DIGITRAX DSC100 MAJOR PROBLEM

[TEAMYAKIMA]

Hello

 

As I have said many times, I am not a DCC expert, let alone a Digitrax expert. I did join a Digitrax users group but left almost immediately because I couldn't understand the questions on there, let alone the answers.

 

So here is my problem ........................

 

I have just exhibited at the 3 day Bristol show. For two days my system which uses a DSC100 and a DB210 booster worked perfectly. Then at 2pm on Sunday everything stopped. 

 

I unplugged the transformer box and left it for approx one minute and replugged it in. The DSC100 began beeping and so I unplugged it again. We tried to put something together using a spare old DB150.

 

I have looked up the meaning of the beeps in the DCS100 handbook. I have to admit that under the extreme pressure of knowing that the layout was at a complete standstill I did not stop to count the bleeps, but to me it sounded continuous.

 

I have looked at the manual and there seems to be two options. It lists a continuous 'clicking' but I would say that I heard 'bleeps'. It also list 16 bleeps - the maximum - and it says this means ....................

 

"Software timeout failure. No action is required, the unit will resume operation"

 

Eventually, we bodged something together using the DB150 and, as it was quite near closing time we got away with it, but we could not go to another show with the DSC100 in that state.

 

I plugged it in today just using one (of 12) boards as a test and the DCS100 works perfectly - so, the question is does any of this make sense ?

 

The DCS100 was s/h when I bought it and at a show it gets quite a bit of use - would this software issue explain the shutdown and subsequent return to normal.

[kev]

Sounds like you had a overheating problem make sure you get a good air flow round the unit as you say ok next day all I can think of

[luke_stevens]

Hi Paul,

 

You mentioned that the DSC100 was second hand when you got it. Do you know quite how old it is? Digitrax stopped production of the DSC100 in 2016 having started back in 1996...

 

If it is anywhere near that old (26 years) then no wonder it is falling over! I know it doesn't get much use but when it is used it is pushed to the maximum.

 

As far as I can work out when the layout is going hammer and tongs it can have over 24 loco's on it, most sound fitted. At one time there can be in excess of 8 working at once. I know you now have the cooling system installed in the "control box" so it shouldn't be overheating. It certainly doesn't overheat the way it used to before the cooling system. I wonder if we are simply drawing too much power?  From my experience on the layout, especially at Ally Pally, a lot of the loco's are being pushed as hard as they can / hauling long and heavy trains. I wouldn't be surprised it the layout sometimes pulls more than the command station can manage. The DSC100 manual says it is a 5 amp set up and that it can run 6 - 10 HO locos. But if we're using sound fitted logo with heavy trains then even though we have power districts I can see us overloading an ancient DCS100.

 

Is it time for a new command station? Maybe one that isn't too old for a Young Person railcard?

 

Luke

[Nigelcliffe]

11 hours ago, luke_stevens said:

.....At one time there can be in excess of 8 working at once. I know you now have the cooling system installed in the "control box" so it shouldn't be overheating. It certainly doesn't overheat the way it used to before the cooling system. I wonder if we are simply drawing too much power? ....

 

Measure the current being drawn from the DCS100 before deciding that "more power" is needed.   Some locos, from some makers, draw lots of current.   Other locos, from other makers, draw hardly any current.    So a layout which uses power-hungry locos in OO/HO can run into problems on a 5A system with only four moving.  A layout with power-efficient locos will be happy with more than fifteen moving.   

 

Paul said there's a DB210 booster involved.  And a spare DB150 was used to substitute for the DCS100.   A DB150 has about the same power output as the DCS100.  If the DB150 ran the layout on its own, then there isn't a power problem from a 5A command station. 

 

For large layouts, a common arrangement is to have the command station not feeding any track.  So, its power output is irrelevant as it is not powering any track.  Boosters feed the track, divided into different sections of the layout to spread the load.  

 

 

An old DCS100 may be as simple as the internal batteries are expired.  There are two of them inside, change one at a time and the DCS100 keeps any settings by using the second battery (assuming its got enough power to keep things going).    That's a very cheap thing to check/change.  

 

 

 

- Nigel

 

 

[TEAMYAKIMA]

25 minutes ago, Nigelcliffe said:

Paul said there's a DB210 booster involved. 

 

For large layouts, a common arrangement is to have the command station not feeding any track.  So, its power output is irrelevant as it is not powering any track.  Boosters feed the track, divided into different sections of the layout to spread the load.  

 

An old DCS100 may be as simple as the internal batteries are expired.  There are two of them inside, change one at a time and the DCS100 keeps any settings by using the second battery (assuming its got enough power to keep things going).    That's a very cheap thing to check/change.  

 

Nigel

 

Nigel

 

Thanks for your interest/comments.

 

The arrangement is that there are four tracks, the  command station feeds two tracks and the booster feeds the other two.

 

We have had issues in the past with too much current draw and so I bought the 8 amp booster to hopefully solve that issue and it appears to have done so.  What is confusing me is the beeping - what is that telling us?

 

When I first went down to investigate, the DCS100 was beeping like mad - I admit that in my 'panic' I did not stop to count them (I know that there are various coded signals depending on the number of beeps). I initially unplugged it for a minute or two - still beeping when I replugged it. I then waited maybe 30 minutes - still beeping. Next day - no beeping, all good, but only plugged into one board.

 

I don't remember the DCS100 beeping when we had the original cut out problem several years ago before I bought the DB210.

[Nigelcliffe]

 

The DCS100 an old product, long time since it was replaced by Digitrax.  So, ask a dealer if there are any servicing options for it, or if you're on your own with it.   That might alter decisions taken. 

 

The battery change is simple.  Its a standard coin cell used in lots of devices.  My local high-street electrical shop sells them.    Its a CR2032.    Digitrax designed things so you insert a new battery into the empty slot, then remove the old one from the other slot.    For £3 spend, its a simple thing to try.  

 

Much of the documented beep numbers from the DCS100 don't have an end-user option to deal with them.  So, in many respects the information in the number of beeps isn't particularly helpful.  

 

As you've got a DB150 available, that means you have a backup for show failures.   So, the question is whether the DCS100 will do a show without problems (perhaps with a new battery to find if that was the cause), or whether its now unreliable.    Which is then, is service an option, or is replacement the only route ?   

 

Replacements from Digitrax range are approaching £400.  A DCS240 with a power supply is about £370-£400.   Alternatively, the EVOX system costs about £370, with a DCS210 command station, and a DT602 throttle and power supply in a complete package - that's possibly a better option as there's a spare throttle for the same amount spent.   
There are non-Digitrax LocoNet compatible options, possibly cheaper, but given the technical skill levels on the layout going to multiple brands may be dangerous with lack of support. 

  

 

 

[Oldddudders]

I used an elderly DCS100 for the first time in ages the other day, and it just beeped and beeped. A quick renewal of the CR2032 battery sorted that completely. My other DCS100 is  continuously powered and hasn't needed a battery change in many years. Incidentally, the older one was bought in 1998.

 

As far as the Digitrax online community is concerned, I note the list-owner himself still uses a DCS100, as do many others. Do not splash your cash until you are certain yours is actually duff. 

[TEAMYAKIMA]

26 minutes ago, Oldddudders said:

I used an elderly DCS100 for the first time in ages the other day, and it just beeped and beeped. A quick renewal of the CR2032 battery sorted that completely.

 

That's very interesting. Did that co-incide with a total power shut down as in my case?

 

I ask that because from the info I've managed to glean from various sources, I have assumed that the battery only maintains settings like consists and therefore a dead battery would not cause a shut down and so I await your answer with great interest.

 

The problem with this issue is that there are so many variables which might play a part in the issue - please see my reply to Nigel which I will post shortly.

[Oldddudders]

The older DCS100 had been unused in some time, so the CR2032 had slowly flattened, as happens when not powered up. 

 

EDIT : And, thinking back, while the DCS 100 was beeping continuously, it refused point blank to accept any input from the plugged-in DT500, thus suggesting that loss of battery power does indeed affect overall performance. 

 

Incidentally, when I first bought into Digitrax in 1997, the DB100+ came with a dire warning that opening it up would invalidate the warranty. Yet opening it up is needed to replace the battery! Digitrax manuals have not improved much over the years, sadly....

Edited May 3, 2022 by Oldddudders

[luke_stevens]

1 hour ago, TEAMYAKIMA said:

 

The arrangement is that there are four tracks, the  command station feeds two tracks and the booster feeds the other two.

 

 

Paul, when the layout "fell over" was it the 2 tracks from the booster, the 2 tracks from the command station, or all 4 tracks?

 

Luke

[luke_stevens]

2 hours ago, Nigelcliffe said:

Some locos, from some makers, draw lots of current.   Other locos, from other makers, draw hardly any current.    So a layout which uses power-hungry locos in OO/HO can run into problems on a 5A system with only four moving.  A layout with power-efficient locos will be happy with more than fifteen moving.   

 

 

Most loco are Bachman China (plastic), but some are from other manufacturers (plastic) but there also many brass locos (both steam and diesel).

 

From my experience with the layout many of the locos are being used to their maximum. Frequently we can't get a train to run at anything more than a slow walk because of the weight behind it: remove wagins / coaches and the train speeds up. Run a loco more than one circuit at a time and the decoders will shut down / stop / stutter / flash lights.

 

Luke

[TEAMYAKIMA]

TBH, I have just spent two hours writing an in depth reply and somehow RMweb has deleted it before I could post it.

 

I will have to write a shortened/simplified reply.

 

There were all sorts of other DCC 'issues' over the first 3 days (including the Thursday) and I have no idea if they are relevant or merely co-incidental. I really did spend 2 hours preparing a reply to post here and I really don't know if I can go through all that again today.

 

However, the move to the DB150 was not without its problems. Both the DCS100 and the DB150 now work perfectly in my shed connected to just one board (there are normally 12 boards in total) - my worry with that is that it might mean that the problem may not be the DCS100 or the DB150, but an inherent layout problem. Maybe the move from my storage unit which has a dehumidifier on 24/7 to a packed, hot, humid exhibition hall had something to do with it.

 

I have to go off now, but I will try to list all the other issues in sequence at some time to see if there is a pattern.

 

Thanks to one and all for your input.

 

Paul

 

 

[RobinofLoxley]

For further reference, if you are writing a long item its safer to write in word and to paste in here.

 

The usual reason for text 'disappearing' is that your login times out in the writing period, so when you submit reply its not added to the thread. In that case, login again, return to the thread, click on the empty reply box and you should get a dialogue asking you if you want to restore the original (missing) text. Its really frustrating to lose a long post.

[AndrewC]

Paul, others like Nigel have commented around the DCS100. I've still got one and an even older DB100a. 

Having fed and watered them for well over 25 years, these are my findings and suggested order of trouble shooting. 

 

1: cmos battery. Do this first. Do this annually. 

2: loconet voltage being dragged down. Too many throttles without batteries or non-powered UP5 panels can drop the loconet voltage. If there are send issues with the command station you'll get 9 beeps. (then another 9 and another 9... ) 

3: bad loconet cable. Use a cable tester from where you connect the DCS from and every terminating point on the layout. This means each throttle panel, stationary decoder, detection device, etc. I had one major failure where a bit of grit had got into an RJ12 socket on one of my panels and nothing beyond would work as it blocked one pin from connecting to the plug. A bad connection or cable can also cause the 9 beeps of frustration. 

4: slot max. If you have JMRI connected, keep your eye on the slots, and the loconet messages (from 2 and 3) One strange design "feature" of Digitrax systems was to have a series of slots once you select a loco. If you just grab a new address you don't clear the slot. Eventually you run out. Check your system to see how many it is configured for and purge them through out the day during an exhibition. There are loads of JMRI python scripts that will do this automatically. 

5: Check the input voltage and current rating of the power supply. If you have a multi-voltage supply it is easy enough to knock it from 16v to 9v. 

6: Support. Not happening. The 100 is regarded as out of support and beyond repair. 

7: cooling is important. That's why it has fins. I built a Digi-box based on what Chris Shinn did on the Digitrax list many years ago. It contains a fan for cooling. 

8: Inrush current and sound locos. I've found that after any sort of short where there are multiple sound locos the DCS misbehaves afterwards. IF the short was on the booster, no problem. I found a PSX makes a big difference. In fact I now break all my power districts into no more than 3a sections, all protected by PSX units. 

[luke_stevens]

Hi Andrew et al.

 

Knowing the layout...

 

1) damned good idea! quick, simple and cheap. I didn't known there was a cmos battery in the digitrax command system but in other situation I have has dead / dying cmos batteries cause serious IT concerns.

 

2) We used to use wireless throttles but they cause so much anguish that we moved to plugged in ones which gave us much better performance (at Ally Pally) but we didn't have batteries in them. This is another easy fix.

 

3) Some of the loco net cables could do with replacing (cables plugged into extension cables, broken plugs). Again another easy fix.

 

4) The layout doesn’t used JMRI. We have had slot=max issues in the past but it is less common now. We only had on occurrence at Ally Pally. It’s possible that the DSC100 had reset to 22 slots when it was repaired but it is/ has now been reset to 120 so this shouldn’t be a problem.

 

5) I think the power supply is reasonable new and is a Digitrax one but Paul can confirm.

 

6) Ha! Yes we'd realised that!

 

7) In the past (Fareham?) we had serious problems with this so Paul built / had built a control box with large fans, etc. Since then we haven’t had an overheating issue (as far as I’m aware)

 

8) The layout is divided into 4 districts, 2 powered via a booster, 2 via the command station. It’s not obvious from Paul’s post which districts went down. Maybe it would be worth Paul getting another booster so all 4 districts are run vi a booster, putting less stress on the command station?

 

Thanks you all for your help! It’s a great looking layout which is even better when ti works properly!!!

 

Luke

[Nigelcliffe]

To Luke's list above. 

 

Point 2 - should power all UP5's (they can share a power supply, see Digitrax documentation), thus throttles get power from the UP5, and don't drag the LocoNet down. 

 

Point 4 - if any of the team have a copy of LocoNet Checker, that's a really good PC tool for slot management.  Unfortunately, it was withdrawn by its writer some years ago, so its only used by those people who have old copies.     

 

Point 8 - optional, and I suspect not needed.   Using command station only for commands does mean it will run cooler (not powering track), but I think its not the key issue here.    If booster is available cheaply, do it, but otherwise look at the other factors first.  

 

 

 

 

[luke_stevens]

Hi Nigel,

 

Thanks for the response. It at least helps me get a better idea of what is going on.

 

2) I don't know if the UP5 are separately powered but it sounds a good idea...

 

4) I use DCC more than the rest of the operating team so it would have been me if anyone, but I don't have. Sounds like it could have been really useful :(

 

8) Makes sense: nice idea but not needed.

 

Whatever the current problem turns out to be it's the first time it has happened in 6 exhibitions so it is not a common problem (for us) or something likely to reoccur on cue. I know Paul is very concerned that it might happen again but until it happens again / we can make it happen again it is going to be difficult to cure.

 

Thanks all for you thoughts and comments, I'm sure they will help and reassure Paul.

 

Luke

[luke_stevens]

On 03/05/2022 at 13:55, TEAMYAKIMA said:

Maybe the move from my storage unit which has a dehumidifier on 24/7 to a packed, hot, humid exhibition hall had something to do with it.

 

I wouldn't think that sitting on the floor of an exhibition is going to be that hot and humid...  I think I remember that you keep the control box in you larger (heated) shed rather than you smaller (unheated) one which must help.

 

On my quick calculation the layout has been to 6 exhibitions and set up in your garden 3 times so in total the layout has had about 20 "operating" days. Whatever the problem on Bristol day 3 was it is not a common problem; that is, I'd be very surprised if you could replicate it in your shed. I'm sure it is a function of the layout being used intensively for many hours on end. I'd guess that you would need to replicate exhibition conditions to get it to happen again.

 

It could be useful to setup a system where the DCS100 can be quickly swapped for the DSC150 so if it does happen again we can quickly recover and get the layout working again.

 

At this stage the layout works :) We just need to make small tweaks to make it more reliable, removing / changing little things to improve performance, starting with the first few things Nigel suggested.

 

The layout is getting there. It works and (most of the time) works well!

 

Luke

 

[WIMorrison]

1 minute ago, luke_stevens said:

I wouldn't think that sitting on the floor of an exhibition is going to be that hot and humid...  I think I remember that you keep the control box in you larger (heated) shed rather than you smaller (unheated) one which must help

 

Definitely wasn't hot and humid at Ally Pally - I was on a stand fairly close to you, and we were freezing!!

[TEAMYAKIMA]

59 minutes ago, luke_stevens said:

By my reckoning, the layout has been to 6 exhibitions and set up in your garden 3 times so in total the layout has had about 20 "operating" days. Whatever the problem on Bristol day 3 was it is not a common problem; that is, I'd be very surprised if you could replicate it in your shed. I'm sure it is a function of the layout being used intensively for many hours on end. I'd guess that you would need to replicate exhibition conditions to get it to happen again.

 

Yes, that is the BIG problem. I could hire a van, hire a church hall, put some of my team up in a local hotel and operate the layout intensely for two days and have it work perfectly, but unless we can categorically identify why the DCS100 failed at Bristol, having it work for two days in a virtually empty church hall in Watford doesn't mean necessarily, that it will work perfectly at a show with 300 people crammed into the hall 200 miles away. 

 

I must be honest with you, I have only mentioned the big problem we had at Bristol. We had other relatively minor issues and, again, that, in itself, is a problem. Were those other 'issues' part of the same problem or just co-incidences? In other words are these other issues relevant or merely red herrings?

Edited May 4, 2022 by TEAMYAKIMA

[TEAMYAKIMA]

Ok, I will try again to produce the full list of 'issues' and as I have said before - are these relevant or red herrings?

 

1. On the Thursday, we had a chance to do some basic testing Two tracks were perfect and two tracks had running problems. I am 99.9% certain that the two problem tracks were those controlled by the DCS100. I left the hall on Thursday making a mental note to prioritise that problem. I turned up early on Friday and the problem did not exist. Relevant or red herring?

 

2. On Saturday morning, we had some route selection issues with four sets of points controlled by a DS64 powered by one of the tracks controlled by the DCS100. I disconnected all points in question and when I reconnected them yesterday they all worked perfectly. Relevant or red herring?

 

3. After the DCS100 failed we substituted the DB150. It also gave problems. The two tracks powered by the booster ran perfectly, but trains on the two tracks controlled by the DB150 seemed to run incredibly slowly on full power and then we discovered that they were actually accelerating away - we were told by a DCC user who came over to help that it appeared that the command station had been 'altered' to always provide acceleration - he said it was possible. However, I have tested those locos since I came home and it appears that their chips have been modified to have acceleration as they still show those characteristics whereas the locos which ran on the booster tracks always ran without acceleration and still do. How did the locos on the tracks operated by the DCS100 get their chips altered? And is it relevant or a red herring?

 

4. When the DCS100 failed, everything stopped - all four tracks. It always beeped as soon as it was plugged in. I did leave it for approx 20-30 minutes at one time, surely that would give it time to 'cool down'?

 

5. The DB150 also gave problems. It failed 5/6 times, but it never beeped. This s/h unit had been our original unit and at our first two exhibitions when it powered all four tracks. It was OK at our first exhibition when our steam locos didn't have sound. It dropped out due to overload/overheating several times at our second show by which time, several of the steam locos had sound. It was after that show that we bought a DB210 booster and a s/h DCS100. Since then we had had no problems.

 

6. I had assumed that as both command modules had failed on Sunday, the problem must be inherent in the layout rather than in the command stations. But as the DB150 did always (on Sunday) run for a few minutes before dropping out, was it merely replicating what it had done in the past? On Sunday it was only running two tracks, but one was generally a double headed long steam train with both locos having sound and the other was a double headed diesel hauled 17 coach passenger train. So, could the DCS100 problem and the DB150 problem be unconnected?

 

I am guilty of not making notes of all issues at the time e.g. I did not count the beeps - it just seemed continuous to me.

 

Thanks for all your comments/advice.

[Nigelcliffe]

Paul's point (3) above appears to contain garbled information.  
It sounds like (but without programming track readback information to check) that something caused the locos to have CV's for acceleration changed.  There are several ways this could happen, but it is particularly easy to do with a DB150 by a throttle mistake by a user (any programming action would hit every loco on the DB150's output, so one mistake on one throttle can reprogram everything on the tracks connected to the DB150.).     
There's nothing in a command station itself which sets acceleration (garbled information??).   

 

Point (4). Some thermal cut out devices can take quite a long time to reset.  I don't know what is fitted to the DCS100. 

 

 

The advice hasn't really changed. 
1 - change the DCS100 battery.   That's still the most likely issue.  And its only £3 to change it.  
2 - look at the Power Supplies - are they failing to provide all the power asked of them ?  
3 - make sure the UP5 LocoNet sockets have power supplies.  

4 - find a way to measure the current drawn from the DCS100.  A good approximation is measuring the power going in at the power supply.   Or, fit an output overload cut-out set to 3Amps, if that doesn't trip, then not drawing over 3A.     That will answer whether the layout is drawing too much power (so needs additional boosters) or not.   
 

 

[TEAMYAKIMA]

14 hours ago, Nigelcliffe said:

Paul's point (3) above appears to contain garbled information.  
It sounds like (but without programming track readback information to check) that something caused the locos to have CV's for acceleration changed.  There are several ways this could happen, but it is particularly easy to do with a DB150 by a throttle mistake by a user (any programming action would hit every loco on the DB150's output, so one mistake on one throttle can reprogram everything on the tracks connected to the DB150.).     

 

Nigel

 

Thanks for your advice - I have today bought the batteries but not fitted them yet.

 

Yes, I questioned the DCC guy when he suggested that somehow the command station had been altered - I said surely it's a chip issue? That's when he said that it was possible/likely that the command station had been altered and I didn't know enough to challenge that. He said he did DCC but not Digitrax and so what he said may be true of other brands.

Edited May 5, 2022 by TEAMYAKIMA

[WIMorrison]

May I suggest that if you think that the decoder settings are being changed inadvertently by a user then you simply lock the decoder to prevent any accidental changes happening?

 

On Zimo decoders this is CV #144, I believe that Digitrax do something with CV #15 and CV #16 to achieve a similar function (they may even have a CV that simply locks the decoder )

 

[Nigelcliffe]

1 hour ago, TEAMYAKIMA said:

 

Yes, I questioned the DCC guy when he suggested that somehow the command station had been altered - I said surely it's a chip issue? That's when he said that it was likely that the command station had been altered and I didn't know enough to challenge that. He said he did DCC but not Digitrax and so I have no idea if what he said is true of other brands.

 

You're correct - chip issue may be the case (and I outlined how that could happen with a DB150 in play).  

Can't see how its a command station issue, for any brand.   

 

 

Iain is correct about decoder locking - Soundtraxx also has decoder locking options.  

 

 

 

 

 

- Nigel