Aaarrghhh - ECoS2

[Philou]

Chums,

 

Apologies if this has already been started as a topic - if it has, could you direct me there?

 

Right, I have 'issues' with my ECoS2. I did a software update in mid-January to the latest version 4_2_2 and followed the on-screen instructions (no switching off before the message etc.).

 

As I had no need to use it at that moment, away it went until yesterday when I took it to the club to run a Class37 that one of the yoof wanted to listen to the sound. It wasn't even powered up as I forgot the track lead connector. We did a work around and listened to the loco via a Roco Multimaus instead.

 

This morning I chipped up my latest Class 66 'Evening Star' and wanted to listen to the sound and address it correctly. Plugged up the ECoS, waited for the intro screen and found that it simply would not respond to the stylus nor my phat phingers. It did struggle to respond to the paddle and button and just about got to the throttle screen but no more. Switched off and switched on a few times (as you do) and lo! a response to the stylus and my fingers - except - when you touched the screen it responded a few centimetres away (for example on the keyboard strike a key on the top row and a key on the middle row to the right would operate) - or not at all depending on its mood.

 

So off and back on again - I was able to access the screens using the stylus - sporadically - and fingers. However, all to nought as it refused to acknowledge the presence of the Class66 - even using Railcom - nothing - nada. 'Aha' I thought. I exchanged the 66 for the Class 37 as I wanted to give a four-figure address. It knew it was there but refused to read the chip - yet the loco worked perfectly the day before.

 

I have sent a message to the ESU site forum - but do you have any ideas? I don't want them saying 'It's bust' and I have send it back as it was fine before the update and I don't have a layout - just a yard of track to programme locos - rest of the time the unit is packed away.

 

Cheers,

 

Philip

[GoingUnderground]

Can't remember anyone else reporting this problem either on here, or on ESU's own forum.

[RAF96]

Just a thought as I don’t have experience of an ECoS, but could you try to reinstall the update, maybe the first one was a tad flaky.

 

Edit: or is there a unit reset?

Rob

Edited March 12, 2018 by RAFHAAA96

[Philou]

@ GoingUnderground @ Rob

 

Hi - there was one other person on the ESU forum that had issues after the latest update - ESU's reply was more or less 'It's bust - send it back' which is what I'm trying to avoid - if I can. There is a reset - factory reset - which takes you all the way back to version 3_7_0. I would only lose loco addresses as I have no layout. I just use the ECoS to test and address chipped locos on a 1 yd piece of track - it's just the pain of uploading all the updates again.

 

@ Rob

 

I see you're involved on the H****y forum too. I'm Froggie there .

[GoingUnderground]

The factory reset has absolutely no effect on the firmware. All it does is delete the user created data, locos, accessories, routes, track plan, and loco icons. It's the same as doing a factory reset on a digital TV which deletes all the tuning and channel data.

 

In fact once you have installed the new rescue system for firmeare 4.x.x you cannot go back down to firmware 3.7.0 or earlier even if you wanted to.

 

To reinstall firmware you boot the ECoS in Rescue Mode and reinstall the firmware from there. The reinstall process is identical to the normal firmware upgrade process, which means that you will must have the ECoS connected to and communicating with your computer. Reinstalling the firmware normally leaves the user created data unchanged.

 

Each edition of the firmware is a full copy of the firmware containing all the updates up to the release of that version of the firmware. So there is no need to install firmware in release date order.

Edited March 12, 2018 by GoingUnderground

[Phil Bullock]

The factory reset button has absolutely no effect on the firmware. All it does is delete the user created data, locos, accessories, routes, track plan, and loco icons.

 

In fact once you have installed the new rescue system for firmeare 4.x.x you cannot go back down to firmware 3.7.0 or earlier even if you wanted to.

 

To reinstall firmware you boot the ECoS in Rescue Mode and reinstall the firmware from there. The reinstall process is identical to the normal firmware upgrade process, which means that you will must have the ECoS connected to and communicating with your computer. Reinstalling the firmware normally leaves the user created data unchanged.

 

But always create a back up volume first...

 

Phil

[GoingUnderground]

@ GoingUnderground @ Rob

 

Hi - there was one other person on the ESU forum that had issues after the latest update - ESU's reply was more or less 'It's bust - send it back' which is what I'm trying to avoid - if I can. There is a reset - factory reset - which takes you all the way back to version 3_7_0. I would only lose loco addresses as I have no layout. I just use the ECoS to test and address chipped locos on a 1 yd piece of track - it's just the pain of uploading all the updates again.

 

@ Rob

 

I see you're involved on the H****y forum too. I'm Froggie there .

IIRC, that person had different issues to you. If there was a major problem with 4.2.2 the ESU fourm would be overrun with posts about it.

 

As I've replied already on the ESU forum, it might be problems with condensation. My concern is that condensation, which you indicated might be a possibility. may have electrically cross connected components and damaged them.

 

Laptops, tablets and smartphones have made us all think that we can move electronics from cold to warm environments and turn them on without first letting them warm up to the temperature of their new surroundings. If you wear glasses it brings home how often condensation happens when moving from a cold to a warm space.

[Philou]

@ Keith

 

I saw afterwards that once you updated, there was no going back. The unit is here next to me downstairs in the warm. I hope there is no damage as otherwise it could get expensive . I shall give it a spin tomorrow and let you know how I get on - both here and on the ESU forum.

 

Thanks for the suggestion.

 

Cheers,

 

Philip

 

Edit: PS As I haven't the manual close by, could you remind me of the keys to press for Rescue Mode, should I need it. Thanks.

Edited March 12, 2018 by Philou

[GoingUnderground]

Do what I do, download the manual from the Downloads part of the ESU website. It's in PDF format, so you should be able to read it on a tablet, or smartphone, or your computet.

[Philou]

@Keith

 

Thanks - I had completely forgotten about it being available on the ESU site.

 

Cheers,

 

Philip

[Philou]

Hello Keith,

I did as you suggested and had the ECoS in a warm room since this last night and fired it up again today. It was sluggish regarding the tactile screen so I switched on and off a couple of times, went away and came back a while later and it was responding to both the stylus and my fingers. I took the two locos that I used the day before - Class66 with a new sound chip and Class37 with sound chip but needed a new address. I placed the locos (one at a time) on the programming track and despite the apparent difficulties I had my end on-screen the day before, it remembered both their new addresses that I had inputted that day. I was able to 'read' the chips - both indicating Loksound v4.0 and 927 CVs.

However, when testing for the various functions - absolutely nothing - no lights, no sounds. No response from the throttle from either.

Is there something basic that I missed in setting them up, or is there a greater underlying problem? (I hope not). I seem to remember sometime ago I had what might have been a similar problem with one but I cannot recall now how it was resolved - or if it 'went away'.

I have not yet done a rescue start as I wanted to explore matters a bit further first.

BTW, Railcom IS enabled at the moment and it is default D28 in speed steps - I know that some chips, especially Hornby TTS ones, have to be D128.

Cheers,

Philip

 

PS: You will find an identical message on the ESU forum.

[55020]

If you have a responsive ECoS, I would try a reload of the latest firmware (as Rob suggest earlier in the thread).  It's a "no jeopardy" option that might just make a difference.

 

 

 

Steve

[Philou]

@55020

 

Steve,

 

Thanks for the advice - I suppose it would be better. Stupidly I get wound up about it due to the need of ensuring a good link at all times during the upload. Usually nothing happens and it all goes OK. As Underground (Keith) was explaining to me (if I understood correctly), each new update has the previous versions attached anyway.

 

I'll give it a go tomorrow evening and if you hear a scream from about 1000miles to the south of you - it'll be me .

 

Cheers,

 

Philip

[55020]

Yeah, the updates aren't incremental and contain a complete "load" of the firmware.  So a reload might just sort out your problems. 

 

Let us all know how you get on, assuming we don't hear your scream first    

[GoingUnderground]

Make sure that the "Programming track currentless" box in the Configuration - Setup 2 menu is not ticked before you try reinstalling the firmware. If it is, "untick" it and try your locos again. As you can read CVs then the programming track outputs are working in programming mode, and I'm guessing that you're leaving Programming mode but still using the programming track outputs when trying the motor and functions. If that is what you are doing, then configuring the programming tracks to be unpowered when not programming would account for the lack of response by locos.

Edited March 14, 2018 by GoingUnderground

[Philou]

@ Keith,

 

Thanks for heads up - I didn't know that. I'll check that box, but I shall do a reinstall of the latest update anyway as it's a little unresponsive at start up. Once done I'll let you know here and on the ESU forum how I get on.

 

Cheers,

 

Philip

[Philou]

Hello chums,

 

'Tis I again, reporting back as promised - and I didn't scream - not much anyway.

 

A bit of a curate's egg this. I will try and cut down on the waffle, but I need to give some detail for analysis as I don't really know what happened.

 

I did as suggested, once it had fired up, and checked that the box was not ticked 'Programming track currentless'. It was not, though I did tick it thinking that just maybe the command had stuck. I couldn't untick it despite jabbing it with my stylus and finger. I had to come out of the screen and go back in and then it unticked. Tried my Class37 that was known to work but nothing. I changed the track output lead from programming track to main but it made no difference and I decided to bite the bullet.

 

I went into the ESU site and called up the manual and found the page to do the Rescue Start. I also called up the official ESU You Tube video so I could have the update tutorial visually in parallel.

 

The Rescue Start came up advising that I was in 4.2.2rescue mode. Before returning to the video I happened to see in the top corner an IP number which was different to the 'standard' IP for my ECoS here - luckily I kept a note of it as nowhere does it mention that this is the IP you have to use. Also, in reading the manual something else caught my eye that is not mentioned in the video is that you have to take the batteries out of the ECoS - something that I had not done the previous times (except possibly the very first time I upgraded). I do not know whether it is a factor or not - but I took them out anyway.

 

I reloaded the software and let it do its thing and it rebooted. I turned the ECoS off and re-installed the batteries and booted it up again. It may be my imagination but it seemed to boot a little more quickly and the screen seemed more responsive to the stylus and my fingers - things were looking good. I put the Class37 back on the rails under Programming Track, the 'currentless' box was still unticked and set about reading the decoder - nothing, nada. It just would not respond and I deleted the loco thinking that the ECoS would pick it up as a 'new loco found'. Well it wasn't going to in a month of Sundays as I hadn't plugged the track cable in the back - I add this simply to show how even the most obvious things can be overlooked.

 

Having checked the wiring and replaced the loco back on the tracks - the decoder was successfully read. I gave it it's new address as I had deleted it previously and it still wouldn't respond. I was about to stop everything at that point when it suddenly announced 'found new loco' and here you are, a new address for you (1002). I didn't want that number and typed in the four figure number I wanted. The ECoS advised me that number was taken already. I hit the 'overwrite' button, the 'Stop' button lit up and it closed down and rebooted without prompting - needless to say I was confused.

 

I went back in to the loco list and called up the Class37 under its ealier address and deleted it. I then called it up under number 1002, did all the function tests - which worked! I then renumbered it to what I wanted, checked the functions again and being OK turned to the Class 66.

 

Again, I had no problem in reading the chip. So I gave it this time the default 3 address to test the functions - again nothing, absolutely nothing.  After a lot of faffing I deleted all instances of that 66 from the list and it still nothing - not even anything from Railcom. I was about to give up when it flickered 'Class66' on the throttle screen - and it was working! The address was one that I had given yesterday and overlooked in my deletions (I have another 66 and may have left it thinking it was that one). Again re-addressed it to what I wanted, checked the functions were still working and packed it away.

 

So, I have two locos addressed and working - but I have absolutely no way of knowing how I really got there.

 

Questions arise:

 

Was the update 4.2.2 corrupted when it transferred from my laptop to the unit originally, and did the removal of the batteries make a difference?

 

Why did the ECoS shut down when asked to overwrite the loco number?

 

Was the autodetect of 'new locos' sluggish? It certainly was not as instant as it used to be. Did it only work with the Class37 because all instances of previous addresses had been deleted?

 

Certainly the reloading of the update did help as the screen became very responsive - so thank you all for that suggestion and thank you for your advice and help.

 

I am a cautiously happier bunny tonight.

 

Cheers,

 

Philip

 

Edit: @ Keith, I have redacted this version and pasted it onto the ESU site. Thanks again for your advice.

Edited March 15, 2018 by Philou

[GoingUnderground]

Hello chums,

 

'Tis I again, reporting back as promised - and I didn't scream - not much anyway.

 

A bit of a curate's egg this. I will try and cut down on the waffle, but I need to give some detail for analysis as I don't really know what happened.

 

I did as suggested, once it had fired up, and checked that the box was not ticked 'Programming track currentless'. It was not, though I did tick it thinking that just maybe the command had stuck. I couldn't untick it despite jabbing it with my stylus and finger. I had to come out of the screen and go back in and then it unticked. Tried my Class37 that was known to work but nothing. I changed the track output lead from programming track to main but it made no difference and I decided to bite the bullet.

 

I went into the ESU site and called up the manual and found the page to do the Rescue Start. I also called up the official ESU You Tube video so I could have the update tutorial visually in parallel.

 

The Rescue Start came up advising that I was in 4.2.2rescue mode. Before returning to the video I happened to see in the top corner an IP number which was different to the 'standard' IP for my ECoS here - luckily I kept a note of it as nowhere does it mention that this is the IP you have to use. Also, in reading the manual something else caught my eye that is not mentioned in the video is that you have to take the batteries out of the ECoS - something that I had not done the previous times (except possibly the very first time I upgraded). I do not know whether it is a factor or not - but I took them out anyway.

 

I reloaded the software and let it do its thing and it rebooted. I turned the ECoS off and re-installed the batteries and booted it up again. It may be my imagination but it seemed to boot a little more quickly and the screen seemed more responsive to the stylus and my fingers - things were looking good. I put the Class37 back on the rails under Programming Track, the 'currentless' box was still unticked and set about reading the decoder - nothing, nada. It just would not respond and I deleted the loco thinking that the ECoS would pick it up as a 'new loco found'. Well it wasn't going to in a month of Sundays as I hadn't plugged the track cable in the back - I add this simply to show how even the most obvious things can be overlooked.

 

Having checked the wiring and replaced the loco back on the tracks - the decoder was successfully read. I gave it it's new address as I had deleted it previously and it still wouldn't respond. I was about to stop everything at that point when it suddenly announced 'found new loco' and here you are, a new address for you (1002). I didn't want that number and typed in the four figure number I wanted. The ECoS advised me that number was taken already. I hit the 'overwrite' button, the 'Stop' button lit up and it closed down and rebooted without prompting - needless to say I was confused.

 

I went back in to the loco list and called up the Class37 under its ealier address and deleted it. I then called it up under number 1002, did all the function tests - which worked! I then renumbered it to what I wanted, checked the functions again and being OK turned to the Class 66.

 

Again, I had no problem in reading the chip. So I gave it this time the default 3 address to test the functions - again nothing, absolutely nothing.  After a lot of faffing I deleted all instances of that 66 from the list and it still nothing - not even anything from Railcom. I was about to give up when it flickered 'Class66' on the throttle screen - and it was working! The address was one that I had given yesterday and overlooked in my deletions (I have another 66 and may have left it thinking it was that one). Again re-addressed it to what I wanted, checked the functions were still working and packed it away.

 

So, I have two locos addressed and working - but I have absolutely no way of knowing how I really got there.

 

Questions arise:

 

Was the update 4.2.2 corrupted when it transferred from my laptop to the unit originally, and did the removal of the batteries make a difference?

 

Why did the ECoS shut down when asked to overwrite the loco number?

 

Was the autodetect of 'new locos' sluggish? It certainly was not as instant as it used to be. Did it only work with the Class37 because all instances of previous addresses had been deleted?

 

Certainly the reloading of the update did help as the screen became very responsive - so thank you all for that suggestion and thank you for your advice and help.

 

I am a cautiously happier bunny tonight.

 

Cheers,

 

Philip

 

Edit: @ Keith, I have redacted this version and pasted it onto the ESU site. Thanks again for your advice.

ESU do say in the manual to remove the batteries. However, in the years that I have owned ECoSs I have often forgotten to do it and haven't had any problems leaving them in so far........  But ESU are the experts, and there must be a reason behind the advice. After an update with the batteries removed, you can leave them out, permanently, but you must always use the Stop button to shut down your ECoS at the end of a running or programming session. If you do keep the batteries out, you could lose all use configuration data if there is a power cut between the ECoS starting and finishing its shutdown procedure, However as that only takes a few seconds, it is a very remote risk, unless you live in an area that regularly suffers from power cuts.

 

The Rescue Mode shows you the IP address that the ECoS has been given when it started. That is a sensible precaution as if the firmware has become corrupted, that corruption could have affected the way that the ECoS has stored its IP address. If the Rescue screen didn't show you the current address, you would have a very, very, long guessing game trying to find the address it was using.

 

It does take the ECoS a little time to recognise locos using Railcom Plus.

 

You mention the overwrite button, I take it that you mean the write data to decoder button, the one that looks like a piece of track with an arrow pointing towards it, but that's just semantics on my part. The important thing to note is that after using the write data button, you must also use the green tick button to save the new address to the ECoS's own internal database. 

 

Also if you have RailCom Plus activated, the ECoS will not let you program a decoder with an address that it already has in its database. I think that is why it stopped.  You can have decoders with the same address, but only if they do not have RailCom Plus, or if one or both of the decoders that you want to share an address have Railcom Plus, you must have turned off RailCom Plus in the Plus capable decoder(s), which for ESU V4 decoders is done using CV28. It works, as I do this myself.

 

Remember, RailCom Plus is extra functionality that uses RailCom. You can still have Railcom turned on in a decoder, as that is controlled via CV29, even if you have turned off Plus. Once my ECoS has registered a new loco with a Plus decoder, I always turn off Plus in the decoder as there is little use for Plus thereafter as my DCC locos never leave home. The only benefit to keeping Plus turned on in a Plus capable decoder is if you take your locos to other layouts, when Plus will prevent your visiting loco having the same address as another loco on the visited layout, always assuming that the DCC controller on the layout being visited uses Plus.

 

I understood that the update process does include a check that the ". bci" firmware file hasn't been corrupted during its journey from ESU to your ECoS.  There certainly used to be such a check as the download instructions on the Firmware download page refers to a "Wrong CRC" error message which implies some sort of file integrity check. The ESU's FAQs have more information on the reasons for this error message appearing. So I think file corruption is unlikely. If it was corrupted, it would have been during the original update process. But as I think you said that you turned off your ECoS immediately after doing the 4.2.2 update the first time we don't know if it was installed correctly.

 

I still think that your problems were caused by cold and condensation and not file corruption, but I may be wrong, I often am, according to SWMBO.

 

How you post your update on the ESU forum is up to you, it's your experience, and you need to retell it in your own way.

 

But I'm very pleased for your sake that all is working again, and thank you for providing us all with a timely reminder to let electronics warm up when moving them from a very cold environment to a warm, humid, one before turning them on.

Edited March 15, 2018 by GoingUnderground

[Philou]

@ Keith

 

It may have well been have been the cold and then the warm humid room. The first time of trying, I remember the screen was becoming more responsive as time went on - but there we are - hey ho.

 

You wrong sometimes? I thought that honour was reserved for me .

 

Thanks anyway for your help,

 

Philip

[Philou]

@ Keith

I saw your reply on the ESU forum and I thought I ought to try and clarify the 'overwrite?' request.

 

This was at the point that the ECoS 'found' my Class 37. A sub-menu opened up (about the size of the 'Welcome' screen). Within it it gave the address of the loco that I given previously - in this instance '3742' and details of the chip provider - in this instance 'SWD Class 37' (the detail may not be precisely the same but it is the gist) plus a key board that had an address 1002 already in the yellow screen.

[This is an automated sequence for addressing locos. It could not give me 1000 or 1001 as these numbers I have already used on another pair of locos. The automated addressing can be controlled from another window which, I think, is under the Set up screen 1 with the loco symbol and 1 - 9999 under it. (I may not have the right location). I had told the system ages ago that the addresses should start at 1000.]

The sub-menu gives you the option to accept the address given - in my case 1002 - which I did not want. I wanted 3742 that I re-typed in - the sub-menu advised me that the number 3742 was already allocated (partially hidden under the green tick box). It gave me the option of overwriting - I assumed to overwrite 3742 onto the allocated loco. I tapped the 'overwrite' button and it was at that moment the ECoS 'Stop' button lit up (as if there had been a short) and it shut down and re-booted without any prompting from me - that is when I was confused - to say the least! To be absolutely certain, it was NOT the write button that looks like a piece of track.

 

I got there in the end, thanks to you and all the others and that for me was the most important.

 

Cheers,

 

Philip

 

I have posted this in a similar form of the ESU site..

[GoingUnderground]

Apologies for my misunderstanding. I did say that I make mistskes at times.