Problem with updating Elite

[pidder]

Chrissaf  Thanks for your quick response. Just a thought. You probably know I have Windows 10 so that idea is a non starter. Back to the drawing board.

[96RAF]

Correct Chris, they are best attempted on an XP machine and you may have to remove boot-loader program first as early installers used several variants of bootloader - your pc will tell you if its not the right one by failing to install. I had to revert to my XP laptop earlier this week to back install early revisions to see how far back a problem went - since fixed, but historical.

 

It just goes to show how much Hornby has improved the installer that it will install the update on any machine now with total confidence. In fact if you think about how easy operating a PC is nowadays compared to even 10 years ago much less 20-30 years. E.g frustrating tape recorder loading of programs onto a Spectrum in the 80's.

 

You have nothing to lose pidder by trying and you may get it back on its feet and working for some aspects. E.g my one updated fine even with a duff programming output and HBM has one that has blown the main hi-amp track circuit but still works as an accessory power source and programmer.

 

Something else that was a problem long ago was flaky USB connectors due mostly to heavy handling (not limited to Elites, but also laptops and PCs) these seem to be more,reliable these days. Another possible update or RM comms failure route, that could be checked.

 

I could go on but any such list leans toward endless.

Rob

[Fishmanoz]

Electronic equipment failure rates follow what is known as a bathtub curve.  From new, there tends to be an initial higher than usual rate, then it drops off and maintains a low value until components start to wear out with age and the rate goes up again.   And these days, components tend to be more reliable so the other end of the bathtub is a long way off,  although accelerated failure rate may occur with poorly designed hardware working close to or just exceeding its specification ratings.  And from Rob's experience, this may be the case for the output components to track and programming track.

 

Pidder, chances of yours starting to work again at an earlier firmware state?  Very low. Something has failed be it in the firmware storage chain or elsewhere and it will stay failed. It's not as if a higher firmware version works something harder and it can't take the extra strain but will be ok at previous state. Not saying don't try but chances of success are low.

[Yelrow]

As it was ok, prior to update, which was not a necessity, lessons should be learnt. If it aint broke, etc. john 

[RB51]

Isn't the point that everyone is making is that it might not have been the update process that caused the failure? A multitude of components, the age of the machine etc. all have a bearing. Nothing lasts forever. R-

[Yelrow]

The point everyone WAS making, originally, was that the update was 100% safe, and we could go ahead in that knowledge. Now one has gone wrong, suddenly, its the Elite age/ components that were at fault. This was never brought up when everybody was telling me to update.   Closing ranks, springs to mind. john

[RB51]

Regardless of what springs to your mind John, we do not know that the update process is not 100% safe. There are too many other variables. R-

[Fishmanoz]

What springs to my mind is the update process has proven to be 100% reliable in the latest version.

[Yelrow]

Only, it has not, as poster has proved. The fact that his controller was fine before , and not now, destroys your 100% claim. Instead of maintaining all is well, we should be looking at at what age/ batch  of Elites, will not be 100% certain  of updating, without damage, and leaving their owner with a broken machine. As for other variables., this is even more reason why posters need to know the suspect batch numbers, as  they can then, avoid the risk. and continue to use them on older firmware. john

[RB51]

The point is John, and I am a little exasperated making the same point, you cannot say that the update caused the failure. It was a software update. A component has failed. The two issues are not related. This is the only reported failure of an Elite, which just so happened at the same moment as the update. Objectivity and logic John. R-

[Yelrow]

logic says was it ok before, yes, what has changed, oblectivity, the object was updated, putting different software through.. I know it does not suit to accept the two may be related, but we are aware of 2 batches of the Elite which are failing. Consequently, if this was one, putting any pressure on components, could have led to the failiure. I am not at all bothered, if i am out of step, or cause exasperation. This forum has a duty of care to posters. We should not put their controllers at 1% of a risk, if we know that certain batches are failing. There is another thread that clearly brings this to light. john

[RB51]

You make the point admirably John, some of the early batches of Elite do seem to have some functionality loss without the software update. You are, I think, confusing the two issues and potentially confusing your readers. The duty of care you speak of also means not allowing erroneous and unsubstantiated conclusions to go unchecked. R-

[96RAF]

Now we have 'batches' failing....John. I don't think so.

 

The facts are a very few Elites were sent for repair from a single batch code in 2011. I repeat 'very few of that batch' according to source, so maybe a bad component or more likely a workmanship issue at the factory in China, that passed initial quality control and production testing but then failed shortly after in service. Any others in that batch if faulty would have followed suit by now five/six years down the road.

 

We have also seen on these forums that a few have reported Programming output failure and one unit has suffered a Track circuit component failure. Then we have pidders ancient second hand device failing coincidentally with an update. I would like to have seen the inspection report before making a judgment on that one to see if the failure was in the memory area or somewhere else totally unrelated.

 

I would hardly consider that as BATCHES John and where duty of care fits into a public forum of opinion raised by private individuals I am not sure, unless you think Hornby should remove any and all vaguely adverse comment and provide a lifetime warranty for everything in their inventory. Ain't gonna happen either way.

 

This is said as an individual and nothing to do with the red title or Hornby.

 

[Yelrow]

Well, there is a very easy way to resolve this. Can we have from Hornby  written confirmation, that none of the  early, or more recent Elites, will be affected, by their  software updates, and if they are, will they stand behind them. This would clear the air, and satisfy me.. john

[pidder]

Didn't intend to start WW 111 about this. However, I've just tried another update on my ancient Elite, this time went all the way through with the updating bar, then said update failed. Is there any point on me wasting more time on this? Perhaps we should all just close this book and move on....

[96RAF]

@pidder

I have seen this update state before...

... trying to remember the context.

PC type I think. Try notching it down 1 and trying again. You have nothing to lose but your time.

 

If the Elite is bricked as we call it (I.e. Between engagements as the acting community would say) then retrying is the only option. If it works in any way then it would be interesting to know which bits work - the track and/or programming  then we have Boost to an isolated power district via an R8247 or hanging a Select on the Xpressnet socket as a walkabout.

 

If as may have been the case the Elite was duff in some way before the initial update then the update should still go through as all my Elites have some fault or other but all of them still update, except the one that smokes if you leave it on too long

[Fishmanoz]

Pidder, why try to change the firmware status of a duff Elite.  Even if you achieve a change of firmware, it will still be duff.  It's not as if v1.43 might somehow now not work, all the evidence is that it does.

 

john, 1%, now there's a nice neat figure but where on earth do you get it from?  We do not have any data to suggest it, or any other percentage, all we have is one Elite found to be faulty after an update was done.  We have no evidence one way or the other to tell us if the component or components now faulty were the ones being written to, in fact no evidence of a causal link between Update and failure at all. As a result, the idea that we might now suppose that updating an Elite's firmware might cause it to fail is technically and scientifically ludicrous.

 

My last word on the matter.

[pidder]

RAF Now there's a point. I had used my Select as a walkabout, connected to the Elite by a suitable cable, worked o.k. Of course it wasn't connected when I started this update lark, but could it have affected something? Perhaps we shall never know, unless "someone" tries the same trick...

[96RAF]

Pidder

If your Walkabout worked OK and was not connected during the update there is no reason to think it could have affected anything, so I think we can rule that out.  I have no idea what would happen if you tried an update with it connected without having a circuit diagram to chase the various circuits.

 

The simplest mistake someone could make by not reading the instructions properly is to connect the Boost terminals to their existing track in the presumption it will give more oomph, which of course does but the wrong sort of oomph.

 

Same sort of thing could happen if you left the Select power unit connected when using it as a walkabout, another disregard of the clear instructions.

 

These sorts of error can damage the Elite and maybe its only when an update fails the Hornby bench check flags them up, who knows? I am sure there are other schoolboy error methods of damaging an Elite that could induce latent faults.

 

Who knows what abuse your Elite had in its previous owner life, but it still seemed to work OK until the update.  A bit like going to hospital for one check and they find something else you didn't know you had.

Rob

[pidder]

Hi RAF Thanks for your further thoughts,...my Internet very slow today. I shall have to consider a new controller....

[Augustus Caesar]

I have followed this thread and the other with similar detail and am surprised at the level of thought that a firmware update can brick an Elite unit. This is not the case.

Logic tells us that IF a particular firmware update file is 'breaking' one Elite then it MUST by pure logic alone break others. This hasn't happened here. Fair enough some early units may have had a small issue some time back but this is now unworthy of thought in this case.

If an update fails with this version (1.43) then a simple reapplication of the update with slight changes to the PC type option and reloading of drivers etc will usually gain a solution in a short space of time. The Windows OS version is largely irrelevant with 1.43 so no issues there.

What appears to have happened here is that the Elite is at least attempting to read the firmware file but it is likely other chips on the circuit board have a fault which is stopping the write process from happening. The update file itself CANNOT brick an Elite. Sure, it can fail to update the unit but break it? No.

If this update had faults within it would have been corrected a long time back and Elites would have been able to take a new update version straight away.

 

Firmware, by its very nature, can be problematic only if the procedure is not carried out via the correct instructions, laptop powered by battery instead of mains so power loss causes a fail, other tasks are being done on a computer that is attempting the update to the Elite or, worst case scenario, the firmware chip itself is damaged so while a unit may work well under normal use until a firmware update is applied that will never get going because the chip is bad. This is not a very common problem but I have noted a couple on motherboards when the chip is unusable therefore unable to accept new firmware or BIOS changes.

 

Other chips that would aid in the writing process can also be damaged which will stop an update in its tracks. This has been alluded to in previous posts of course.

So the upshot really is an update bricking an Elite is extremely unlikely no matter what. Yes it can fail but a repeated attempt or two will sort it usually.

 

@pidder

If Hornby sent the unit back to you it may be because the firmware chip is blown and cannot economically be replaced therefore leaving the unit as it is now.. unusable because the chip cannot be written to.

 

@Chris

I've added a link (Update Elite with Windows 10) on the Help Site which, after reading this thread over, rather duplicates your process for updating the Elite via methods you point out elsewhere on the forum. I have not copied those of course but it does more or less mirror your method. However, a couple of small differences are there.

 

If anyone feels the need to take a look at this link it is on the Home page of the Help Site listed near the top of this sub forum.

[pidder]

AG Thanks for your further thoughts, I had thought this thread had reached it's end.  Hornby gave me no indication why they thought my Elite was economically beyond saving, I would assume they had seen the date of manufacture, perhaps tried an update themselves, but nothing said. I don't understand why, on one update try, I did get as far as the download indicator bar reaching the end, but then a report "update failed", (or some such) . I did think of opening the case myself, just to see if anything looked fried, but, as you know, it appears to require a special screwdriver. Think I shall just abandon it and waste no more time. Even if I bought a new unit, no guarantee that it would have the latest operating version... 

[Augustus Caesar]

It would be absolutely apparent to myself that if you can get the progress bar to reach the end of the update procedure only to fail at the last then what I say about the failure of a firmware chip is not likely to apply to yourself unless an intermittent fault is not all that obvious.

With the earlier Elites this failure was common and I had a few myself when my unit was based upon v1.3 I believe when bought (I may be wrong and it may be earlier but thus far that is the version I remember). However, persistent attempts and fiddling with all kinds eventually got each new update installed.

This latest version installed first time as stated and I haven't had the unit out of its box for a couple of years as I now use the eLink and RM..

 

I don't want to stomp over old ground and will leave this in your hands but if it were myself I would lay down some scenarios and record what I did each time and compare each attempt with others to see if common gorund can be dug up. Sometimes, you never know, something may just hit home and a certain combo of situations may just do the trick.

A bricked Elite though? I would find that very difficult to swallow even in your situation as an update should not do that. It will always accept an update another way if all is well inside. An older OS may help at a push. It depends upon what Hornby did when they looked at it and why they say 'uneconomical' to repair.

Have you considered asking WHY they said that? I would want to know but you have been digging at this for a while and maybe later on down the road you may want to take a fresh look.

[pidder]

AG Thanks for your response, certainly something for me to consider. Don't think there is any point me going back to Hornby,  but then I may.  I'll give further updates a try, perhaps with wifi or antivirus off. However we have also tried with my son-in-law's laptop with the same result, so I'm not really hopeful. Thanks,Peter.

[pidder]

AugustusCaesar  Just noticed I keep addressing you as AG, sorry, mind wandering as usual...