!!

Not Registered?

Welcome!  Please register to view all of the new posts and forum boards - some of which are hidden to guests.  After registering and gaining 10 posts you will be able to sell and buy items on our N'porium.

If you have any problems registering, then please check your spam filter before emailing us.  Hotmail users seem to find their emails in the Junk folder.


Thanks for reading,
The NGF Staff.

Author Topic: DCC power supply  (Read 227 times)

0 Members and 1 Guest are viewing this topic.

Offline malc-c

  • Trade Count: (0)
  • Full Member
  • ***
  • Posts: 85
  • Country: gb
    • Awards
DCC power supply
« on: September 11, 2019, 12:14:36 PM »
Hi,

I've built my own controller around the MiniDCC module and booster.  I originally powered this from a Meanwell LPV-60-12 which is a 12v constant voltage DC power supply marketed as a LED driver (data sheet https://www.meanwell.com/Upload/PDF/LPV-60/LPV-60-SPEC.PDF )

I have two loco's, one a DCC ready Dapol 33, and the othre a modified GF 25, both have the same decoders fitted, BACHMANN 36-556RA's

On this supply the loco's didn't appear to run at a reasonable full speed, so I removed it and found an old 3A laptop supply that outputted 18v - Performance is fine, and both loco's ran at more than scale speed.  But it was suggested that the decoder may get burnt out having to deal with a high DCC signal, possibly swinging 30+ volts.

Should I remove the laptop supply and seek a different supply, if so what would be the max "safe" voltage. 

Offline ntpntpntp

  • Trade Count: (0)
  • Full Member
  • ***
  • Posts: 3262
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #1 on: September 11, 2019, 01:18:20 PM »
N gauge is considered nominally a 12V system, though unless you zoom locos round at full tilt or it's a very low geared loco you probably don't get up to the full 12V on analogue DC with normal running.

NMRA DCC spec. says the DCC voltage should be DC + 2V, so around 14V.    Therefore the supply voltage you need depends how your booster generates the DCC output signal, how much voltage loss is taken up by the booster's electronics etc. ?

Taking a well-known commercial system, the NCE PowerCab, as an example: NCE say the output DCC voltage for a 12V system should read around 13.8V AC (so pretty much ties in with NMRA spec), though that is presumably trying to use a standard sinusoidal AC meter on the DCC square wave so only an approximation?  It's interesting that NCE say their power supply for the PowerCab is 13.8V DC.

Decoders are generally supposed to be able to cope with reasonably high peak voltage, I've seen specs saying up to 35V, but check the specs for the decoders you're using? Indeed I've run certain decoders intended for 12V HO on my G scale line at around 18V with no problems, and that is presumably a DCC peak-to-peak of 36V. 

Bottom line is I would think an 18V supply is probably a bit higher than you need, and 14V-ish would be better. What does the supplier of MiniDCC recommend?

« Last Edit: September 11, 2019, 01:20:47 PM by ntpntpntp »
Nick.   2016 celebrating the 20th anniversary of "Königshafen" exhibition layout!

Offline NinOz

  • Trade Count: (0)
  • Full Member
  • ***
  • N Gauge Society Number: Ex3504
  • Posts: 748
  • Country: au
  • Gender: Male
    • Awards
Re: DCC power supply
« Reply #2 on: September 11, 2019, 01:43:51 PM »
so I removed it and found an old 3A laptop supply that outputted 18v - Performance is fine, and both loco's ran at more than scale speed. 
How about dropping the 18V down using a regulator or buck converter to 12V DC.  Lots of cheap dc-dc or ac-dc modules available online.

https://www.ebay.co.uk/itm/LM2596S-DC-DC-Buck-Converter-Adjustable-Power-Supply-Step-Down-Module-5V-12V-24V/254234341174?hash=item3b318c2f36:m:mrCsMGO6qxOowVupHRamMuA
https://www.ebay.co.uk/itm/LED-Display-DC-5V-12V-24v-3A-Step-down-Voltage-Regulator-buck-Module-Converter/282215589884?hash=item41b55c27fc:g:HJIAAOSwh-1W4j-p

The laptop supply is DC or AC?
To be called pompous and arrogant - hell of a come down.
I tried so hard to be snobbish and haughty.

Offline malc-c

  • Trade Count: (0)
  • Full Member
  • ***
  • Posts: 85
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #3 on: September 11, 2019, 02:22:05 PM »
Thanks for the advice and suggestions. 

Both supplies are DC.  For now I'll swap out the 18v supply for the 12v, at least it will fall in line with the standards

Offline jpendle

  • Trade Count: (0)
  • Full Member
  • ***
  • N Gauge Society Number: 23638
  • Posts: 881
  • Country: us
    • Awards
Re: DCC power supply
« Reply #4 on: September 11, 2019, 06:55:42 PM »
This issue has come up a number of times and I believe there are some misconceptions regarding the track voltage needed for N Gauge DCC.

For example my Z21 was set to 18V track voltage out of the box, and I have had no issues with Zimo chips. I have now reduced the track voltage to 15V as I have some Revolution Pendolinos with ESU sound chips fitted and they recommend that track voltage should be no more than 15V.

The 12V recommendation is for N Gauge motors, NOT N Gauge DCC chips. The DCC decoder is regulating the supply to the motor not the DCC controller.

I would recommend a 14 or 15V DCC track voltage.

Regards,

John P
Check out my layout thread.

Contemporary NW (Wigan Wallgate and North Western)

https://www.ngaugeforum.co.uk/SMFN/index.php?topic=39501.msg476247#msg476247

Offline malc-c

  • Trade Count: (0)
  • Full Member
  • ***
  • Posts: 85
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #5 on: September 11, 2019, 08:09:43 PM »
John,

Thanks for the reply.  I must admit on the 18v (unladen) supply the locos respond better, and slow speed is more controllable and responsive.  The loco's are warm where the decoders are fitted, after a run, but not excessive and I can't differentiate between 12v and 18 supplies.

I'll look out for a PSU rated at 15v which should give the best of both worlds, a decent speed range but at a track voltage that is kinder on the decoder.

Offline sprogman

  • Trade Count: (0)
  • Full Member
  • ***
  • N Gauge Society Number: 17786
  • 2mm Association Number: 4557
  • Posts: 374
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #6 on: September 12, 2019, 10:23:58 AM »
found an old 3A laptop supply that outputted 18v - Performance is fine, and both loco's ran at more than scale speed.  But it was suggested that the decoder may get burnt out having to deal with a high DCC signal, possibly swinging 30+ volts.

Another misconception by whoever suggested that. It's like saying the 240V mains is actually 480V.

With an 18V DC supply, from your laptop charger, and ignoring the small losses in the controller and decoder, the track is never more than 18V, not 30V. Half the time it's 18V measured between the two rails, the other half of the time it's -18V between the same two rails. The "differential voltage", the difference between the two, is indeed 36V, but this voltage never appears across any components. The decoder rectifies the +/-18V DCC signal and only ever sees 18V internally.

If you were using an 18V AC transformer and rectifying that for your DC supply then it would be a different story...

PS I agree, 18V is high for N. I recommend 12V supplies for SPROGs with N and have had very few complaints.
« Last Edit: September 12, 2019, 10:25:09 AM by sprogman »

Offline malc-c

  • Trade Count: (0)
  • Full Member
  • ***
  • Posts: 85
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #7 on: September 12, 2019, 11:02:01 AM »
Thanks for the reply.

So because it's a DC supply which gets inverted as part of the DCC signal process, the rails are alternating between 0 - 18v.  So in essence one rail will be -18v the other 0v, then the rail that was -18v becomes 0v and the 0v rail becomes +18v, so the decoder is only ever receiving 18v at any one time ?  So I assume that the decoders don't actually rectify the DCC signal, they simply invert the -18v to +18 so that they are running at +18 all the time ?

I have an IR laser temp gun, and last night left the Dapol loco which has the decoder just under the roof, running for half an hour at a scale speed for an express passenger (based on comparing the sound of wheels over joints in the track with real life sound).  After that the loco was stopped and the temperature over the roof was read.  Max temp was 48.3c - Not sure if this is normal for this loco, but would have thought that this is well within the tolerance set for these devices, and in all honesty in running the layout I doubt that I would ever be running the loco constantly for that length of time.

It may well be that the MiniDCC is not very efficient which is why the speeds are very low when a 12v supply is being used ?   I can't look at the signal on the track as I blew up my USB scope last month... (long story) so can't confirm the actual peak to peak voltage on the track.  Setting a Digital Multimeter to AC volts reads 18.7v across the rails

Offline sprogman

  • Trade Count: (0)
  • Full Member
  • ***
  • N Gauge Society Number: 17786
  • 2mm Association Number: 4557
  • Posts: 374
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #8 on: September 12, 2019, 11:19:57 AM »
Thanks for the reply.

So because it's a DC supply which gets inverted as part of the DCC signal process, the rails are alternating between 0 - 18v.  So in essence one rail will be -18v the other 0v, then the rail that was -18v becomes 0v and the 0v rail becomes +18v, so the decoder is only ever receiving 18v at any one time ?  So I assume that the decoders don't actually rectify the DCC signal, they simply invert the -18v to +18 so that they are running at +18 all the time ?

Not quite. Rails are A and B, say. At the output of the booster, measured against the 0V connection of the booster, rail A switches between 0V and 18V, rail B switches between 18V and 0V (i.e. they are doing the same thing but out of phase). When A is 0V and B is 18V the difference B - A is 18V. During the other half of the cycle the difference is -18V.

Out on the layout we have no 0V reference, all we have is the two rails where the voltage between the two rails switches between 18V and -18V. The decoders do rectify this. No component ever sees more than 18V across it, one way or the other.

Andrew

Offline sprogman

  • Trade Count: (0)
  • Full Member
  • ***
  • N Gauge Society Number: 17786
  • 2mm Association Number: 4557
  • Posts: 374
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #9 on: September 12, 2019, 11:25:15 AM »
I can't look at the signal on the track as I blew up my USB scope last month...

Neither of the rails ever actually switches to 0V due to the components in the booster. Never connect a single 'scope probe across the rails unless:
1. You know it's completely floating, e.g., connected to a laptop on battery power with no mains connection, or
2. you have isolated the ground connection to the 'scope (potentially dangerous in other situations).

To be sure you should use a differential probe. A quick and dirty approximation is to use two probes with the tips connected to the rails and their ground clips connected together (and to nothing else). Use the 'scope math function to subtract one from the other.

If you are close enough to the booster then you can connect each probe across one rail and the booster 0V ("home ground" as its known on some systems) and use the 'scope math function.

Offline malc-c

  • Trade Count: (0)
  • Full Member
  • ***
  • Posts: 85
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #10 on: September 12, 2019, 11:44:42 AM »
Andrew, I found that out the hard way.

Working on a totally different project, with the scope connected to the PC I probed what I thought was the gate of a MOSFT but it was the drain, and the voltage was rater high.  Out went the power as the panel was tripped....everything came back up bar the two year old PC that I had built.... Diagnosed the issue was the motherboard, so it cots me dearly.  I also wanted to see if the Hantek scope still worked, plugged that into an old Asus 900A netbook and it killed that too !! - Both went into the bin !!

I'm now a lot more careful when building and testing circuits.

I've been trying to find the specifications on the decoder to see if the voltages are within specification for it, but it's proving difficult.

Offline Nigel Cliffe

  • Trade Count: (0)
  • Full Member
  • ***
  • Posts: 656
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #11 on: September 12, 2019, 01:24:19 PM »
Back at track voltages.....

In my experience, the heating effect of higher track voltages can cause problems.   The worst I've seen was in 2010 at the 2mm Scale Association 50th anniversary show.  The room air-con was off for Sunday, and it got very hot inside.  Those running smaller locos on higher track voltages experienced a lot of run-aways.  Certain decoders were worse than others, but the big common factor was track voltage. 


Separate issue, quite a few people are now fitting stay-alive circuits into their locos, often with components which are on-the-limit at 16vDC (after the decoder rectifier stage).   Some of those are home-brew, some commercially supplied.  Allowing any tolerance on the components means going a bit lower on the track voltage. 


I'd run N at a DCC track voltage of around 13-14v.   Higher invites problems.


Nigel

Offline malc-c

  • Trade Count: (0)
  • Full Member
  • ***
  • Posts: 85
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #12 on: September 12, 2019, 01:37:22 PM »
Hi Nigel,

Thanks for the comments.  I'm going to look for a 14 or 15v DC supply, which should give me a the running speed I want but with less chance of issues anticipated at the 18v I'm currently using, and hopefully produce less heat in the loco

Offline themadhippy

  • Trade Count: (0)
  • Full Member
  • ***
  • Posts: 711
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #13 on: September 12, 2019, 03:07:34 PM »
If youve got 18v DC available why not chuck a low drop out 15v regulator on the output, LM2940CT-15 is good for 1A and works down to 16.75v

Offline malc-c

  • Trade Count: (0)
  • Full Member
  • ***
  • Posts: 85
  • Country: gb
    • Awards
Re: DCC power supply
« Reply #14 on: September 12, 2019, 03:26:12 PM »
If youve got 18v DC available why not chuck a low drop out 15v regulator on the output, LM2940CT-15 is good for 1A and works down to 16.75v

That's one idea, but I would have though 1A is low for a DCC supply.  Most seem 3A - 5A rated.  I guess a DC-DC converter as suggested above might be an alternative option

 

Please Support Us!
September Goal: £55.00
Due Date: Sep 30
Total Receipts: £30.00
Below Goal: £25.00
Site Currency: GBP
55% 
September Donations

anything