Dave,

Do you know that the PCP issue are pointing to is an issue from 2005 boards?? My boards I have are 2017 and everything is WORKING.
Even the 2005 board is working. What do you want me to do? You want me to delete my findings??

I don't get it.

Inobu

Dave,

You don't have to accept it but I know I saw the same problem that the OP said he had. I went through this procedure and cleared it. I got the components new out of the box and it showed the same problem.
If you want to say that I have hardware issue that ok.....but like I said I did not start this thread Larry did. His hardware has nothing to do with mine but he is having the same problem.

I think if you look in the past you pretty much said the same thing and recommended for the guy to go with the FTDI and USB Serial route. I think that he was experiencing this same issue.

I would venture to say other people will come across the same issue. In any case I know what to do the get these components going and that important to me and my clients.

Hopefully Larry comes back and tries it as I did it to help him out as well as anyone else who has the same "hardware" problem that I have.?

Inobu ?

Inobu,

On 1 Jan 2020, at 4:05 PM, Dave Heap via Groups.Io <dgheap@...> wrote:

It says there's something wrong with your particular system. I'd suspect a below-spec RS485 chip that is causing cab bus flooding and buffer overflow in either device and hence locking up communications. This was what was happening when we worked through this original problem with NCE and came up with the PCP fix Your "workaround" is possibly just avoiding this.

I know you have tried several different Power Cabs. But you haven't reported whether your PCP has the extra-resistors fix documented at the following link:
<>

The site claims that only a limited range of Power Cabs are affected, but I've had reports of NCE USBs that show the symptoms of sub-spec chips.

Setting up an unterminated RS485 bus (bad practice) and relying on special lockup-proof transceiver chip models wasn't a robust move. It was done for what seemed a good reason at the time (reluctance to put terminators in the Power Cab due to its secondary applicability as a slave cab on already-terminated Power Pro systems). When we tracked the problem down it was realised that putting terminators on the PCP (of which there should never be more than one) solved the problem completely. The retrofit was recommended and all newer production runs of the PCP were fitted with terminators.

Dave in Australia

开云体育

It you allow the NCE USB interface to go RED all bets are off. The handshaking will never start.

What you do is watch JMRI when launches Panel Pro. Connect the coil cable from the Power Cab to the NCE USB Interface. This starts up the micro processor as the USB portion
starts up. Within that initialization sequence the USB Interface will connect the two sides and you will be good.?

If this happens too far apart the ports are deemed bad by both sides and each side is labeled? bad.

With a normal NCE USB setup, the 12V power to the NCE USB and Power Cab is established and the microprocessor started up long before JMRI is launched.

Which panel editor? The LayoutEditor panel will allow you to click (on blank part of screen) and drag (a selection rectangle). Then click inside of it and drag to move it.

Bill,

On 1 Jan 2020, at 9:19 AM, WHurry <Wttr@...> wrote:

System consists of A NCE Power Pro-R 5A command station/booster, PB5 booster, and a stand alone Loconet.

While shopping for a programming track booster and have the ability to change CVs on sound decoders,

Does your current Power Pro have trouble with reading CVs on sound decoders? More recent Power Pro's than mine have a circuitry change that reportedly solves the problem.

I have:
- An older Power Pro that has trouble with some sound decoders.
-- I briefly had a PowerPax booster but it was unsatisfactory.
-- I now have a PTB-100 booster, I always leave it in-circuit and has been fine with almost everything I've thrown at it.
- An NCE Power Cab. It has been fine with everything I've thrown at it.
- A SPROG IIv3. It has been fine with almost everything I've thrown at it.

- The Power Pro with PTB-100 has good compatibility and a speed edge.
- The Power Cab has best compatibility, you can use the Power Cab as an extra throttle when not programming. It can also be taken portable as a standalone system, with or without laptop.
- The SPROG has good compatibility but is slowest. It is compact for portable use but needs a laptop accompanying it.

Dave in Australia

That falls in line with what I am saying.

It you allow the NCE USB interface to go RED all bets are off. The handshaking will never start.

What you do is watch JMRI when launches Panel Pro. Connect the coil cable from the Power Cab to the NCE USB Interface. This starts up the micro processor as the USB portion
starts up. Within that initialization sequence the USB Interface will connect the two sides and you will be good.?

If this happens too far apart the ports are deemed bad by both sides and each side is labeled? bad.

The one piece missed was the connection of the coil cable. This is the on switch (12 volt power) being applied to the USB Interface at the same time the JMRI side is?
trying to establish communication.

I tested this all kinds of ways and I got everything up no problem. I was experiencing the same issue that Larry was with the Raspberry Pi4. Using this procedure
I was able to turn it up everything from scratch.

开云体育

Group,
Something I noticed when constructing my Control Panel using CP editor.? I have the panel finished and it extends over 4 22" panels, in a 2 x 2 arrangement.? I recently purchased a large 55" screen to see if I could get the panel on just one screen.? However, because the screen resolution is what it is, the panel needs to be zoomed to a smaller size to fit on the screen.? When I do this, the icons do not function correctly.? In other words, the toggle switches and control levers do not function.? Has anyone else see this?? Any solution?? Also, when zooming the panel, the resolution of the icons really degenerates.? Any solution for this< or is it an function of a bitmap image rather than a postscript type?
Any insight to these issues would be appreciated.
Thanks in advance,?
Keith Williams
Hoschton, GA

Our railroad, the Castle Rock and Pacific, is 2500 sq.ft. on two levels.? It is freelance, point to point - Galveston, TX to Castle Rock, CO.? There is a classification yard in each of the end point cities.? Beyond Galveston and Castle Rock there are two double ended staging yards that are the true endpoints.? The staging yards are not active during an op session, that is, there is no switching in staging - what ever car order goes in is that same as what comes out the next time that train runs.? We are currently running? a 5:1 fast clock so four hour sessions only represent 20 hours of a full day.??

Now for CAR ORDER on the Castle Rock and Pacific...

We have 4 through freights that run staging to staging and return during a 24 hour day - so 8 through freights.? These trains pick-up and set out cars only at the two major class yards as they run across the railroad.? Before each session, I check and block any of these trains that are in staging.? If a train is coming out of Galveston staging, Galveston cars are in block #1 and Castle Rock cars in block #2 followed by the remainder of the train.? I use Operations Pro and love it.? I do not run in real time but rather build, update, print and terminate all manifests in sequence ;for all trains that are expected to run in that 20 hour fast clock session.? When a through freight comes out of staging for the first time during the session it is blocked because I blocked it before the session.? When it enters the second and last classification yard before entering staging on the opposite end of the railroad, the yardmaster a that yard has orders to reblock the train so that when it reappears on its next run out of staging it is already blocked.? Since I have generated all of the manifests in advance, I know how the train should be blocked when it reappears.? To keep things kind of "railroady" I print the blocking instructions on a pseudo "31 Order" and place it in the "restaurant rail" on the fascia above the yard and staple a copy on to the associated manifest. (a "restaurant rail" is one of those linear rails that are used to hold food order tickets in a restaurant)? I also stamp the Operations Pro generated manifests and switch lists with a green inked stamp that says "Blocking Required see 31 Orders".? In a pinch, the YM can delay the blocking until the train returns because no one else will see the train until it has passed through his/her yard the second time but we encourage them to get it done at the first opportunity.? This works for us and keeps things in order.

Not really on topic but it goes with staging on our railroad - we swap open loads.? A train may go into staging with an empty flatcar but when the train is built again by Operations Pro it may have any number of different open loads for the next trip to a different industry.? When an operator gets a train out of staging they check the manifest for cars that have open loads and change them if necessary.? We keep a file cabinet full of different open loads.? All the operator has to do is remove any improper load and replace it with the load listed on the manifest.? In this way a flat with a steel plate load going into staging may come out with a load of? pipe or tractors or...whatever.? The downside of this is that you don't have all of the neat detail of chains and such to fasten down the loads but it does allow you to have sensible loads going to each industry.? Food for thought.
Dave...
Castle Rock & Pacific
Franktown, CO

Inobu,

On 31 Dec 2019, at 6:16 PM, Inobu One via Groups.Io <one2beros@...> wrote:

The key is starting the line of communication is the timing of the coiled cable connection to the NCE USB Interface. When that connection is made
the preamble is sent out. Once the handshake is established the port is considered good and open for communication. This is the missing piece
that's been alluding us for years. Even if the combination was right if the handshake wasn't timed it would fail. Creating the wild goose chase.

Your symptoms indicate that you've probably got one of the sub-spec RS485 driver chips in one or more of your devices, probably the NCE USB and don't have the PCP modification necessary to make these work. The result is that the NCE USB connection will need unplugging and replugging after power-up to unstick a stuck RS485 driver chip.
<>


The NCE USB will report the version number even if it hasn't established communication with the Power Cab. All it needs is 12V power from the cab bus and correct jumper settings to match the baud rate JMRI is using. Communication on the NCE Cab Bus only starts when other commands are sent by JMRI and that's when the lockup can happen, shown by a stuck-on Cab Bus LED.

Dave in Australia

So basically, if you zoom out farther, it permanently changes the size of the viewing area (at least until you re-load the panel), but it doesn't change the viewing area to be large enough to see the entire panel while?it's zoomed out that far. I don't know what would make it do that (yet).

I also have an Arduino (Mega) DCC++ base station.? I can run a loco on the main track and I can program a loco on the programming track, but I can't run the loco on the programming track.? Does that work on your system?

Alright, I've done more testing with the program and I have more information.

The problem appears to be difficult to replicate because it's very unusual. It only shows the part of the layout that can be shown with "no zoom" setting once you zoom out. However, it gets weirder if you zoom out farther- if you zoom out to a certain point, it will be able to show the entire thing at the lower zoom factor again.
I don't know what that means regarding what causes the glitch yet, but I hope it helps you with troubleshooting.

Inobu,

On 31 Dec 2019, at 6:49 PM, Inobu One via Groups.Io <one2beros@...> wrote:

Your first post was dated Dec 8 and its Dec 30 and you are still no where. Get a windows laptop from Goodwill and be done with it. Its not going to kill you to learn another platform. I've don't particularly care for Apple products but I have them because portions of my customer base uses them.

That's a crass and unwarranted statement. I've satisfactorily used a PR3 on both Mac and Windows and both were easy to set up and use.

I've seen no evidence that this is a Mac-specific problem.

The fact that the setup worked in the past may point towards a hardware problem with the PR3.

Some if these issues are really difficult. i just figured out a problem with the Raspberry Pi that was dated 2 years ago.
If Marc, Ken, Dave, Dave and Bob (off the top of my head) haven't talked you through this then its a difficult issue.

We are all limited humans and don't have time to talk through every issue. I already spend far too much of my day talking people through problems.

Dave in Australia

Ken,
The reading of the signals meaning is partly what was/is known as "route knowledge". Every driver had to be separately qualified for every route before they could be allowed to be in sole control of a train on it because they needed to "learn" the signals.??

You are correct that there are two types of dolls seen here - posts and lattice (white posts on the left and "black" lattice on the right).? The color of the latter is probably more to do with rust and accumulated smoke than a "choice" they along with the bracket would most likely have been painted white originally. Other than ensuring that posts and dolls higher than 40' were all lattice (for reduced wind-loading - some of these posts could be >70' high!) the choice might well be because of the parts to hand when the signal was altered or repaired or indeed which company or region happened to be in charge then. The LMR would have tended to be putting up posts, while the ER would have hankered for/retained lattices. Many lines changed hands several times during their lives. But that does not designate a material signalling difference as you guessed.? Luckily this allows almost any combination to be explained away if a precise prototype is not being modelled!

There are some useful hints to help read the signals without route knowledge, you'll notice that the dolls are visually grouped by how they are placed along the gantry into a set of two dolls and a set of three dolls with a larger gap between the sets.? The sets are then placed (in this case) to the LHS of the line to which they apply under the bracket. Each set of dolls can only show one Aspect at a time. I say "in this case" because one of the four regions of British Rail (WR) inherited independent ways from its pre-nationalization forebear (GWR) and not only continued to use lower quadrant signals but placed them (usually) on the RHS of the track - this was naturally because their steam engines were Right-Hand drive unlike almost everyone else's....?

In every case, if there were obstructions to "sighting" in the normal signal position it could be altered to allow for a clear view.? Another common trick (and frequently a use for that 70' post) was to place two arms one at the top and the other near the bottom and have them "co-acting" as mirrors of each other. This would allow a driver to see the next signal's aspect even if several overbridges intervened both from a distance and close-up as long as they knew where to look.?

This route knowledge was not only because of the precise meaning of the individual signals, but also because the signals would come in sets ALONG the route. In a UK semaphore world, a distant signal arm (yellow with black chevron) read "through" a set of stop signals (red and white stripe - often referred to as "home" even though that is not precise) - that is to say - several stop signals in a row would have to be "off" for the distant arm at the rear of them to be moved to "off". The stop signals for a box would be positioned to allow the track to be divided for shunting and low speed work but often not be far apart enough for adequate braking at high-speed.? If the distant for a box was "on" as the train approached, one or more of its following stop signals were almost certainly at danger, and so they would all have to be approached at a low enough speed that the train could stop in front of any one of them.? The boxes were not placed at equal distances apart either.? Knowledge of how many stop signals there were in a signalboxes "set"? was therefore needed so that a driver could know when the next block (under the control of another box) was entered and so a higher speed could be maintained/resumed until the next distant arm could be seen. You'll see here that the distant arm for the following box is placed under a stop arm controlled by a closer box on one of the dolls in each set, because the two boxes are positioned too closely together for a separate distant signal post and adequate braking distance. In this case the stop/home and distant were mechanically interlocked (slotted) so that the distant arm (from the following box) would only go off if the stop arm above it (from the closer box) and all following stop signals it controlled were also "off" even if the following box had cleared the distant (after first clearing all its own stop signals, of course). A side benefit of the stop/home & distant was that it allowed denser positioning of trains and this was an advantage retained when 3-color lights were slowly introduced to replace semaphores.? ?

What was also important in the British context was that high-speed passenger trains doing >100 MPH would routinely share the road with low-speed unfitted (mechanical brakes on each wagon and a guard's van) goods doing <30 MPH and fitted freight (engine controlled vacuum brakes) doing ~70MPH. It was up to the Signalmen to sort all this out - via communications usually made only on "block-bells" and "block instruments" (that is there was no routine use of the "speaking instrument" which was felt to be prone to idle chit-chat and safety-reducing confusion). The middle doll in the RHS set here has a home arm for Up Fast to Up Slow and that might be used to allow a fast fitted freight to be moved from Fast to Slow so it could passed by an express passenger train, after the fitted itself had earlier gone past a pick-up or coal wagon goods valiantly chugging along on the Up Slow, for example.?

Here's a??of three dolls - it is a single bracket (mast) controlling a single track that divides into three roads - each of which has not only a stop/distant pair also a subsidiary "calling-on" arm as well - to allow a train or engine to proceed even if the road is occupied (to add a carriage or an engine, for example). All these nine arms can only show one Aspect at a time. The relative importance of the lines can be judged by the relative heights of the dolls.

Hope that helps at least a bit.
??
Adam

What version of the program are you using? I'm using version 4.18 and the issue is still there pretty consistently.

I never figured out the cause of the original problem (which I now can't reproduce (with or without the setClip disabled).
I'm going to leave it disabled (since it was causing the scrollbars to get overwritten).
If (/when?) the original problem shows up again I'll look into again.