Chipping the Alco F units

I mentioned before that I had ordered some Southern Digital frames and Digitrax DZ126 decoders for the FA1/FB1 loco set that I want to put DCC in. I only ordered two frames, enough for one set, just in case they didn't work out. I could have gotten DZ123 decoders a bit cheaper, but the 126 has BEMF and better motor control, so the added cost was more than justified.

They arrived and I set to work to put them together. I'm happy to say they worked out great, I did make one modification that I will show to connect the wires and I did have to deviate from the instructions a bit for the motor wire length, but overall they were well worth the purchase price.

The first thing I did was strip down one of the locos and compare the frames.

 The left half of the frame is virtually identical to the stock frame, about the only difference (not pictured) is the tab that contacts the upper motor brush is missing on the Southern Digital frame. The right side is very different, having the relief cut out for the decoder. The pocket that the motor fits into is also deeper to allow room for the decoder wire to run to the bottom brush. This becomes important when modifying the stock frames to use on the B units.

I assembled one of the frames as per the directions and discovered that the decoder wire lengths that are specified are too short for the motor leads, they both needed to be another inch longer than the instructions said so that they would reach to the relief where the decoder sits.

The instructions also said the white magnet on the motor went to the top, but both of my motors had the white magnets on the bottom. I left the white magnets on the bottom and when I was done the locos went forward when the NCE PowerCab had forward selected and backwards when the PowerCab said reverse, so they must have been right.

The second thing is the way the wires connect to the frame. The directions say to solder the wires to the frame, but I found the solder just would not stick to the frames. So I figured out another way to do it.

On the left half there are indents cut in for some brass clips to fit in, these clips hold the plastic bearing blocks for the worm gears in place. Being brass they were easy to put some solder on for the black decoder wire.

I had noticed some indents down the sides of the frame and thought maybe the red wire would fit in the one closest to the rear of the frame. I checked, and sure enough it did. So I took my rotary tool and a cutoff disk and cut a wire channel in the top of the frame.

Next, I inserted the truck pickup strip into its spot on the right frame half and marked it where the slot in the frame was. Then I tinned the contact strip.

With the contacts figured out I soldered the wires to the motor, put some electrical tape in for added insulation, and put it in the frame with the gears.


The frame was then reassembled and the black and red wires soldered to the frame tab and truck pickup strip respectively.
 

 The red wire was then tucked into its new slot in the frame;

 and the decoder was nestled into its new home.

The trucks were put in and the headlight wired up to the blue and white wires. The directions say to just hook the white wire up but I did them both.

A note of caution, if you ever do one of these be sure to cut the traces on both sides of the light PC board. They are both connected.

I then put the assembled loco on the track and tested it out, and it all worked swimmingly. This completed chassis then went under one of the A units.

I then took a close look and noticed that with a bit of judicious filing the old frame would work for the B unit, I just had to take off enough material for the decoder to fit in the depression where the cab would be on the A unit. I used a rotary tool and cutoff wheel to accomplish this. I then cut a slot for the red wire, on the front of the frame this time, and then while holding the frame halves together cut a slot along the spine of the frame from the front to the top motor contact tab to clear the motor wires. Lastly I cut out the top motor contact tab on the left frame half.

With these modifications the decoder fits on the front of the frame with no problems. The light wires were just clipped short and taped off. The contacts were soldered on the same way as before, and the finished locomotive was tested and put under a B unit shell.

That left me a spare Southern Digital frame. As luck would have it I had another set of locomotives to modify, so the work was done and I now have not one but two FA1/FB1 sets in Erie paint.

All I have to do now is speed match them and I'll be all ready to go pull some freight.

UPDATE: When I went to speed match the locos I found the chipped units were much much slower than the non-chipped units. The non-chipped units ran a max of about 106 scale MPH (prototype top speed was about 65) while the chipped units were fortunate to find themselves above 35. I took the units apart and found the decoder wires were causing the motors to bind on the modified stock frames. So I had to cut a relief in the modified stock frames for the decoder wire and I also had to adjust the tightness of the screws holding the two frame halves together. The screws have to be tight enough to hold the frames together to slip inside the shell but no tighter otherwise the motor binds. Screw tightness also accounted for the slow speed of the Southern Digital framed units.

With those adjustments and modifications made the chipped units once again ran at the same excessive speed as the non-chipped units. I set CV's 2, 5 and 6 on each  individual unit to give each loco a starting speed of about 2 scale MPH with a top speed of 65 and a mid speed of 35. Now they all run together as they should.

I now have four chipped units and two DC only units. I'm thinking of selling the two DC units because they have the same numbers as one set of chipped units, but I may just keep them for DC only operations.

But I have them all the same.

N&W Varnish

The heavyweight cars that I had in the paint shop to be put into N&W livery are done, and they are beautiful. I am still waiting on four coaches, but I have a string of 6 that I could run...if they had the proper couplers. I have a set of MicroTrains trucks and couplers on their way, but in the meantime I have installed a dummy coupler on one end of the baggage car.

The smoothside cars that I had in the paint shop to be put into N&W livery are done as well, and yesterday while I was mucking about putting the decoder into the SD7 I had them behind N&W 601 doing laps.

I am told that the Pacific that was formerly in Lackawanna paint is almost done (I was putting it into N&W paint), and the streamlined Hudson shell is almost done with its fantasy Erie paint job as well. So soon I should have my four N&W coaches, my fantasy N&W streamlined Pacific and my Erie streamlined Hudson all set up.

Next step will be to put a chip in that Hudson, which may be a lot of fun, and then I'll be able to run the Erie consist. I'd like to take it to the next show in December but I'll have to see how it goes. I definitely will take N&W 601 and associated cars. I'd take 611 as well...but everyone has 611.

I must say that I am now convinced that DCC is the way to go. I still have my DC/DCC control panel set up on the home layout, but about the only time I use it any more is to isolate the front siding so I can use it as a programming track. I still have enough DC only locos to keep it, but they are getting to be a smaller and smaller percentage of my roster.

One day I might even get crazy enough to chip my Nn3...

Developments

When last we spoke of J class locomotives and DCC I had failed in my attempts to chip N&W 601, a locomotive that you may recall I had purchased in a non-running condition with wiring issues. As you may also recall I attempted to repair those wiring issues with a complete point to point rewire (using the color coded wiring from a used decoder), which I now actually believe was successful but did not work out the way I had planned.

I did order the Bachmann wiring harness, and it came in last month. I took the locomotive, the new harness, and the decoder that had been installed, uninstalled, reinstalled and once again uninstalled with me to a friends house yesterday. He has a Digitrax system connected to a laptop with JMRI software installed, which I hoped would be useful in diagnosing the problems with the install.

Using his work bench (since mine is currently in the living room until its new home can be prepared) the new wiring harness was prepared and installed. Once installed and tested the decoder installation was started once more on the PCB in the tender.

Right away I ran into the issue of finding one of the pads burned off of the PCB. Not to worry, there was still plenty of copper left on the underside of the board, which was actually better because the trace was connected on the underside anyway. To access this I pulled the board off the tender...and discovered the copper contacting strips, which were supposed to be contacting the trucks...weren't.

Not to worry once more, I tinned the truck contacts and prepped them to have the PCB wiring connected to them directly in the same manner that the PRR K4 wiring is done (which I believe to be a better arrangement even though it makes the trucks, for all intents and purposes, non-removable; in itself once again not an issue since I don't plan on having to remove the trucks for any reason, and if they do have to be removed they can be desoldered). I then installed the decoder, connected the trucks, connected the loco harness and reassembled the tender.

When the loco was placed on the programming track the Digitrax/JMRI computer was able to find the decoder right away and determine its programming parameters. The small hand held controller I had used in troubleshooting had programmed the loco with number 531, not 601, and even though I attempted to change that address with the MRC controller it obviously was not effective (which explains why the loco would not work with the MRC but would with the small handheld). Once that was fixed the loco was put on the layout...and it runs beautifully.

I ran a couple of laps on his layout but since he uses Atlas C55 track and apparently all of my cars have pizza cutter wheels I was not able to run it with a consist. It does however start and stop smoothly, it responds well to the throttle and the lights work as they are supposed to. He was amazed that such performance could be had from a Bachmann (he is one of those who judge the B-mann by their hideously inferior white box models and not by their vastly improved Spectrum line).

Since my own layout is in disarray at the moment I am unable to run it with my MRC system, but I have no reason to believe it won't perform just as well at home as it did away. And now I have two N&W J class locomotives to play with (and I can double-head them if I want to!), with a slew of cars in the paint shop to split between them.

And speaking of cars in the paint shop, the Erie cars came back yesterday as well, and they are beautiful. I need to get trucks and couplers for them, and a handful of weights as well, and then they will be ready to go on the layout. Pictures, as promised, when available. I might, however, need to get a set of the ConCor PA1/PB1 locomotives in the matching Erie paint to go with them now since the PA1 I have right now is in the blue and yellow livery instead of the green and cream (and I don't have a B unit for it).

The only sad part of the day was discovering that one of my B-mann GP7's is once again acting up,  same issue as before with the dimming lights and slowing to a stop. This time the tab bending trick didn't work, so it may be time to replace that decoder board.

And that's the news from the Path Valley Railroad.

Once Again

Today I attempted once again to chip N&W 601.

First I disconnected the tender from the locomotive. Then I carefully installed a new in the package DZ126 decoder, paying close attention to which wire went where.

Then I took the extra wiring I had clipped from the decoder during the installation and used it to repair the loco to tender wiring harness, paying careful attention to what wire went where on each end. I carefully ohmed every wire out at both of the above steps, both to ensure continuity end to end in the loco wiring harness and to ensure that neither the loco wiring harness nor the tender decoder installation had any shorts from wire to wire.

Finally I reconnected the loco and tested it in DC. It worked perfectly, the correct LEDs lit up when the loco moved in both directions and the motor stopped and started smoothly in both directions.

However, when I attempted to program the locomotive the NCE PowerCab could not find the decoder.

I tried a new in the package DZ146 decoder. No joy. I reinstalled the old suspect DZ126 decoder. Nothing. I even connected all three of the decoders (one at a time) to alligator clips and clipped them first to the programming track rails and then directly to the output wiring of the NCE, and still the system did not see any of the decoders.

I then placed a good DCC loco on the tracks and attempted to reprogram it. I had no problems at all doing so.

I have several older locomotives using Bachmann Spectrum tenders, so I installed the decoder in one of those tenders. Still no luck. Finally I uninstalled the decoder and put the DC shorting clips back in, and once again the loco runs perfectly on DC. I reinstalled the decoder and verified that yes, it still does run fine in DC.

I then attempted to run it on my little snail speed DCC controller (which used to be sold through Z Scale Track but seems to have, unfortunately, been discontinued) and was able to program the loco to run in DCC mode with that controller. However, when the loco is programmed to run with this controller it will no longer work in DC. And since the NCE couldn't find the decoder I couldn't tell what the address was on it so I was not able to run it on the NCE.

I finally disconnected the loco from the tender and was able to get the NCE to recognize the decoder about half the attempts as I rolled the tender back and forth on the programming track. I was able to put it back in dual mode and program in the address, both long and short. I reconnected the tender and once again I was able to get it to run in DC, but not in DCC.

I have no idea what's up with this locomotive. I'm hoping that installing the stock Bachmann wiring harness will fix the issues and I'll be able to chip it and get it to work. But until that happens I have removed the decoder and reinstalled the shorting clips, the loco runs fine in DC, and it will stay that way for now.

And maybe I'll just leave it that way.

Distressingly Familiar

I attempted to chip N&W J class 601 today.

The tender to loco wiring is just as bad on 601 as it was on 611.

Looks like I'll be placing an order for a new wiring harness tomorrow.

Success!!!

My last DZ126 Digitrax DCC decoder has been installed into the SD7 SD9 (explanation below).

There were a few things that were kinda screwy, and it's probably all my fault anyway but I did get everything straightened out.

First off, when wiring in a decoder the red wire goes to the right side pickups of the locomotive and the black wire goes to the left. The motor brush that once was connected to the right gets the orange wire and the other motor brush gets the gray wire.. The blue wire goes to both headlights, the yellow wire goes to the back light and the white wire goes to the front light.

Before the mod the bottom brush on the locomotive was connected to the right contacts and the top brush was connected to the left contacts. I wired up the decoder as specified with the orange wire to the bottom motor contact and the gray to the top motor contact, and then I put it on the track.

Operating in DC mode the lights did not come on at all, which I thought was kind of screwy but OK. Then I connected the DCC controller and programmed the loco...and the wrong headlight came on. The motor however ran just fine, just like it was supposed to, so no big deal. Back to the soldering bench, reverse the yellow and white wires and now the headlight on the leading end of the locomotive comes on when it is supposed to.

However, there was another problem, and the headlights should have clued me in. You see, when I actually paid attention to the NCE controller display I found the locomotive was running backwards.

How this happened I have no idea, but the instructions were clear so I should have known; the instructions clearly state to connect the gray wire to the bottom motor tab and the orange wire to the top (the link goes to a TCS decoder installation page but the wire colors are identical). Why the locomotive ran properly in DC is anyone's guess, perhaps I was incorrect in thinking the red side contacts were the positive leads.

At any rate the issue is easily solved, back to the programming track and change the locomotive direction bit (CV29 bit zero) and all is well. This bit is supposed to be used to reverse the direction of the loco in consists or when you want to run it backwards; IE if it is designed to run long hood forward like the SD9.

In my case I wanted the loco to run short hood forward; if I had wired the motor correctly it would have and I would not have had to swap the lights. The direction bit isn't supposed to be used to correct motor wiring, but since I had already switched the headlight wiring it worked out.

Or maybe, since the only difference between the SD7 and the SD9 was the prime mover (the diesel motor), and since (according to the Wikipedia article) Burlington locomotive 346 is actually an SD9, I had it wired correctly all along...except for the original headlight wiring, which was incorrect because I mistakenly thought the short hood was the front.

UPDATE: Or not...the F on the short hood end indicates the front, which means I did have the motor wired backwards to begin with...and now the headlights too. Of course. And, the locomotive is clearly marked as an SD9. And also of course, the answer was right there all along.

Anyway, it now has a decoder, the decoder works, it programmed correctly, and I have it running in the direction I want it to run. Also, with the previously DCC'd K4, the entire railroad is now DCC when running the usual locomotives, and I now have seven locomotives I can run on the club tracks (well, six actually since the first loco I successfully converted, a Model Power Pacific, is in for repaint).

And now I am all out of decoders, at least until the broken one returns from Digitrax. I haven't yet decided what my next victim will be...

DCC'ing

I have four Digitrax decoders.

I put one into a Minitrix K4. It doesn't work. I open it up and find the black wire has somehow gotten cut and is touching the gray wire. D'oh!!!

I have three Digitrax decoders.

I install one in a Minitrix K4. It doesn't work. I open it up and can find nothing wrong. I remove it and the locomotive still doesn't work. I clean the motor brushes and now it works. I give up on this Minitrix K4.

I may have three Digitrax decoders.

I install the suspect one into a Bachmann Spectrum N&W J class. It works. I am happy.

I have two Digitrax decoders and a locomotive with DCC.

I install one into a different Minitrix K4. It works. I am happy. I also now have four Pennsylvania Railroad locomotives with DCC; two Bachmann GP7's with factory decoders, one Model Power Mikado that I got the bugs worked out of last weekend, and the K4.

I have one Digitrax decoder...

Sigh

I tried to put a decoder in a K4 last night.

I failed miserably.

One of these days I'll be able to do this without all this bother...I hope.

More Fun with DCC

Did you know that when you are programming a Digitrax decoder and it asks you what you want your minimum, maximum and middle voltages to be, they don't really mean voltages?

It's true!

So, with the "malfunctioning" DCC decoder once again installed in the Mikado, and properly programmed, the Path Valley Railroad ran one day totally digital (and totally Model Power powered). The Lackawanna Pacific pulled the passenger train while the Mikado handled the freight duties.

In the process I've found that my little layout just really isn't set up to run two trains simultaneously...but then again, that's not the way it was designed in the first place. It was designed as a switching layout with the possibility of roundy-round train running, and it fills those roles admirably.

If I wanted to run two simultaneously I'd really need to put in a dedicated mainline and turn the existing main into the switching siding, that way I'd be able to orbit the passenger train while I was switching the freight. Unfortunately I just don't have that kind of real estate to play with.

Now I just have to fiddle with the SD9 until I can get it to be a good locomotive on DC again, and then I'll try once more to put a decoder in it. I don't know why I'm so hell-bent on putting DCC into that locomotive, but it's become a quest of sorts.

I'm also going to put a decoder into the K4, I just haven't decided if I'm going to wait for the new motor and gear to get here before I do it.

DCC...or maybe not...

The replacement Digitrax decoder came in the mail Friday.

I immediately put it in the SD9.

It ran about as well on the pulse power as the Pacific did, so I called that one a win and turned on the DCC system. The decoder programmed and all seemed well...until it was time to run the beastie.

The lights came on (in the wrong direction) but the locomotive would barely crawl. I checked the programming several times to no avail.

So I pulled it out of the SD9 and put it in a Model Power Mikado. Same deal, it programmed just fine but it wouldn't move the loco at all.

So I pulled it back out and gave up on ever getting a decoder in the SD9.

I've been told I set the speeds wrong, so I'll put it back in the Mikado and try it again when I have time. If it still doesn't work I wonder if Digitrax will take it back again...

DCC

The local model railroad club, which I joined recently, is a DCC only club. So if I wanted to bring locomotives to play on the display layouts they have to be DCC equipped.

Currently I have a pair of Bachmann GP7's that came from the factory with DCC decoders, but that was all. I decided to go ahead and convert a few. First I selected my Bachmann J class only to find out the wiring going between the tender and locomotive was shot. I have an Email to Bachmann about repairing it, but I've ordered the wiring harness in case I don't get an answer. Their customer service tends to suck a bit.

Next I decided to tackle the LifeLike SD9 that I run on my home layout. Following these instructions (with the exception of completely disassembling the shell, not necessary since it comes off in one piece) I first modified the trucks for improved pickup...which I highly recommend, if you have this locomotive...and then following these instructions (with the exception of modifying the rear weight as described in these instructions) I installed the decoder.

It was then I screwed up. I didn't test it on DC as the Digitrax instructions say to do, I just put it on the programming track and attempted to program it. The system couldn't find the decoder, all it did was blink the lights. The instructions say this is A Bad Thing, so I took the power off immediately. I then tried to run the loco on DC and the lights would just flash once when I turned the throttle up.

I tried disconnecting the motor and then the newly installed LED lights one at a time to no avail, so then I just connected the power wires to alligator clips and connected the decoder directly to the programming track. The system could still not detect the decoder, so either the decoder was bad from the factory or somehow I fried it. Fortunately Digitrax will replace it free, even if I screwed it up. Now THAT is customer service!

So then I turned to my newly acquired Model Power Pacific. It ran great on DC, so why not. Installing a decoder in a Model Power Pacific is a dead simple task as long as you aren't trying to hook the headlights up, and it only took a little time to do. The loco ran forward and backwards with no problems on an oval powered by a cheap train set power pack, so off to the programming track it went!

The programming went smoothly and soon I was happily running the locomotive around the track. The problem is it won't run on DC now. Whenever I try to run it on DC it surges; full power, down to a crawl, then back to full power. My layout DC is a PWM throttle so that may have something to do with it, but it is disappointing to say the least.

And now my best running Pacific on DC is once again the ancient Atlas.

As a side note, I also acquired a new Pennsy locomotive, a Minitrix K4.

I am disappointed to say that it runs just as poorly as Spookshow says it does (I have the one with the red gear, of course), mostly because the two drivers on the fireman's side that are wired for pickup have traction tires on them (!!! Really? You couldn't have put the traction tires on the middle drivers, neither of which are wired for pickup???), but fortunately I found some instructions to make it run better.

I have the Bachmann trucks on order (along with the J class wiring and a couple of Pennsy shells for my GP7s) and as soon as they get here I will install them and hopefully overcome it's power pick up problems. Hopefully the Motorman will return my email and tell me the good news that he has a motor upgrade kit for it as well.

So kind of a mixed bag lately railroad wise, but all in all not too bad.

Trials and Tribulations

I finally got the DCC setup installed and working.

Installing the system was simple, I installed a DPDT switch connected to the common rail and Atlas track switch panel, connected one side of that DPDT switch to the DC controller and hooked the DCC connection panel up to the other side of the DPDT. That way the layout is either all DC or all DCC so there is no danger of running from one into the other and causing system or locomotive damage.

Second was to program my locos. I have two locomotives, a matched set of Bachmann GP7's, that are DCC equipped. I threw all the track switches except the front main to the center off position and placed the locos one at a time on the front main for programming. One of them went just fine, but the other one didn't want to run. The lights would come on when commanded, but when I pushed the throttle up the loco moved, then the lights dimmed and it slowed to a stop.

I threw the master mode switch back to DC and tried it that way with the same results. I put the first loco back on the tracks and it wouldn't move at all. I threw the master mode back to DCC and the first loco ran like a champ again. Definitely something in its programming then, a problem for another time.

I took the shell off the second locomotive...which was a chore in and of itself, those things are on TIGHT...and removed the decoder. There are two little springy contacts that supply power to the decoder, I bent them out a bit and reinstalled the decoder, and then the loco worked as it should in both DC and DCC.

On to the first loco. I tried a factory reset according to the manual but this did not solve the issue.  tried a factory reset on the second loco and then it wouldn't work on DC although it worked like a champ on DCC, just like the first one (and completely opposite of what I wanted). Off to the internets!

The issue, according to the internets, was the decoder dual mode function was set to off. The dual mode function was included in a thing called CV29, whatever that is. More digging reveals a CV is a Configuration Variable, and there is a table that shows what they all do. All very well and good, but how do I know what to set them to? More internet digging!

Fortunately the Digitrax site has some Control Variable (CV) calculators that I was able to use to determine the bit that sets the dual mode (DC/DCC) on or off. As expected, when I checked it the bit was set to off. I programmed them each for dual mode again and once more I am able to run both locos on either DC or DCC as I wish.

In my experimentation I found two of my turnouts aren't quite up to par as the locomotives short on the frogs, so I'll have to work on them a bit, but so far everything is working as it should. One thing that annoys me about this particular setup is I am apparently not allowed to run DC locomotives using the DCC controller. Not a big issue I guess since I am able to switch the master modes over, but if I had a larger layout it might become a factor.

At any rate, the NCE Powercab is working as it should and I'm having fun with it.

A Visit from St Nick

While I was away on vacation the following items arrived at the PVRR Corporate Office:

The DCC system will be installed on the Clarinda layout and one of the three long haul tenders will be used for an Erie K5A project. The other two will likely find themselves behind Atlas locomotives, most likely Pacifics as well. But first, more on the Erie K5A.

Photo from http://lists.railfan.net/erielackphoto.cgi?erielack-02-24-07/Erie_2943_Marion_9-26-49_Ron_Dukarm.jpg

This is Erie K5A number 2943. Her last engineer was a fellow named Jones, who just happens to be my wife's maternal great-grandfather. She was replaced by an Alco PA1 and Mr. Jones was the first engineer on that locomotive. I gathered this information from a couple of pictures that my wife's grandmother had and thought it might be a nice build.

The next step in the K5A project will be the procurement of a suitable JNR C57 chassis for the boxpox driver sets. I have a spare Atlas Pacific shell and various bits and pieces off of both Atlas and Model Power Pacific locomotives for what I hope will be the proper cylinders and trailing truck. The rest will be Details Associates pieces and polystyrene.

The DCC system is a new venture for me. As far as I know only two of my locomotives have decoders, a matched Bachmann GP7 set, so I am going to be learning all about DCC with them. Time will tell if I decide to stick with DCC, but since the model railroad club I am joining next month is all DCC it is likely I will be converting at least one or two steamers, too...including the K5A most likely.

I hope Santa was good to you as well.