points decoder

[MetmanUK]

Mousehole said:

As reported in other posts, I have only just become operational with eLink and Railmaster and have started looking at point control. On my layout all my points are operated by Peco PL10W point motors on Peco points controlled

with Peco point switches. All work well, even where two are operated simultaneously on crossovers and on Peco three-way points.
I have a Hornby 8247 decoder that I have set up successfully on Railmaster and it will operate a point motor when stand-alone

but when attached to a point there does not seem to be enough energy to throw it. I have set a pulse length of 600ms.
Is there any reason for this failure? Are there any other accessory decoders that will be more successful by providing more power? I have

seen several comments on the ADS-8, would these be a suitable alternative?

Hi Mousehole,
My experience with Peco motors is that they require a fair amount of power to operate. Even with a dedicated Peco CDU (R8247s were not available) I had

to allow a long (re-)charging time between throws. I was using the e-Link 1amp power supply at the time.

As for the ADS-8s, even with the 1amp supply, they work very well with almost no recharging time required.

[yerffoeg]

Can I continue this thread??

I have just made some basic mistakes:

Most of my track is Hornby, however some of my points are Peco and some of my point motors are Peco.

I have four of the E-Z, ESU decoders and two Hornby four way decoders.

What a mess!

First the Peco points need a lot more power to drive them.

I am trying a Peco point with Peco point motor using the E-Z decoder. The decoder has 22volt DC on the power bus.

It still will not operate reliably.

Looking at the pulse output of the E-Z it is only 20ms long!

This works fine for the Hornby points but is way too short to drive the Peco points.

I have to have this system ready for my Grandson on Christmas day.

21 points...

I only have Railmaster and select, 8213, controller. Is there a way to change the CV values of the E-Z units to increase the pulse width to say 100ms +?

I have alll the decoders programmed to the points in the range from 50 up to about 75 no problem. I just need to change the CV values of the E-Zs.

Lesson: Don't mix this stuff unless you know exactly what you are doing. I will still  have a bunch of Hornby point motors on Peco points which I think will never work.

[Chrissaf]

I only have RailMaster and Select, 8213, controller.

 

Since you can't officially connect a Select to a PC [no USB port] ... I assume you mean that you have both a RailMaster / eLink combination as well as a Select controller, given that the 'Select-a-Link' cable was cancelled and never released by Hornby.

 

I am trying a PECO point with PECO point motor using the E-Z decoder. The decoder has 22 volt DC on the power bus.

 

It is not totally clear from your post which Bachmann points decoder you are using, but from your description stating the 22 VDC input on a separate power input to the DCC power. Then this indicates that you are using the ESU SwitchPilot which, I believe, is also badged as a Bachmann E-Z product.

 

Now assuming that this deduction is correct, this decoder does not have an integrated CDU, so although you are using an external 22 volt Power Supply to operate the points, that power supply has to be a high current power supply. I suggest it needs to be a minimum of 4 amp rated for reliable point motor operation. What is the current rating of the 22 volt DC supply you are currently using?

 

ComMod RDS on this forum is using the SwitchPilot with a 19 VDC 4 (possibly 5) Amp ex Laptop Switch Mode PSU and reports on this forum that his point operation is reliable with this power configuration.

 

I would also draw your attention to Page 21 of the SwitchPilot manual, that suggests tweaking CV values when using PECO point motors.

 

If I read the manual correctly, CVs 3, 4, 5 & 6 [for the four ports] need to be set as a value 1 - see yellow highlight in extract below [Page heading & manual section 9.3].

 

I think Page 21 also covers setting pulse times [Manual section 9.2]. But again, if I have read the manual correctly, the PECO setting and the pulse time setting are mutually exclusive as they are set via the same CVs 3, 4, 5 & 6. So it is an either / or configuration choice.

 

Note that there are three different SwitchPilot manuals listed on the ESU website. My extract was taken from the 2008 SwitchPilot V1.0 Edition, you need to look at the manuals listed and decide which one matches your actual decoder and confirm the details I have provided as being still valid..... scroll down the page that opens on the clickable link above.

 

/media/tinymce_upload/c3d6ba56fe081d23cbfbb10288696b52.jpg

 

THIS IS A LONG REPLY ... PLEASE COMPLY WITH THE TIP BELOW. Using the 'Blue Button' will duplicate this reply unnecessarily. Not only that, but if you use the 'Blue Button' your reply will be held back for 'image approval' by admin because I have used an image in my reply.

 

TIP: As a relatively newbie poster on the forum, just be aware that the 'Blue Button with the White Arrow' is not a 'Reply to this post' button. If you want to reply to any of the posts, scroll down and write your reply in the reply text box at the bottom of the page and click the Green 'Reply' button.

 

Particularly as my reply includes an image, using the 'Blue Button' may result in your reply being held back for image approval, even though it is an existing image.

 

See also – further TIPs on how to get the best user experience from this forum. TIPs include 'How to post images' and 'How to make links clickable'.

https://www.hornby.com/uk-en/forum/tips-on-using-the-forum/

 

[96RAF]

@geoff

You say you have RM and Select. Select does not work with RM, only eLink or Elite.

Being able to write new CV values using RM is possible but can only be done external to RM using the Select IF it is at the latest firmware state. Any less than firmware v1.6 means it needs updating by Hornby to be able to write CVs.

[yerffoeg]

Thanks for the quick replies. - Everyone.

It has been a long day..

The answers to all your questions and assumptions is yes.

I have the interface to the track, I think it is an RM ...

I also have the little box R8213 - select.

I thought it might help writing the CVs.

So at the moment the ESUs are generating 22.5ms 2 amp pulses to the solenoids. Very short, not 65ms the min???

My 22v PSU is 1.1 amps. So not enough. I will check the rail when the points are triggered to see if it is dipping out and crashing the decoder. Should not affect the programming though?? My track psu is 15v at 7amps.

When I read the ESU I get daft results:

Tracka,b and power inputs connected together. feed from prog output of the rail interface, (need RM part no here.)

I read back something like port 1 address as 16234 and port 1 pulse length as 16nnn.. Really just junk.

I set them to 73, 74, 75, 76 and 100ms each. Hit write and it performs okay. Read back and get the same junk.

so no good.

I have looked onn the web for a programmer for the ESU but they are all £200+. I have just purchase sound versions of the flying banana, a deisel electric and two coach local train. £1000+. Not sure I will be allowed to spend any more.

I have tried terminating the DDC prog bus with 1000 ohms. The waveform has very poor edges but I would expect the decoders to handle these with  motors on the bus.

Also I noted when programming there is a lot of time with no activity on the bus. So no power to the points decoder. I assume you send a bunch of clocks to get the power up before sending data? Is there enough of these to get the decoder functional. Just a thought.

I design gate arrays for a living.

My wife tells me there are just 4 days to Christmas...

Detail,

I am running the latest RM software. I think the interface box is a bit early. Can I flash the firmware myself??

I did try to talk to the decoder with the select box but I really don't know what it is doing.

 

Geoffrey

[yerffoeg]

It is an elink encoder.

All I want to do is increase the pulse lengths to the points solenoids.

[Chrissaf]

I've already given you a solution which was to write the pulse length you want into CVs 3 to 6 as per Page 21 of the manual.

 

But to put it in a 'step by step' spoon-fed style.

 

1) Use the RailMaster / eLink. Not the Select. As Rob said earlier, if your Select was version 1.6 firmware, you could write direct to CVs 3 to 6. Using RailMaster as documented below, will write the required value to those four CVs for you.

 

2) Connect the ESU SwitchPilot DCC Power Input to the 'PROG A&B' output of the eLink. You must also connect your additional DC power supply [i.e your 22 VDC one] to the ESU as well to power it whilst it is being configured. This is actually stated as a note on the screen bottom in step 5 below.

 

3) Open the 'Accessory Decoder' configuration screen in RailMaster, using this icon from the RailMaster navigation bar:

/media/tinymce_upload/9d527e7375b8e4e25caa22f938007780.jpg

 

4) The following screen opens:

 

/media/tinymce_upload/6ee56c1a05d65283679aa4b41355cd34.jpg

 

5) Use the 'pull down box' to Select the "ESU SwitchPilot 4 Port decoder" this entry is listed alphabetically in the list, so it is above the Hornby R8247 that displays as the default so you need to scroll the list upwards. With the ESU decoder selected, the screen changes to this one.

 

/media/tinymce_upload/3a3a6fa2573f90905fbed3a22a8ce713.JPG

 

Note that the default pulse length that RM sets for this decoder is 520ms. Either leave at this value or make it any other ms value that you want. Note that the value entered in this screen MUST be a multiple of 65ms. The 520ms value is 8 times 65ms.

 

IMPORTANT NOTE: You have indicated that you have already set the DCC port Addresses for these decoders. You need to make sure that you enter the appropriate address for each decoder in the 'Output Port' number column, else you will over-write the previously set DCC addresses all with the value 1 if the 1 that is being displayed in the image above is not changed.

 

I suspect, that if you have been trying to write values that are not multiples of 65ms, then that is probably confusing / corrupting the configuration. You said in your last reply that you have tried to set them to 73, 74, 75, 76 and 100ms each. These values are not multiples of 65ms. So for 100ms you would write 130ms as being the nearest valid value [2 x 65ms].

 

6) Once happy with the settings, click the lower left Pen icon to write the settings to the ESU decoder. The green tick is used to read the decoder not write to it. After writing, use the red cross to close the screen.

 

NOTE: What I have documented above is the process recommended by Hornby using RailMaster. There have been many posts in the past where users have struggled to get any success using the documented method with the ESU SwitchPilot decoder, well at least to set an address. This is the first time I recall someone asking about setting pulse times. So in essence, you are the 'guinea pig' doing this. It may work, it may not. I still feel that replacing the power supply with a higher current version is more likely to deliver a working solution.

 

Additional Comments & Information.

 

You say that your 22 VDC power supply is 1.1 Amp rated. I am still of the strong opinion that this current rating is too low, as the ESU decoder does not have an integrated CDU in it.

 

You also said that you are using a 15 volt 7 amp power supply with your eLink. That is obviously not an official Hornby power supply and you potentially risk damaging the eLink or the Select if you are using the 7 Amp power supply with that instead.

 

I am running the latest RM software. I think the interface box is a bit early. Can I flash the firmware myself??

 

Be specific, which interface box are you referring to:

 

• The eLink [this is updated when a new version of RailMaster is installed, if an update is available. The current version is version 1.07].
• The Select [firmware updates are return to Hornby only. The current version is 1.6 Select Mark I or 2.0 Select Mark II].
• The ESU SwitchPilot [no idea, as not a Hornby product ... you would have to ask ESU].

 

I have tried terminating the DDC prog bus with 1000 ohms. The waveform has very poor edges

 

I'm not surprised, the correct termination to filter the DCC ringing [leading edge spikes] is a 100 Ohm resistor in series with a 0.1µF Ceramic capacitor rated at 50 volts or better. I recommend a resistor with a 2 Watt power rating as it will get very warm to the touch and this is normal.

 

I would also point out that the 'PROG' track is very low power level and putting any form of terminator on the PROG track is likely to make the PROG track ineffective. The RC filter documented above should only be used on the main DCC TRACK not the PROG TRACK.

 

Also I noted when programming there is a lot of time with no activity on the bus. So no power to the points decoder. I assume you send a bunch of clocks to get the power up before sending data?

 

This is completely normal on the 'PROG' track. No power is sent to the 'PROG' track when it is idle. And when power is sent, the power is current limited. Some decoders [Hornby R8216 for example] need to be 'pre-charged' before they are connected to the 'PROG' track. The Hornby R8247 has internal circuits that support the low power programming scenario. This issue can also affect locomotive decoders that are fitted with large value 'stay alive' capacitors. It sometimes becomes necessary to disconnect the 'stay alive' during programming.

 

This low 'PROG' power is a designed in feature, so that any decoder installations that are miswired or faulty cannot damage anything. The theory being, that any issues will be found and rectified at the programming stage before real damage is done by putting the decoder on the full powered DCC supply.

 

This reduced power in terms of both level and duration is why you should not put any other form of termination on the 'PROG' track as mentioned in the previous comment above.

 

This is also why RM recommends connecting DC power to the ESU SwitchPilot during configuration.

[yerffoeg]

I have been trying all the above with no success in changing the CV time constants. So at present the Railmaster scheme does not work for me.

Detail: Railmaster version 1.73 and elink version 1.07. Not sure what the elite is but is it worth obtaining one to replace the elink?

When you input 100ms into the box Railmaster amends it to 195ms so that is good.

When you input address 73 into the first box you get 74, 75 and 76 in the other boxes.

With  the separate 22v rail attached to the ESU and the elink prog connected to the track inputs there is no response from the ESU as per pop up on the Railamaster. It appears to respond when both the power and track are connected together on the ESU and then connected to the prog on the elink. But going back to my power generation from startup clock pulses, I guess there is not enough power getting to the ESU during programming and read back.

I could work on this by pulling the data lines to slightly higher levels but not yet.

I need a programming solution for the CV3, 4, 5 and 6. At the moment railmaster is not doing that. Not your problem but mine.

I just thought I could use the select at a lower level to address a CV and pretend the decoder was a loco?

My select is stamped 31/15 It displays 15 30 03 at start up.

I will next try reducing the 22 volts while programming to say about 15volts.

[Chrissaf]

For the third time. I suggest you need a DC Power supply rated at 4 amps not 1 amp.

 

Not sure what the Elite is but is it worth obtaining one to replace the eLink?

 

No ... that will not make any difference whatsoever when used with RailMaster. The RM interface is still the same. The Elite can however be used 'stand alone' to configure CVs, but a brand new Elite is £200 plus. You can update your Select for less than £20 to do the same CV writing tasks ... see below ... so why spend more than 10 times more than you need to.

 

My select is stamped 31/15 It displays 15 30 03 at start up.

 

Your Select has got version 1.5 Firmware. You need version 1.6 Firmware in order to write to CVs 3, 4, 5 & 6.

 

The Select is a 'Return to Hornby' firmware update process. The firmware update files are not downloadable in the public domain. Not only that, but special programmer hardware is used in the Hornby workshop to attach directly to the Select circuit board.

 

This latest 1.6 version completely transforms the user experience of the Select controller and adds the ability to write (not read) all CVs up to CV255. As well as access to the full F0 - F28 range of functions. The Select also sends more power to the programming output than both the eLink and the Elite because it doesn't have a separate 'PROG' output.

 

With the Select, you use the 'TRACK' output for programming. This means that you disconnect the layout track from the Select and reconnect it to whatever it is you are programming when performing a programming task.

 

The upgrade charge is typically £15 plus your sending postage costs. You need to get a 'returns number' from Hornby Customer Services - see 'Contact Us' link at the very bottom of the page.

 

If you do arrange to get your Select firmware upgraded, then you will need both the Select version 1.5 full manual and the version 1.6 addendum manual. As the operation of your Select (how to use it) will be completely different in certain areas of operation compared to your current firmware version.

 

You can download these two manual PDFs from here:

https://www.hornby.com/uk-en/downloads/view/index/cat/12/

 

If nothing appears to happen, check your nominated browser download folder for the files as they can download silently in the background.

[Chrissaf]

This is a long shot, but you can try this. I used something very similar to this process to write 8 to CV8 to reset an R8247. I see no reason why it might not work for your ESU SwitchPilot too. I have basically taken my CV8 process and re-written it for your CV3 to 6 requirement.

 

It is important that you follow the procedure exactly as written and do not deviate from it.

 

1. Remove the ESU SwitchPilot from your controller DCC track output connection.
2. Connect your ESU SwitchPilot to the Controller 'Programming' PROG output.
3. In RailMaster click on the 'Steam Train' icon on the top row between the 'Cog' and 'Stop Watch' icons. This is the "Set up Locomotives" option. Yes I know that your SwitchPilot is NOT a loco. But this is a procedure to follow that has worked for me writing CVs to an Accessory Decoder.
4. The "Locomotive" window will open. Ignore everything on this window except the " i " icon on the bottom left hand side row. Clicking this" i " icon opens the "Locomotive Settings" window.
5. In the "Locomotive Settings" window column 1 is 'CV' number, column 2 is 'Description' and column 3 is 'Value'. DO NOT TRY TO READ THE CV VALUES with the green tick. Place your cursor in the 'Value' column for CV3, and type the number 3. Also enter a value of 3 in the value box for CV4, 5 & 6
6. Now click the 'Pencil' icon (Write CVs to Loco) located on the bottom far left side of the "Locomotive CV Settings" window.
7. Wait for the write process to complete, this may take about 10 - 20 seconds.
8. Click the Red X to Exit and close window on the "Locomotive CV Settings" screen
9. Click the Red X to Exit and close window on the "Locomotive" screen.

 

The value 3 for CV3, 4, 5 & 6 should equate to a pulse length of 195ms.

 

DO NOT try to read the CV values back again. Now test the operation of the SwitchPilot with it reconnected back onto the main track layout.

 

If this doesn't work, then I have exhausted all possible suggestions to offer you and I strongly suggest you heed my suggestion about uprating your 1 amp supply to a 4 amp one.

[yerffoeg]

So current conclusion:

Hornby point motor do not operate the Peco points. - too much load.

Peco point motors and Peco points do not work with the Hornby point decoder - not enough drive. I could attach a picture but it looks like they are getting removed??

As for the ESU that could probably do the job. It needs a working programming scheme which Hornby have not got.-  Not their problem.

In detail: The ESU will only respond to RM messages if both the power and data lines are connected together and connected to the prog lines of the elink. The ESU appears to be quite a load as the signals are quite low in amplitude. See attached, not.

RM has some sort of unexplained issue when programming a decoder: You put the decoder address in and then ask it to read the decoder back. At this point it says to you want to abandon changes?? (So what address is it talking to?) However it does get a replies back from each port. see attched, not. It also writes to the decoder and the decoder gives a short response. see attached, not. All the replies are constant 6629 16640ms, 6630, 1640ms, 6631, 1640ms, 6632, 1640ms. so nothing changes regardless of what is supposed to be programmed. The pulse length to the solenoids is still 40ms - not a defined value.

RM will only accept a minimum pulse length in ESU box of 195ms?? max is 2080ms.

So no solution.

I don't like being moderated either.

Geoff Frost

[Chrissaf]

I could attach a picture but it looks like they are getting removed??

 

They are not being removed, they will appear in due course ... see below

 

I don't like being moderated either.

 

Posts can not appear for two main reasons. You have included an image in the post. This automatically triggers the post to be passed to a queue for a Hornby admin staff member to approve. ComMods can not approve an image. If we could, then the process would be speeded up considerably. You are not being singled out, this is the same for everyone with relatively low to medium post counts.

 

The second reason that a post might not appear straight away [one that does not contain an image] is that a typing error has introduced a banned word, and the post has been captured by the automatic 'profanity filter'. For example, many people miss spell by accident the word Hornby and leave out the b. This is a common occurrence.

 

PLEASE DO TRY my suggested alternative in my last reply, it might work for you. You won't know if you don't try it.

 

[RDS]

... The ESU will only respond to RM messages if both the power and data lines are connected together  ...

 

Not true.  I have 7 ESU devices on my Layout, Powered from a separate 18v ex Laptop Power Supply.

I only operate the Points using RailMaster and it works perfectly, every time.

[96RAF]

I thought it was clearly recommended in an earlier reply that the ESU decoder needed an external supply to function, hence likely why it won’t work without slaving the power and data lines together in lieu when programming.

[Flashbang]

I thought it was clearly recommended in an earlier reply that the ESU decoder needed an external supply to function, hence likely why it won’t work without slaving the power and data lines together in lieu when programming.

Just what is all that meant to mean?  "Slaving - Power and data lines together in lieu when programming"?   Please keep replies simple to understand. Many on here have basic train sets and a DCC console and accessory decoder(s) for accessory operation.  

 

I feel you may mean ....Use an external suitable power supply for the ESU Accessory decoders power?  But  its very unclear!

 

From memory ...  The ESU accessory decoder offers the options of,,, A) Using the DCC power to operate its connected accessories and receive the DCC data. Or B) Use an external power supply to operate the accessories and just use the DCC data to operate the decoder outputs as required.  

[RDS]

From memory ...  The ESU accessory decoder offers the options of,,, A) Using the DCC power to operate its connected accessories and receive the DCC data. Or B) Use an external power supply to operate the accessories and just use the DCC data to operate the decoder outputs as required.  

Brian, that is correct but B) is maybe better explained as: Use an external supply to power the Decoder and just use the DCC data to operate the Decoder Outputs as required.

[96RAF]

I thought it was clearly recommended in an earlier reply that the ESU decoder needed an external supply to function, hence likely why it won’t work without slaving the power and data lines together in lieu when programming.

Just what is all that meant to mean?  "Slaving - Power and data lines together in lieu when programming"?   Please keep replies simple to understand. Many on here have basic train sets and a DCC console and accessory decoder(s) for accessory operation.  

 

I feel you may mean ....Use an external suitable power supply for the ESU Accessory decoders power?  But  its very unclear!

 

From memory ...  The ESU accessory decoder offers the options of,,, A) Using the DCC power to operate its connected accessories and receive the DCC data. Or B) Use an external power supply to operate the accessories and just use the DCC data to operate the decoder outputs as required.  

 

 

This is what he said

The ESU will only respond to RM messages if both the power and data lines are connected together and connected to the prog lines of the elink.

 

I used his terminology, except I said slaved instead of connected. The back ref was to using separate power supply and keeping the DCC for data flow, i.e. your B option.

[RDS]

 

The ESU will only respond to RM messages if both the power and data lines are connected together and connected to the prog lines of the elink.

 

 

 

 

Sorry but that is not correct.  I have been using ESU Decoders for 7 years now. 

The Power and Data lines do NOT need to be connected together. (unless of course the eLink operates differently to my Elite)

I have, in the last 12 months added 2 more ESU units to my layout and I program them away from my layout using my separate Power Supply but taking the Data from the Elite.  Once programmed with their addresses, I fix and wire them up in position under my layout.

[96RAF]

 

The ESU will only respond to RM messages if both the power and data lines are connected together and connected to the prog lines of the elink.

 

Sorry but that is not correct.  I have been using ESU Decoders for 7 years now. 

The Power and Data lines do NOT need to be connected together. (unless of course the eLink operates differently to my Elite)

I have, in the last 12 months added 2 more ESU units to my layout and I program them away from my layout using my separate Power Supply but taking the Data from the Elite.  Once programmed with their addresses, I fix and wire them up in position under my layout.

 

Yes I know, which is why I quoted his own words and referred back to the separate power supply and data (Brians Method B).

I was surprised that he got it work by connecting wires together like that as I presumed if using DCC only, that connection would be internal to the unit.

Regardless of that - we know you use these things extensively and you have them working correctly so your explanation is definitely solid and should be followed.

[yerffoeg]

Okay. Problem sorted:

I purchased an LOK programmer to find out what is going on. I ran a bunch of tests using different CV 3 values, and power supply voltages. Here are the results.

1 You need plenty of supply current - we already figured/new that.

2. The decoder integrates the current in the points solenoid with time. I.e. the greater the current the shorter it fires the solenoide for. (Must be measuring the current somewhere)

3 With a fixed solenoid value varying the supply voltage alters the solenoid current. Sort of expected. However the pulse width changes as well.

4. With the switch in the mid position CV3 >2 does nothing to control the pulse width it is all down to the current flowing in the solenoid.

These were my measured results:

12v  - 1.8amps for 258MS

13V - 2.1amps for 132ms

14v - 2.2amps for 90ms

15v - 2.5amps for 70ms

16v - 2.62amps for 52ms

17v - 2.8amps for 43ms

18v - 3.0amps for 40ms

19v - 3.16amps for 35ms

20v - 3.28amps for 31ms

The energy input to the coil is around 2 for high voltages and creeps up to 5.57w/s at 12v

However to give the armature a good kick a higher voltage would be more appropriate.

So the down side is that trying to get a longer pulse out of the unit at say 19v is not very practical at the moment.

The solenoid used for testing was an under board PECO part.

Running any more than 20v could well damage the decoder darlington output stages. 24vdc is certainly destructive.

However it is an interesting alternative to CDU?? Not sure I like it much.

[96RAF]

I have a LokProgram er and didn’t know it could be used for all that. Pray tell how.

[yerffoeg]

I have a LokProgram er and didn’t know it could be used for all that. Pray tell how.

Measurements made with a Tek scope and dc current probe plus Tek bench powersupply.

The esu works fine with the split power rails but does need the power and track inputs joined together to be programmed.