A journal following the history, design, construction and operation of Bernard Kempinski's O Scale model railroad depicting the U. S. Military Railroad (USMRR) Aquia-Falmouth line in 1863, and other model railroad projects.
©Bernard Kempinski All text and images, except as noted, on this blog are copyrighted by the author and may not be used without permission.

December 31, 2018

Happy New Year

Sunrise at Daytona Beach Shores, Florida
Another year is upon us. It will be a special year for me, as January 30, 2019 will mark the tenth anniversary of my starting work on the Aquia Line and starting this blog.  I have a special treat planned to mark that event, and no, it is not tearing out the layout.

Tapered legs
We were  in Florida for two weeks during the Christmas vacation for  family time and golf. Nonetheless, I managed to get some layout related work done.

Planing the slabs
Domino holes for table top, didn't get them quite centered across the width, but they were all consistent. 
Table top glue up
Mock up of desk. Final assembly will happen in Virginia
      I hauled down a pile of maple lumber from Virginia to my brother's wood shop. There we used his tools to joint and plane the slabs, and cut the tapered legs. We also used my new Festool Domino machine to make floating tenons in the desk top. I returned to Virginia with the wood pieces to do final assembly and finishing.

Meanwhile, Seth and Stephen were modifying the Arduinos software and making a few other improvements to the circuits.  We'll get back on the telegraph system as soon as the revised parts arrive.

In the meantime, I have been super busy with a secret project. I need to wrap that up soon, as the back log of projects is getting out of hand.  My 2019 Do List is chock full of interesting projects, such as models for clients, and  new products for Alkem Scale Models, but it is starting to get scary.

Best wishes to all my blog readers. I hope you have a great new year. Thanks for your support by reading this blog.

December 13, 2018

It's Alive!

Tom Pierpoint helped wire the first telegraph station today. Most of the process was straight forward, but we did encounter some curveballs.  Tom had ordered some jumper cables from Adafruit. They were nicely made, but the female end did not fit securely on the posts of the rotary switch. So I had to solder each connection, instead of simply pushing each one on.

First wires to be attached
Also, we had some confusion on how to wire the train numbers to the rotary switch. The Arduinos can handle 11 inputs. We  need to be able to OS train numbers 3-12, but we also need train number 1 to program the station number into the Arduino. That is because the Arduino uses the rotary switch settings at start up to set the station location. We think we sorted it out by making train 12 on the panel actually be train 1 to the Arduino.  We'll see if that works.

After Tom left, and we had a dinner break, I went back through the documentation for the tenth time and finished the final wires. Sure enough, I got the system to work. All train numbers except 6 work properly for regular and extras.

When I select train 6 and push either switch, the small LED on the back of the Arduino flashes, indicating it got a signal, but the message doesn't start playing. The wire connections appear good, so it might be a software issue.

Fully wired with some test leads to the sounder
Another glitch is that the reset procedure is not working. This station was set to Brooke, probably because we had the rotary switch set to 2 when we fired it up. It should be Aquia Landing. Stephen suggested that I reset while repowering and that worked. The station now has the correct station code. Yeah!  But now 5 is sending Train code 6 and 6 still doesn't work. Oh well, it's getting closer.

We still need to build 4 more station panels and the dispatcher desk, so lots of work to go but, so far I am pleased with the system. This is US Army Dot Code, a simplified version of Morse code adopted by the US Army during the civil war to allow new operators to be trained quickly. It was easier to learn than RR Morse, but did not allow messages to be sent as rapidly as regular railroad morse.  We have the speed set so that it sends messages slow enough, that even a new operator can take down the dot code, and then translate the messages with a cheat sheet.  Perhaps, now we can have a dispatcher job position on the RR. It won't be the most exciting job, but it will be a change of pace. And everyone else in the layout will enjoy the sounds of the telegraph clicking away.

December 12, 2018

Telegraph System Design

Two prototype telegraph sending stations for comparison
We have been working on the telegraph system for the Aquia Line. I had mentioned this before here Ripple Effect - Adding a Telegraph.
Last spring, Seth Neumann and Steve Williams, of Model Railroad Control Systems,  built an automated system to my specs. I wrote up a spec sheet and they designed, wrote the code, built the circuit cards, and provided documentation for an Arduino based system  that meets that spec,  plus they added some embellishments.

This was my turn to be a defense contractor after many years of being involved in defense acquisition from the other side as an operational tester and budget analyst of defense systems. If only our defense contractors were as capable, customer oriented, and under-budget like the guys at Model Railroad Control Systems. They did a superlative job, and are willing to help get it working too. I can't say enough good things about these guys.

Circuitry behind the sending panels
Now, with some help from Tom Pierpoint and Amby Nangeroni, I have been working on getting the system built.

I started with designing the telegraph sending stations. The photo at the left  is the prototype set up before we add the wiring harnesses. It will be a partially automated system.

Jumper ribbon cable
The lead photo shows two designs for the telegraph sending stations that I considered. I am going to use the upper design, as I believe it looks more like a 19th century instrument to me, even though it uses some modern control components. Remember, during the civil war, there was no electric power except for battery powered telegraphs.

Each of the 5 stations on the railroad will have a sending station like this. When a train departs a station, the conductor selects the train number using the rotary switch, then pushes either the regular or extra button to OS. The Arduino does the rest and the sounder in the dispatch office plays the USMRR dot code. The dispatcher will get the message and annotated the train sheet with the appropriate data.

Tom found some neat ribbon cable that we will use to wire up the rotary switches to the terminals on the circuit cards. We will also have to run a Cat 5 cable from each station to the control panel. There is no data bus, each station has to have its own cable run.

At the dispatcher desk, a colored LED on a control panel will light up indicating the calling station. Thus the Dispatcher can tell at a glance what station is calling without having to know the USMRR dot code. However, the system will also generate the appropriate USMRR dot code for those wishing to listen and decode the message, to get the rest of the data such as train number and class.

The system will also generate a set of typical military railroad messages at random intervals. These messages are based on actual telegraph messages that I have obtained from the National Archives.  Those messages serve no functional purpose except to add audio scenery to the layout, and give the DS a chance to practice decoding dot code.

Design of the dispatcher's desk.
I plan to build a small desk for the dispatcher over the Christmas break with the help of my brother, Rob and his wood shop in Florida.  The desk design is shown here.

Right now, the desk is a pile of maple lumber on its
way to my brother's shop in Florida.

 The desk will reside in my home office, so it is somewhat small to fit the space I have available. It's a basic table design based on one I saw at the Station Museum at Ellicott City, MD.  The central portion houses the control panel with  the indicator lights. Not shown is  the sounder and telegraph key.

I'm taking design cues for my dispatcher office
from the Provost Marshall office in the Ellicott City Station Museum 

December 11, 2018

Rappahannock Bridge Mock Up

Mock Up of Rappahannock River Bridge
The Aquia Line is coming up on its tenth anniversary. I've been thinking a lot lately about what comes next. Of the options that I considered that expand the Aquia Line, the one that seems to appeal to me the most is the Fredericksburg Expansion. That expansion is a what-if scenario that I discussed earlier  such as, Fredericksburg Plan RevisedConcept Art for Fredericksburg Expansion and Finding a Foundry.

One of the worrisome aspects of this plan is that the fascia under the bridge will be quite low once I subtract the height of the bridge from track height.  This is because the track in this area has to be no higher that 50 inches from the floor to clear under the stairs.  Obviously, I'd have to relocate the TV, which shouldn't be a big issue.

Also problematic are various water levels. The track at Aquia Landing is 51 inches above the floor, while water is at 50 inches. To model the 600-ft long and 75-ft high Rappahannock Bridge would require the river level to be 32 inches from the floor, while the bridge would be 12 feet long. I don't have room for that, so I scaled the bridge to 6 feet between end abutments and 12 inches above the water, making the water surface 38 inches from the floor.

Would the lower water level here cause a visual disconnect? To get a feel for the situation, I made a mock up of the bridge and fascia with red rosin floor protection paper. I placed a 8-car train and one engine on the existing PoLA benchwork above the bridge mock up.  Then I took a photo of the mock up and used photoshop to enhance the scene with a backdrop and some scenery.

The good news is that the scene seems to work. Because you really can't see the bridge and Aquia Landing at the same time, the geographic incongruity is not that noticeable. To test it out, I asked my wife to look at the mock up and see if she saw anything unusual. She didn't comment on the water levels, but she did ask, "you're going to get rid of PoLA, which everyone likes?" Well, that is the question, isn't it?

I am leaving the bridge mock up taped to PoLA's benchwork to get a longer term feel for the scene. In the meantime, I've been working on several other projects. I am making slow but steady progress on the bulk carrier for PoLA.   The garage renovation is almost done, and I have been busy with videos and customer projects. I've also been sketching layout plans that would replace the Aquia completely.

December 3, 2018

Paul Dolkos's Baltimore Harbor District 10th Anniversary

Mat, Pete, John, Lance, Doug and Paul. (L to R) I took the photo.
Today I had an opportunity to operate on Paul Dolkos's Baltimore Harbor District HO scale layout. This op session marks the tenth year that the layout has been in operation. Participating in the event were John King, Mat Thompson, Pete LaGuardia, Lance Mindheim, Doug Kirkpatrick, Paul Dolkos and myself. Congratulations to Paul and Linda.  I put together a short video showing some of the action. Paul might not remember, but I had the honor of being the first outside visitor to the layout 10 years ago when he started. I still remember the small pile of sawdust on the floor.

Also, I should mention that Pete LaGuardia is also celebrating the 10th year on his large HO scale layout.

I ran the Carroll Street Job today.  Several years ago I did this same job and I came away inspired to build a similar layout. Thus was formed the germ of an idea for my PoLA layout. It has basically the same design concept  as Carroll Street, only enlarged to accommodate modern equipment. If you are looking for a small layout design idea, the Carroll Street part of Paul's layout is an excellent place to start. It's essentially a double ended siding with several spurs coming off it, but the spurs are long enough to have room for multiple cars and spots.

December 2, 2018

Op Session 17 - Half John

An interesting situation at Brooke

John Drye at Aquia Landing
Ops Sessions 17 on the Aquia Line and 22 on PoLA are in the books. The half of the crew today were Johns -  John Drye, John Salmons, John Barry. Of the crew not with the first name John were Tom Pierpoint, Brad Trencamp, and first timer Ambrose Nangeroni.

The sessions went well. I never heard a peep from the PoLA crew of conductor John Barry and engineer Brad Trencamp. John was celebrating his birthday too, which was Friday.

John Salmons and Amby Nangeroni at Falmouth
Meanwhile, Train 7 and 10 ran under conductor John Salmon's leadership with Amby on the engine. Trains 8 and 9 were conducted by Tom Pierpoint with John Drye on the engine.  Things were running well until engine Haupt started stuttering near the end of the session. I swapped it out and the crew finished with Whiton. Of course I wasn't able to replicate the problem once I had Haupt on a test track. Arrg DCC!

The crews did a pretty good job in filling out their paperwork too, though Train 9 forgot to do a conductor's report at the end of the session, but Tom can be excused since he was distracted as we finished the session with a discussion of the work on the new telegraph system.

John Barry and Brad Trencamp working PoLA
One of the great things about the Washington DC area is talented people we have in this area.  For example today we had 5 engineers, an architect, and economist in the crew. Two of the engineers are electrical engineers and they are interested in helping build the telegraph system.  Tom gave me some good ideas on how to build the panels, while Amby volunteered to do extra work sessions to help.

 The January session will be a work session vice an op session. Hopefully by then I will have one of the telegraph stations built so we can mass produce the rest.

Group shot of the crew looking resplendent in their Aquia Line shirts