DCS-RC Perpetual Barking Watchdog Generator (DCS-RC WD Gerber Files Added to First Post)

Well, that would get us to at least ten, so we have reached critical mass.  The option will be a kit for $6, or an assembled/tested board for $11.  I'll post the "final" documentation here for download.

BTW, the kits will consist of all the parts and the header to solder to the DCS-RC connector as well.  You shouldn't need anything extra to put them together.

Please email me actual commitments to my profile email and we'll start the ball rolling.  Boards will take around 2 weeks, parts should all be in by then.

John, so I have this straight.  This is a gizmo to fit inside a remote commander to cause it to send a repeating watchdog to a layout area powered by a single TIU channel.  Correct?

The prices sound good to me. Since you are going with the assembled ones, I'll take 4 of those. I trust your assembly more than I do mine. 

Will send you an email.

RTR, great question. I plan to use a 10A relay thru an AIU to turn on my roundhouse/turntable or yard, siding or whatever. Will one Remote Commander cover every area/zone?

RJR posted:

John, so I have this straight.  This is a gizmo to fit inside a remote commander to cause it to send a repeating watchdog to a layout area powered by a single TIU channel.  Correct?

Correct, that's what it does.  At least Stan says it does that, he's the architect of this one.

milwrd posted:

RTR, great question. I plan to use a 10A relay thru an AIU to turn on my roundhouse/turntable or yard, siding or whatever. Will one Remote Commander cover every area/zone?

It will basically cover what a TIU channel would I suspect.  If you are turning on track segment, it certainly should do the trick.  OTOH, if you're turning on the world, it may not be enough.

Having 6 TIU channels and 0 remote commanders, I'll pass on this. 

But, GRJ, I'm surprised at what you say is the finacial point for having boards made up.  You may recall you and Stan2004 gave me much help in perfecting an operating car device, for which I laboriously put together a dozen units.

You're surprised at how much or how little?

How little.

As long as your board specifications fit their totally automated system, they're fairly cheap.  However, anything special and the price goes way up for small quantities.  I think I get around twenty square inches of board or so for the $36 shipped.  Usually there's a few extras in the shipment, but sometimes not.  The boards are decent quality, double-sided, silk-screen and solder-mask on both sides.  The standard is .062' FR4 for the board material.

John-

who are you using to fab and assemble these? Those prices are extremely good for a small run simple board. 

I use DipTrace for schematic capture and board layout, they have an internal link to Bay Area Circuits that directly sends the data for the PCB order, very convenient.  Since their prices are as good as any I found, that's who I typically use. 

OK, parts and boards are on order, ETA around 2 weeks.  I'll send an email out to the folks that committed to boards via email.  Obviously, first come, first served, there will only be about 20 of these produced in either kit or finished form, depending on the exact quantity of boards provided, that is somewhat variable.  I ordered 17 boards, and I usually get a few extras for most orders.

This is the final board layout, and it's what was ordered.

John, I'm still in for two complete parts kits, since it looks like that's how you are leaning. Just let me know when to PayPal your account!

Rod

Can use 1 or 2 John.

John,

E-mail sent for 2 boards, preferably assembled, otherwise as kits.

Thanks!

As I said before, the options are a kit for $6, or an assembled/tested board for $11.  Shipping for one or two will be $3.50, add $1/ea for additional boards or kits.  Send email to my profile address with shipping details and how many kits or assembled boards you want.

Well, that was quick!  I have commitments for 20 sets, so that's all for this round.  If I get enough interest in more, I'll do another board order.  I guess I should have waited on the board order for a day or two!

GRJ, I am in no hurry for mine, anytime is fine with me. I know York is coming up and you are probably already busy with other things. 

OK John this thread got away form me and I missed the ability to buy some of these. If you do another run I will definitely take 4 of the boards

I'm saving all the latecomers in a file and I'll see if there's enough to do another board run.  I have learned a lesson, I should have waited a couple of days to order the boards, I could have increased the order.

Well, if you do make some more hit me up, I'll send you some money

Well, I got a few that came in after the gates closed, so there may be another run, we'll see what develops after these are done.

So for guys that missed the bus and/or already have components, using your neat discovery in the other thread:

can they use the exact zip Gerber files you supplied above and order 3 bare-boards for $4.05 shipped?

I saw that, and considered getting some. Maybe Ill jsut wait and see what John Does, I dont need em right now anyways but will need em once I get to the wiring of the switches in the yards and turntable

It seems I can accommodate more folks buying the boards as I can get them in small quantities in the same ballpark price as the first batch.  I figured to get these three boards in and see what they look like before pressing forward.

Stan, you are correct, I'll post the Gerber and drill files for anyone to use.

See attached file, these are all the files you need to order a set of boards from the OSH Park PCB Fabrication website, you should get exactly what I got above.  No need to wait for me to put them together unless you're looking for the parts kits and/or assembled boards.  It's very easy to order the boards, and as near as I can see, the check of the board files they received look fine.

Update: 10-19-2016, replaced Gerber files with corrected version.

I live in the world of long list of projects short list of time to do them. I will pay a bit more for stuff to save me time having to build it or figure it out

Since GRJ is the only guy kitting and/or assembling the Perpetual Barking Watchdog (PBW), perhaps this is directed only to him.   In fiddling with this more I observed a difference in watchdog timing between DC and AC operation - kind of interesting how it synchronizes to the 60 Hz line cycle during AC operation.  Skipping the details (requires an oscilloscope), I am proposing a performance range to insure all units behave similarly to simplify downstream discussions:

The PBW should generate the watchdog signal every 0.9 to 1.0 seconds. 

The (im)practical impact is GRJ will have to select the R1 resistor value (nominally 1M Ohm).  I can see him rolling his eyes.   There will be part-to-part variance in the 74HC14 timing, C1 capacitance (typically 10-20%), and to a lesser extent R1 itself (1-5%).  So to bring this into a ~10% range of 0.9 to 1.0 sec, it will be prudent to have a stash of resistors on hand of, say, 820K, 910K, 1M, 1.1M, and 1.2M (in other words, roughly -20%, -10%, 0, +10%, +20%).

So the build/test procedure would be (for anyone else trying this at home):

- assemble/solder circuit except for R1

- install the circuit into a DCSRC

- carefully insert 1.0M into the R1 board holes to make electrical contact

- observe timing and adjust; for example measure the time of 10 green LED blinks.  If between 9 and 10 sec then it's in range and solder R1 in place.  If less than 9 sec then it is too fast so increase R1 to slow it down.  If more than 10 sec then it is too slow so decrease R1 to speed it up.

Yes, in retrospect R1 should be a 20 cent trimmer pot but that ship has sailed.  I suspect in GRJ's case all components were bought together and the R1 value will be the same for all boards in the batch.  GRJ, I await a private message from you with what you really think of this.

Private?  I'm going to do it in public!

When I get the corrected boards in, I'll take a closer look.  Since virtually all of them are for assembly, I'll just "tune them.

Dumb question.  Why does it have to be in that specific range?  What would happen if the WD reset happened every 1.5 seconds?  Why is that a problem?

I did wonder if maybe this would have been a candidate for a 555 timer, but that ship has sailed as well.

I was wondering the same thing John. As I understand it the DCSRC watchdog signal lasts several seconds after startup, so whether the PBW triggers once a second, or up to once every 5 seconds or so is immaterial. Hopefully Stan will clarify his intentions.

Happily I have a DCSRC connected now at the rotary switches for my main yard and t'table, so PS-2/PS-3 engines now power up dark and silent until I start them up. Will modify it to continuous PBW operation once your kits are delivered. BTW, did you manage to get the kits done up and shipped yet? Not that I am in any big panic.

Rod

All good questions.

By observation (with a scope) the DCSRC only generate a single watchdog burst about 3/4 second after it starts up (or following a reset).  This burst is less than 1/10th sec with timing that varies a bit whether DC or AC track voltage as previously mentioned.  After this single burst the DCSRC watchdog returns to his doghouse and goes to sleep!

It is the TIU watchdog which barks for 45 sec after initial power up; this period has apparently varied with different TIU revisions.  But that is another investigation which is sort of irrelevant for the PBW.

So my latest thinking is reset the DCSRC as frequently as possible but not too frequently that it never lets that one-time bark come out.

I figure a 10% (or so) window is a reasonable objective (vs. a 1% window or whatever) easily obtained with a modest range of "tuning" values.  Plus, since this is sort of a 3-blind-mice situation (no published requirements on watchdog timing)...and since paying customers are playing the role of beta testers , a tighter spec is better for controlling the rollout.  That's just my opinion of course.

Ahh, I was going to offer a gold star to anyone who could second-guess why I did not use a 10-cent 555 timer IC which has better timing tolerance.  The 555 IC chip has relatively high operating current as it is a bipolar device (vs. the CMOS 74HC14).  Since I am stealing power from the DCSRC microprocessor circuit, I wanted minimal loading of the DCSRC circuit.  Yes, there is a CMOS version of the 555 chip but it is relatively expensive and not so common in the average DIY'ers parts stash.

In retrospect, I never expected anyone else to make one given where the thread was going at the time!  But it's all good.

Good point on the 555, I didn't consider the power draw.

I didn't realize the DCS-RC did such a short WD, it never failed to work with an engine on the track, so apparently it was enough.

I'll stick a 'scope on the units and make sure they're in the .9-1.0 second range for the resets.  I have time to procure the resistors, I'm waiting on the replacement boards.  Bummer that I missed the header being the wrong size.

With the board mistake and the resistor turning, I'm beginning to understand the source of the old saying: No good deed goes unpunished.

I'll be more careful of offering such services in the future.

I think your service is above and beyond the call of duty. 

Of course the party hasn't even started; the fun comes when guys start hooking it up!

I'm thinking after this batch is built and the dust settles, perhaps just put the Gerbers on the OSH Park site with a link to and from this thread.  Anyone who wants to build one can order boards directly from OSH Park for a few dollars and the parts from DigiKey (probably should publish your parts list at some point).

Stan, once I get these out, I was planning on posting them, I'll put them on OSH and link here as you suggest.  The experience in calibrating them will be useful information and will affect the parts list as well.

I take no responsibility on the hookup, my testing will insure they're performing the resets in the proper time frame.  I am providing the matching connector to solder to the DCS-RC PCB, so they don't need anything but a soldering iron, some solder, and enough skill to solder the socket to the DCS-RC PCB.

I agree with Stan about your service being above and beyond the call! Very nice of you to do all this.

Also as for the list I was thinking about, I think your posting at OSH Park and here with links will probably be about as good as it can get for one project. The offer still stands if more of these projects develop over time. I would be happy to make a listing of them all in one place so they could be easily found. 

Thanks again to GRJ and Stan for doing these projects and discussing them here. Good help in learning for all of us trying to follow along and learn a thing or two (or sometimes just keep up). Also very interesting, to me anyway.

Well, the boards came in, and I built one (except for the 1M resistor).  I then clipped a couple different values in, and Stan was right, there is significant timing variability, the timing with a 1M resistor was only around 800ms.  I have different values coming from 820K to 1.2M, good thing resistors are dirt cheap.   It looks like it should fit into the DCS-RC easily, I did take the three pin support strap off to give it a lower profile, total of about 3/8" above the DCS-RC PCB.

FWIW, I got 30 PCB blanks, and I have components to make 30 units.  I have 20 already committed, so there are 10 more that I can deliver.  I have 8 more for people that wanted one, so they get first crack, that leaves two of them still "free".  Act fast, because I might decide to keep one.

For people already lined up for these, they should ship out next week, I figure I can whack them together when I get back from York, the resistors to tune them will be here by then.

I am interested in purchasing one or two but not sure if you have sold out.   Send me an email if there are any available.  Thank you.

Hard to send an email if you don't have an email address in your profile.   I have two left that aren't spoken for, if you want them, send me an email, it's in my profile.

Those looks great, nice and compact! It amazes me how you guys get these things turned out so quickly. Thanks again GRJ and Stan.