Jump to content
 

Managing start up power surges in DCC


young37215
 Share

Recommended Posts

  • RMweb Gold

I am having difficulties with applying track power using my ESU Mobile Control which is part of their Cab Control (CC) system. It appears that the initial power draw/rush of my 25 locos all fitted with Loksound chips (V3.5 x 5, V4 x 15, V5 x 5) causes CC to intermittently trip its short circuit protection. I have adjusted the CC current limit setting to the maximum of 5 Amps and CC clearly provides adequate power for my locos once I get past the initial start up issue because I am able to run trains without any problem. There is a live thread relating to CC where I have received considerable assistance with the challenges I have faced introducing CC to my layout but I dont think that the questions below are specific to CC and suspect that they are generic to any DCC system hence the start of a new thread.

 

The potential solution to CC spuriously tripping its short circuit protection looks like a PSX circuit breaker which purports to include logic entitled 'adaptive load reset' (ALR) to manage initial power rush on start up. I am intrigued at this claim, I am not electrically minded and struggle to see how the PSX can stop the CC sensing whatever it is that upsets it. I have read comment about 'soft start' and am guessing that the PSX's ALR incorporates a variation on this. Can anyone explain soft start or how the PSX ALR works? 

 

In trying to logically find a solution and because I want to get the correct setting for the adjustable trip level settings on the circuit breaker, I want to understand how much power an ESU sound chipped loco initially draws when power is first applied. CC does not incorporate any means of measuring power demand on start up; can anyone give me an indication?

 

Any guidance gratefully received.

 

Link to post
Share on other sites

  • RMweb Premium

I'm not sure how you would measure the start up current as when power is first applied much electronic equipment will appear as a shortcircuit across the power supply as capacitors charge up from zero volts. This is normal for anything electrical beyond basic resistances.

How much current actually flows at power up will depend on many factors such as any apparent resistances (impedance) in the circuit (leads, track feeds, diodes, inductors etc. as well as anything in the PSU)

 

Most power supplies will be able to supply way over the rated output for a brief period until an equilibrium is reached, however fast acting overload trips may cut as soon as power is supplied

Soft start, as I understand it, is when the supply voltage doesn't change from zero to max in an instant but has a time constant whilst it is ramped up, which limits the start up surge. (It's also kinder on components as most electrical failures are at switch on)

 

AFAIK You can have a soft start on a breaker as the start up surge is different to an overload happening whilst running normally.

 

 

  • Thanks 1
Link to post
Share on other sites

  • RMweb Gold

The PSX was a development of the original Powershield circuit breaker, and as you have said, it is designed specifically to recognise a "current inrush" from multiple sound locos. To enable this, you should set your command station breaker delay to its longest setting, while setting the PSX to a shorter one, so it can handle the situation.

 

25 locos all starting up at once is a lot. I have difficulty in understanding the sort of layout where that many locos actually need to be under power all the time. Almost certainly your layout needs dividing into power districts, or at least sub-districts to the command station, each to be protected by a PSX, which come in various multiples. 

 

Even if your layout is a MPD, each siding could perhaps have its own protection. 

  • Agree 2
Link to post
Share on other sites

  • RMweb Gold
7 minutes ago, Oldddudders said:

25 locos all starting up at once is a lot. I have difficulty in understanding the sort of layout where that many locos actually need to be under power all the time.

 

My poor comment, there are 25 locos on the layout but rarely more than 6 actually running their sound function at any time. All locos have sound switched off before I shut down power at the end of a running session.

 

If I inderstand correctly (which is by no means certain) there is an intial in rush of power when first switching on caused by the chips charging their capacitors?  This ocurs irrespective of whether the sound is on or off and it is this spike that causes CC to trip.  I'd guess the power demand would be higher if the sound was on but that is not the case here. What I am trying to calculate is the total rush is in Amps so that I can adjust the PSX settings accordingly. 

 

My layout has no power districts at present but I am open minded to installing them if I can see a clear benefit. Apart from the start up issue which arose when I changed the DCC controller at the start of the year, the layout has run without problems for 4 years with no power districts.  

Link to post
Share on other sites

You could simplify the power districts to using simple toggle switches which allow you to sequentially switch the layout on thus limiting to power surge to a smaller number of locos.

 

simply switch all the switches off before switching the layout on, then switch individual circuits on one by one. You may find that 2 are enough, but 3 nights be better.

 

simple and cheap :)

  • Agree 1
Link to post
Share on other sites

  • RMweb Gold
1 minute ago, WIMorrison said:

You could simplify the power districts to using simple toggle switches which allow you to sequentially switch the layout on thus limiting to power surge to a smaller number of locos.

 

simply switch all the switches off before switching the layout on, then switch individual circuits on one by one. You may find that 2 are enough, but 3 nights be better.

 

simple and cheap :)

I have already tried this, admittedly only 1 DPST switch which created sections that held less locos. The split was about 10 in one section and 15 in the other. The switch made no difference, I suspect that my main fiddle yard is the cause because this is where there are typically as many as 15 locos at any time. The fiddle yard could be divided into sub districts but that would involve a lot of re-routing of wiring which I dont fancy doing if at all possible. If the PSX does what it says then I hope it will resolve the problem. 

 

What is really frustrating is the intermittency, some days I can apply track power on the first push of the start button. Other days can take 5 or 10 minutes and numerous presses of the button.  If I can get an understanding of the various constituent parts that make up the problem, I think I have best chance of designing a solution. That is why I want to know what the maximum start up power draw is of a single Loksound fitted loco is. 

 

 

 

 

 

Link to post
Share on other sites

  • RMweb Premium
Just now, WIMorrison said:

You could potentially reduce the inrush current by increasing the track voltage.

 

12w inrush at 12v is approx 1A whereas at 18v it is only around 0.66A.

 

The actual current is affected by power factor but the principle will still apply.

That doesn't sound right

Increase the voltage and you increase the inrush current.

Link to post
Share on other sites

Even a single loco can take way over 5A, albeit for a very, very brief period.

 

Increasing the track voltage will not help. Capacitors are not simple resistive elements and ohms law does not apply.

 

As above, you may need to split the layout and power up in sections.

 

 

Link to post
Share on other sites

  • RMweb Premium

I've got approx 35 non-sound locos sitting on the layout plus a couple of Diesel railcars with lights.

I'm using a Z21 with a NCE EB-1 set to 3A. It is very unusual for it to trip on power up.

Link to post
Share on other sites

  • RMweb Gold

I appreciate peoples input. I think it is clear that the sensitivity of the ESU Cab Control short circuit system is tuned considerably finer than my previous Guagemaster Prodigy which happily ran the same layout without fuss or similar issues. 

 

In between posts I have been wiring in two PSX 1 circuit breakers running software revision L to create two new sub districts. I appear to have followed the installation instructions correctly because on powering up the layout I was able to apply track power on the second attempt. I then switched everything off, waited 10 seconds and powered up again. This time track power worked on the first attempt. 

 

I am not sure if I have solved my problem with the PSX's. The spurious short has been a regular but intermittent occurence when on previous occassions I thought I had found a solution only to find a few days later that the issue resurfaced. 

 

How long does a capacitor hold on to its charge? By that I mean is my powering off and waiting 10 seconds sufficient time for the power to have discharged back to zero? 

 

 

Link to post
Share on other sites

  • RMweb Gold

I have further tested the PSX's and found that on creating a short by approaching an incorrectly set point, it is the PSX which cuts track power and not the controller. That is a relief, I was uncertain which of the two was set more sensitively. By changing the point to clear the short the auto re-start of the PSX restored power. 

 

At least something seems to work as I want!

Link to post
Share on other sites

  • RMweb Gold
31 minutes ago, young37215 said:

I have further tested the PSX's and found that on creating a short by approaching an incorrectly set point, it is the PSX which cuts track power and not the controller. That is a relief, I was uncertain which of the two was set more sensitively. By changing the point to clear the short the auto re-start of the PSX restored power. 

 

At least something seems to work as I want!

Good. And you now see the advantage of power districts or sub-districts, in that a short on one does not shut down the whole layout. You may find you can use that to advantage. Most shorts caused by running through a wrongly-set point etc tend to occur in shunting etc yards. So separating the main line from the yard can limit the effect such a short might cause. It all contributes to DCC making a layout function more smoothly. 

  • Agree 1
Link to post
Share on other sites

  • RMweb Gold

Absent knowing how long capacitors hold their charge after power is disconnected I decided to switch everything off for 30 minutes and then re-start. Annoyingly it was back to normal, track power took a dozen or more button pushes before connecting. So much for the PSX's adaptive load reset capability! 

 

I have Railcom enabled on my controller which I have read can cause issues with PSX. As a further experiment I will turn this off and see if this any effect of the situation.  

 

 

Link to post
Share on other sites

  • RMweb Gold
47 minutes ago, Grovenor said:

have you got stay alive on these locos? That can significantly increase the start up load.

Another source of problems are Hornby pullmans with lights that need a resistor adding to limit their charge current.

 

Thanks but no to any loco stay alives. I have 2 coaches with lights and stay alives but these live in an electrically isolated siding so have no effect on start up. They  dont get illuminated until I set the route for them to exit their designated siding. 

 

I need to understand what it is that Cab Control measures to determine a short, if I can find this out I have a chance of finding a way around the issue.

 

Link to post
Share on other sites

22 hours ago, young37215 said:

In trying to logically find a solution and because I want to get the correct setting for the adjustable trip level settings on the circuit breaker, I want to understand how much power an ESU sound chipped loco initially draws when power is first applied. CC does not incorporate any means of measuring power demand on start up; can anyone give me an indication?

 

Any guidance gratefully received.

Rob,

 

Have you tried the ESU 'forum'? To quote from the ESU manual:

"16.2. Forum
Since its introduction in late 2006 the ESU support forum has turned into one of the most successful internet platforms. You may ask any questions related to all ESU products in this forum. Our support team endeavours to resolve any issues or problems together with you. Thus everybody will profit from the knowledge off all participants since other users are permitted to provide answers as well.
"

 

They might / should be able to help with the technical side of the ESU.

 

Ian

Link to post
Share on other sites

  • RMweb Gold

Hi Ian

 

The ESU forum is useful but there are more questions on it than answers! I have a thread running regarding this and I have e-mailed technical support at ESU because there are several outstanding questions along similar lines. As others have suggested, I am sure the underlying issue I am facing is a very sensitive setting on the Cab Control short circuit protection. What is measured, how to overcome it and if/how the setting might be adjusted are the sort of information I am after from ESU.

 

I am not persuaded that a booster is the solution to my problem because once track power is on I can run trains freely without any obvious bottlenecks of power or constraints and at £200 for an ESU booster I sincerely hope that I am right. The concept of capacitors loading up power when first connected I can understand, what I am less clear about is why Cab Control treats this as a problem. Hopefully I can get to the bottom of the issue and find a solution, I am impressed with what Cab Control offers and would like to resolve this iritation. 

 

regards Rob

Link to post
Share on other sites

Rob, the ESU Booster (generic DCC term use) is seeing what it percieves as a short circuit on power up, because likely each of the sound locos you have on the layout is drawing a substantial inrush current on start up.  The Capacitators will hold charge depending on draw down.  This can be internal or external draw down- I'd suspect that any caps in use will be fully discharged by the end of 10 minutes, and likely sooner.  

So, when you power up the layout, you have to charge up all the capacitators, and that is what the ESU booster is seeing as a short.  You possibly can change the power on delay on the PSX so that they load sequentially rather than both at the same time.  Otherwise, you will have to further divide the layout. 

There is a fairly easy way to figure out how many sections you need, and that is to remove some locos & put them on the shelf, then turn the layout on.  When you get to where it will turn on without fiddling around, you now have your answer, subtract 1 from that, and then divide your # of locos by the # that the booster will start at the same time.

There are other options- one that springs to mind is to use a Digitrax booster.  They have very slow trip times, in the order of 1/2 a second, which allows for the inrush current to have dissipated- occasionally I will run into a problem, but usually it means that I have a dead short and not anything else.

If you are feeding DCC to accessory decoders, remember that they too will draw inrush current on start up.

 

James

  • Thanks 1
Link to post
Share on other sites

1 hour ago, young37215 said:

The concept of capacitors loading up power when first connected I can understand, what I am less clear about is why Cab Control treats this as a problem. 

 

It's very difficult to distinguish the inrush current at startup from a genuine short circuit. There are two ways to do

- Use a delay and check if the current is decaying as it would in the case of capacitor inrush. The problem is, with modern switch mode power supplies, you can't wait too long as the PSU to the booster would cut out if its a genuine sustained short.

- Use a current limit to force a soft start. This has issues with power dissipation in the booster. I can go into the technical details if you want.

 

  • Thanks 1
Link to post
Share on other sites

  • RMweb Gold

Thanks to everyone for their contributions. I have e-mailed ESU technical support highlighting my problem and asked them to advise what the trigger settings are for the short protection and if they are user changeable. If I can get this information then I can judge what to do next. My issue is very similar to a number of similar ones on the ESU forum so it is clearly a recurring one. Whether ESU are able or minded to do anything is, of course, another matter.

 

I will also see what more I can find regarding PSX's adaptive load reset functionality and if it is customisable. If anyone knows the answer or can explain how ALR works I would appreciate their input here.

 

 

Link to post
Share on other sites

  • RMweb Premium
21 hours ago, Crosland said:

 

It's very difficult to distinguish the inrush current at startup from a genuine short circuit. There are two ways to do

- Use a delay and check if the current is decaying as it would in the case of capacitor inrush. The problem is, with modern switch mode power supplies, you can't wait too long as the PSU to the booster would cut out if its a genuine sustained short.

- Use a current limit to force a soft start. This has issues with power dissipation in the booster. I can go into the technical details if you want.

 

Interesting.

The Z21 uses a SMPS as a power source and I have an NCE EB1 as a track circuit breaker so that some accessories stay powered if there is a track short.

The Z21 never trips on power up, the EB1 very rarely, this is with about 30 non sound locos sitting on the track around the laout, yet the EB1 trips fairly instantly there is an overload. (i.e. no perceptible delay)

So quite clearly some manufacturers can get the balance right.

Link to post
Share on other sites

Keith

 

That is how I used to have my layout wired (until I changed NCE EB1 for Z21 booster) and I never had a single case of the Z21 tripping, it was always the EB1.

 

I have recommend this solution to several people and that have also all found that it works reliably for them.

Link to post
Share on other sites

3 hours ago, young37215 said:

I will also see what more I can find regarding PSX's adaptive load reset functionality and if it is customisable. If anyone knows the answer or can explain how ALR works I would appreciate their input here.

 

I'm not as knowledgeable as other contributors in this thread, but I'm not sure I understand how a PSX helps with the problem that is being described.  My understanding is that a circuit breaker should have a faster response time to a short circuit than the command station booster does, so the PSX that covers the power district in which a short occurs should shut down instead of the command station booster, thus leaving the command station booster able to power the rest of the layout.  That is the reason for including power districts, sub-districts and circuit breakers in the design.

 

My understanding of the PSX is that it is allegedly 'intelligent' enough to distinguish between a true short circuit and a high inrush current on start-up, so if the PSX 'thinks' it detects a short circuit, it will shut down immediately, but if it 'thinks' it has just detected a high inrush current, then it won't.  I have no idea how it actually determines the difference between a true short circuit and a high inrush current and thus 'decides' whether or not to shut down.

 

However, what I think you are describing is that a high inrush current on start up is causing the ESU Mobile Control to shut down.  Therefore, what I would expect to happen would be that the PSX detects the high inrush current, identifies it as such and therefore won't shut down, but the ESU Mobile Control then shuts down instead (ie it acts as though there was no PSX fitted).  My understanding of the situation is therefore that the ESU command station is not as 'intelligent' as the PSX, but has a relatively quick response time (ie possibly faster than some alternative command stations on the market).  I don't see how the 'intelligence' that the PSX uses in deciding not to shut down can be transferred to the ESU command station.  I think the question is therefore whether the ESU command station is configurable in any way (ie make it slower to react to a genuine short circuit).  Making the ESU command station booster slower to react to a genuine short circuit, shouldn't be an issue, as the PSX should shut down in any case, but ESU may not have allowed for this on the basis that many people may use the command station without a circuit breaker and expect it not be damaged.

 

If the ESU command station cut-out time cannot be increased, then I think the only solution is to divide the layout up using switches.  That is, install a switch between the command station booster and each PSX, so that you are only switching on one power district at a time and also include a switch to allow the accessory bus to be turned off as well, which would only be turned on once all your power districts are operational.  If that doesn't solve the problem (because most of the locomotives are in one of the power districts such as a fiddle yard), then there would be a need to isolate some of the roads, so that the inrush current is reduced.  As has been highlighted above, it should be possible to work out how much the layout would need to be divided up by removing all stock and then power the layout up with just one locomotive.  Repeat with two, three, four, etc until the problem first appears and then design the solution around that.

 

I don't think that the fact that others don't have a similar problem with their chosen command station is particularly helpful, as I think the problem is that the ESU Mobile Control must have a faster response time than most other command stations.

  • Agree 1
Link to post
Share on other sites

Sounds like re wiring the FY with dead sections just like DC wiring is the smart move.  A loco with sound or coaches with lights have no value in an FY and if the point motors route the power you don't run the wrong  loco into the wrong end of a wrongly set point and short everything out. 5 amps is a lot of power, and it's the danger of a real short going unnoticed which should be the concern.  

  • Agree 1
Link to post
Share on other sites

Modern boosters have 2 types of protection:

 

1. If the current exceeds the nominal value, it switches off (overload protection).

 

2. If the current suddenly increases within a very short time, it also switches off (short-circuit protection, called dI / dt switch-off). This happens even if the current does not reach the maximum value!

 

Variant 2 is always important when a short occurs at a distant point in the layout and the resistances in the wiring and / or in the track (insufficient track bus) are so high that the rated current is not exceeded because a current of 4.9 A can sometimes make the loco pickups unusable.

 

Unfortunately, at the very first moment of power-up, capacitors cause something similar to a short with current only limited by the resistances in the layout, wiring and decoders.

One solution is to have the capacitors charged in groups one after the other, as has been suggested.

 

Or you can use a soft start device. In the simplest case, this is a resistor between booster and layout that is bridged shortly after the booster is switched on.

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...