additional servo connections, plus a rewrite of the software. Perhaps I should leave that as an exercise for the interested reader .

Good luck with your snow plow - I hope to see the results when you get it done.

@Terry

I'm not opposed to using something other than the PT for control of the servos. But like I said, it's a long ways down the road at this point. So maybe by that time someone else will have come up with an easier solution. This hobby does one thing pretty good; sharing information.

Concise and Clear

It’s hard to write a technical article that is clear in its meaning without being too wordy. Terry has done a wonderful job of in this regard. Thanks for the writing precision and fine selection (and composition) of the photos accompanying the text. 
 

@Matt

Thank you for your kind words - it's great to be appreciated - and I'm sorry about the delay in acknowledging your comments, since I have not been checking this post as often as I should.

Quad servo - completed them

I built three of these and I love them. I find them simple to operate and the work well with with hand made turnouts and commercial. I even have an idea for Kato turnouts but I can't modify the code properly. Kato turnouts require a momentary power signal to operate, much like our snap switches. So I thought how do we do this, why not use the servo to activate a micro switch. So the servo would have to go left and right (activating the micro switch) depending on which route you want and then return to a neutral position until the next command. Any body have a an idea for this as I just have one Kato double cross over turnout.

Thanks

Dan

Anything's possible . . .

Dan,

Glad to hear that you like the QSDD. While it was not really designed for the application you have in mind, with software it's possible to do just about anything. I assume that what you have in mind is to mount two microswitches, one each side of the servo actuating arm, then drive the servo left or right to actuate one of the microswitches for a short time (less than a second), applying a short pulse to the Kato switch machine, and then for the servo to return to a central, neutral position and wait for the next command. The electrical circuit would look something like this - 

The QSDD code would need to be modified to add a flag in one of the configuration variables (CVs) to indicate that one servo was to operate a Kato switch machine, and then the software which receives and executes DCC commands would need to be changed to recognise this flag and make the designated servo behave in the required manner each time a command was received for that servo's address.

It is not practical to use the QSDD keyboard to set the "Kato flag" since all the sensible key combinations are already allocated, so the flag would have to be set via operations mode ('on-the-main') programming using the appropriate DCC commands - this aspect was not fully covered in the published article (Part 2) and would need a bit more explanation.

While I would like to help further with this little project, unfortunately my time is fully committed to two other projects until at least Christmas. If you want to get things done before this then I might be able to offer some limited advice - either to yourself or to anyone else who would like to assist.

Quad servo - completed them

Thanks for the offer and I will continue to experiment with what I want to do in the meantime. Your electrical drawing solves a couple of problems as I was going to have to use relays. Your sample solves this problem

Thanks

Attention to details

Terry, as a software developer, I sincerely appreciated your outstanding attention to making your decoder simple to operate and give sensible feedback. I dabble in electronics and I'm impressed with the quality treatment you gave to all aspects of this project. I would like to say, "thank you," for sharing a useful device and a well written article to go with it. It inspires me to do better in my own efforts.

Thanks - and updates

Thank you Ed for the feedback - it's always great to feel appreciated.

Since you are commenting on the thread attached to the original MRH article, it's possible that you are not aware of the subsequent updates to the QSDD - a hardware revision published in February this year ( https://forum.mrhmag.com/post/quad-servo-dcc-decoder-%E2%80%93-improved-version-12217939), and a software update (Version 5.3) published in May ( https://forum.mrhmag.com/post/quad-servo-dcc-decoder-%E2%80%93-arduino-software-update-12218580) to correct an original oversight.

The latest information on the QSDD can also be found on my website at  https://www.a-train-systems.co.uk/download.htm#Projects.