DIGITRAX DSC100 MAJOR PROBLEM

[peach james]

OK, you have a spare DB150, right?  Rig it up so the DCS 100 is ONLY being used to generate command signals, and feed the loconet bus, and the DB150 is feeding 2 tracks and  the DB 210 feeding the other two.  That way, when things go wrong...if it affects 2 tracks, it is the booster for those 2 tracks, if it affects all 4, then it is the command station.  

I always try to travel with a spare command station to shows which are not right near chez James', as that way I have the bits for when the magic smoke goes poof... but that is a $400 investment or so.  I didn't see any for sale DCS 100's at my last show, but it might be worth trolling some of the North American companies who do 2nd hand to try and get a 2nd one, as it is a bit of a point failure.  The DB150 is not a great replacement 1:1 for it.    (Central Hobbies/Otter Valley/Credit Valley/Georges up here in the great white north)

Other options:  https://www.tamvalleydepot.com/products/dccbooster.html  one each track, supplied with DCC signal from the DCS-100, and fed from independent wall warts.  If I was building what you have, that would have been a good option at the start.  Its rather more work now to redo wiring than it is at the start.  (I know, I stripped our club layout and made it DCC last year...)

I know- I hate spending $ on stuff, but at the end of the day, sometimes that's the best solution.  It sounds to me like it is likely an overheating issue, that or the batteries are flat.  Again, if possible, just use the DCS-100 as the command station, and the "spare" DB150 as a booster.  That way, if it does go pear shaped, you should only have to remove the jumper from the DB150, and unplug the DCS 100 to be able to use the DB150 as a command station, leaving the DCS100 out of it.

But the first thing is change the batteries...

(PS:  My show @#@#$ of the day was the throat entrance turnout servo not working...)

James

[TEAMYAKIMA]

Thanks to everyone for their thoughts. I am trying to digest all the advice and ideas. I have also been in direct contact with Digitrax and they have struggled to understand my psychological problem. Their viewpoint is 'You say it's working fine now, so what's your problem?'. They have told me that shutdown caused by overheating can in certain circumstances last for 30 minutes or more and so on the surface that does fit the known facts. 

 

I suppose that if I could be 100% certain that the problem was overheating caused by running too many long heavy trains, then I could simply cut back on the length of trains and all would be well. So I guess the fundamental question is ' Is there any other possible explanation, other than overheating due to excess current draw, which could explain the shutdown and subsequent reboot?' 

 

If the answer to that is 'no' then that would put my mind at rest. 

 

 

[fulton]

Keep running those long trains, they really set your layout off, they look like the layout was designed for them, find a new control system if need be. The Digitrax system has worked well for me, on four different layouts, I only use the simple UT "buddy" controllers, as it was a problem with operators pushing the wrong button, causing all sorts of issues.

[TEAMYAKIMA]

14 minutes ago, fulton said:

Keep running those long trains, they really set your layout off, they look like the layout was designed for them, find a new control system if need be. The Digitrax system has worked well for me, on four different layouts, I only use the simple UT "buddy" controllers, as it was a problem with operators pushing the wrong button, causing all sorts of issues.

 

Yes, I want to and yes, long trains have always been at the core of the concept. One suggestion has been to buy a DCC CONCEPTS meter which will allow me to monitor power drain. 

[Andymsa]

I did send you a PM about this and overheating but had a lack of reply from you. As for overheating again I repeat put a 80mm fan right next to the dcs100 heat sinks about 2 inches away on the back of the dcs blowing onto the heat sinks. A possible explanation could be after a long time of use at various exhibitions due to this overheating which was coped with by the dcs For a time, a component breakdown could now be occurring as electronic parts do not like overheating. Running long trains will not cause overheating.

 

after reading your original post again the timeout bleeps could be an unrelated issue I’m taking a stab in the dark here but it does sound like a loconet issue. I did have a similar issue myself where an input output loconet chip on an accessory board was giving intermittent problems was causing the 16 bleeps to occur. 
 

as for batteries I’ll put my neck out here that they are not the issue as the batteries function is to hold the dcs option switch settings when everything is shut down.

[TEAMYAKIMA]

Just now, Andymsa said:

I did send you a PM about this and overheating but had a lack of reply from you. As for overheating again I repeat put a 80mm fan right next to the dcs100 heat sinks about 2 inches away on the back of the dcs blowing onto the heat sinks. A possible explanation could be after a long time of use at various exhibitions due to this overheating which was coped with by the dcs For a time, a component breakdown could now be occurring as electronic parts do not like overheating. Running long trains will not cause overheating.

 

after reading your original post again the timeout bleeps could be an unrelated issue I’m taking a stab in the dark here but it does sound like a loconet issue. I did have a similar issue myself where an input output loconet chip on an accessory board was giving intermittent problems was causing the 16 bleeps to occur. 
 

as for batteries I’ll put my neck out here that they are not the issue as the batteries function is to hold the dcs option switch settings when everything is shut down.

 

Yes, many apologies for not replying to your PM. TBH this problem has deflated me so much that I have just been metaphorically staring at the wall for the last 4/5 days. The trouble is that there are fans. They blow air over the transformers which are situated under the plywood in this photo and then escape through the white vents and the command station and booster sit on top of those vents.

 

[Andymsa]

The issue with this setup is your blowing warm hot air onto the dcs/ booster plus they sit above the transformers and any heat from them will rise.

[WIMorrison]

Anything that is sat in a box will need significantly more cooling air blown over it and out without passing over something else than will items sat is free air.

 

take a look at the inside of a desktop computer and will see several fans if it is a reasonably powerful computer and all placed to ensure that high airflow is provided - essentially you can never have too much cooling because heat is hard to expel, but it is very easy to have less than needed and create locate hot spots that are not in the air flow.

 

In addition to have fans dedicated to each item I would take the sides of the box you are using - or at least replace the sides with a wire grating.

 

ideally borrow a thermal camera and take a look through it, you will be surprised at what you see, I know that I was on my layout when I looked at my command station and booster. 

[Andymsa]

I will reply here to save confusion. 

 

So I assume that your coaches do not have lighting units or wagons have resistive axles if so then any stock other than the loco have no effect on the dcs/ boosters in regard to current draw.

 

”As regards your comment in the main thread about component damage - is that consistent with the DCS100 returning to normal when connected to just one board and one light loco in my shed?”

 

I have to say I’m afraid that your not creating the same conditions in your shed as opposed to an exhibition, so any inherent faults might not manifest themselves as the loadings  on the dcs/ boosters will be different. A Component might be breaking down when the overheating occurs.

 

as I said in last post I certainly think your design of the power system and the current setup should really be redesigned.

 

 

 

 

[TEAMYAKIMA]

Just for clarification I am posting some more photos of the command station set up as I feel my previous photo may not have made the situation clear. The command station and booster do not sit in a box, they sit as shown. The power box was made for me by a senior member of MERG who is following this thread.

 

A posed photo showing how the units sit over the vents

 

 

Power box with the plywood floor removed

 

 

 

One end

 

 

Other end

 

[WIMorrison]

That looks like one of the best solutions for retaining heat that I have seen for a long time.

 

it also doesn’t show any cooling of the the digitrax units at all, all I see is natural convection as cooling.

 

 

Edited May 8, 2022 by WIMorrison
Spelling

[TEAMYAKIMA]

37 minutes ago, Andymsa said:

So I assume that your coaches do not have lighting units or wagons have resistive axles if so then any stock other than the loco have no effect on the dcs/ boosters in regard to current draw.

 

Edited May 8, 2022 by TEAMYAKIMA

[TEAMYAKIMA]

Two of the coaches do have lights, but the train is so heavy that we require two diesels double-heading running at full power to pull them - surely that will increase current draw?

 

43 minutes ago, Andymsa said:

I have to say I’m afraid that your not creating the same conditions in your shed as opposed to an exhibition, so any inherent faults might not manifest themselves as the loadings  on the dcs/ boosters will be different. A Component might be breaking down when the overheating occurs.

 

Yes, that is my worst nightmare. As I can't set the layout up at home I would never know until I get to the next show - which is really too late. 

Edited May 8, 2022 by TEAMYAKIMA

[Andymsa]

I’m sorry I have to say that is the most dreadful set up you can have, let’s break this down you have fans at each end with no clear path of air flow I’m assuming one fan pushes and on pulls air through , the air holes under the digitrax units have absolutely no cooling path at all, because air flow is not being pulled up through those holes under the digitrax units. What I find most distressing is this was designed by a merg senior member.my recommendation is scrap it.

Edited May 8, 2022 by Andymsa

[Andymsa]

5 minutes ago, TEAMYAKIMA said:

Two of the coaches do have lights, but the train is so heavy that we require two diesels double-heading running at full power to pull them - surely that will increase current draw?

 

 

Yes, that is my worst nightmare. As I can't set the layout up at home I would never know until I get to the next show - which is really too late. 

I would say not to the point that it would cause the system to shut down by overheating

[WIMorrison]

Andy

 

Being a Member of MERG of whatever doesn’t make you into an competent engineer 😊

[Andymsa]

A further thought occurs the set up is a  significant fire risk

Edited May 8, 2022 by Andymsa

[WIMorrison]

These pictures show how heat is developed. 

The bright orange elements as the SMPS units and the elements under the temperature mark are my Z21 and the booster.

 

I am drawing 2.1A on the Z21 and 2.6A on the booster. You can see that the heat is being effectively dissipated in free air. It also shows that the hottest elements are the SMPS which measure at 24.3c and 26.1c respectively.

 

You need to get effective cooling of your components, and get rid of the wood as it is an excellent heat insulator allowing the heat to build up in your box - plus it is highly combustible. You need direct cooling of the DCS100 and the booster, you need to baffles inside the box to direct air into the bits that need cooling and you need to separate the SMPS to allow cooling from all of their surfaces - they can get extremely hot.
 

I would also move the power strips away for the heat sources as the pictures suggest that are actually above the SMPS units, and some other items which look like they generate heat from the cooling holes in the mounting shields.

[Andymsa]

 

not a good photo but this is my auxiliary power supply which I built. It holds 5 5amp transformers. The top of the unit has a grill which covers the whole top of it, it is fully fused on input 240 for each transformer and is fused for each output.

 

as a footnote this was my original power supply for the dcs100/ db150 

Edited May 8, 2022 by Andymsa

[TEAMYAKIMA]

44 minutes ago, Andymsa said:

I’m sorry I have to say that is the most dreadful set up you can have, let’s break this down you have fans at each end with no clear path of air flow I’m assuming one fan pushes and on pulls air through , the air holes under the digitrax units have absolutely no cooling path at all, because air flow is not being pulled up through those holes under the digitrax units. What I find most distressing is this was designed by a merg senior member.my recommendation is scrap it.

 

I can confirm that both fans draw air into the transformer box. I must add that I have the highest respect for the person who made this box for me  and I have emailed him so that he can read your comments and reply if he so chooses.

[Andymsa]

8 minutes ago, TEAMYAKIMA said:

 

I can confirm that both fans draw air into the transformer box. I must add that I have the highest respect for the person who made this box for me  and I have emailed him so that he can read your comments and reply if he so chooses.

im certainly not calling into question the person who made your power supply box. I was trying to resolve your issue, I’m not going to get involved in a third party conversations. Sorry if this come across as heavy.

 

as to your fans both draw in cold air depending on ambient conditions, and warm air is blown on to an already hot dcs/ booster. I can’t put it any more simply than that your problems won’t go away with the current setup.

[Nigelcliffe]

I'm staying out of any questions around the power supply box, other than to say that I don't think its doing anything to cool the command station or booster stood over it.   If wanting to cool the DCS100/DB210, then a different arrangement is needed to move *cold* air over the fins. 

 

Back at the main issue.  Is the DCS100 overheating due to working too close to its limit of output for too long ? 

Answer to this is "unknown" at this stage, but if one assumes it may be, then I can see two approaches:

 

a)  measure what current is being drawn, over a period of operations.  That can be done several ways, but the simplest two are:  measure the input current going into the DCS100, or measure the output current going to the track. 
Measuring the input current is probably cheaper to do but more complex - it amounts to "identify, buy and fit the right sort of meter". 
Measuring the output current costs more, but is simpler to do - it amounts to "buy a RRampmeter or a DCC-Concepts Alpha Meter" and fit to the track output leads.  Either will cost about the same - about £80.  
This gives a figure, but doesn't fix anything.   A figure is useful, it tells whether the current drawn is regularly going near 5A, or whether its pretty much always comfortably below (say 3.5A or less).   If its always below, then the problem isn't too much current, but something else. 

b) reconfigure the existing devices, so the DCS100 is not drawing much current, thus cannot be over-loaded from current draw.   There appears to be a DCS100, and two boosters available.  The newer DB210 (in regular use) and and older DB150 (backup command station). 

  ( This was suggested by another poster, higher up the thread).

The DB150 could be configured as "booster", and inserted as the track supply for those tracks the DCS100 currently feeds.   Leave the DB210 powering its current tracks. 
Then, the DCS100 is only powering the LocoNet (negligble current), and providing the command station input to the DB210 and DB150 (again negligible current). 
Optionally, the DCS100 could feed a small test track on the layout for any loco repairs and programming.   

**Need to check** whether the DCS100 and DB150 can share the same power supply output, or whether independent power supplies are needed. 

 

 

- Nigel

 

 

[TEAMYAKIMA]

2 hours ago, Nigelcliffe said:

Back at the main issue.  Is the DCS100 overheating due to working too close to its limit of output for too long ?  Answer to this is "unknown" at this stage, but if one assumes it may be, then I can see two approaches:

 

a) measure what current is being drawn, over a period of operations

Measuring the output current costs more, but is simpler to do - it amounts to "buy a RRampmeter or a DCC-Concepts Alpha Meter" 

Nigel

 

Nigel

 

Many thanks, as usual, for your advice - you wouldn't consider moving about 400 miles closer to Watford would you? 😉

 

A DCC-Concepts Alpha Meter will be bought. 

 

I have been in touch with my 'local' MERG man and we are considering all the comments and suggestions that have been made here and we will be posting our thoughts shortly - many thanks to everyone who has contributed.

Edited May 9, 2022 by TEAMYAKIMA

[WIMorrison]

I am looking forward to the redesign of the box that allow it to provide some  cooling for the Digitrax units rather than heating them up.

[Andymsa]

Maybe something along the lines of this for the redesign