DCC power consumption and when does enough become enough for the system

[Dungrange]

18 minutes ago, NIK said:

Nominal Voltage limit not nominal voltage.

 

Table 2.3 says Minimum - 8.5V, Nominal - 15V, Maximum - 22V.  I'd therefore read 15V as mid-range value, not a maximum.

[Bucoops]

1 hour ago, NIK said:

Nominal Voltage limit not nominal voltage.

 

Max voltage is quoted as 22v. Min 8.5v.

[ATME]

So does this mean that all decoders should work in the range 8.5v to 22v despite being informed by Accurascale that ESU v5 decoders require a minimum of 14v?

Edited January 22 by ATME

[melmerby]

51 minutes ago, Dungrange said:

 

Table 2.3 says Minimum - 8.5V, Nominal - 15V, Maximum - 22V.  I'd therefore read 15V as mid-range value, not a maximum.

Z21 claims adjustable from 12 - 24v but then states the maximum input voltage is 24 and the track is about 1v less!

That makes it 11 - 23v (?)

[Bucoops]

9 minutes ago, ATME said:

So does this mean that all decoders should work in the range 8.5v to 22v despite being informed by Accurascale that ESU v5 decoders require a minimum of 14v?

The spec I linked to says a decoder should be able to interpret (so by omission, not actually work) a packet received at the minimum voltage, and it should be able to operate without any damage when the max voltage is received, so no.

[NIK]

3 minutes ago, ATME said:

So does this mean that all decoders should work in the range 8.5v to 22v despite being informed by Accurascale that ESU v5 decoders require a minimum of 14v?

Hi,

 

The NMRA document says peak voltage decoder should operate in and withstand is 7 Volts to 27 Volts.

 

I assume the 7 volts is to allow for voltage drop due to resistance in the wiring and rails.

I assume the 27 volts is to allow for voltage spikes generated external to the command station.

I have yet to find anything in DCC decoder manuals about NMRA voltage compliance.

 

Regards

 

Nik

[Trofimow]

What a fascinating and informative thread - which goes a long way to explaining the performance issues I have been seeing with some locos and why my sole stay alive fitted loco behaves strangely. RM Web at its best.

 

I run Lenz with 10 amps available power, normal load is around 7.5 amps.

 

The system is fed from Lenz TR 150 transformers which provide 15 volts into the command station and booster.

I am seeing 13.8 volts out of the command station, even after maxing out the voltage adjustment CVs.

 

By the time it has gone through the PSX breakers and LDT occupancy detectors, there is only 12.4 volts at the rail.

The loco recoder will reduce the voltage available at the motor even further.

Measurement is done with a RRAmmeter, Alpha meter and my digital multimeter, which all agree.

 

Looks like I need to replace the Lenz transformers with something providing a minimum of 18 volts.

 

Must check what the spec is for the maximum input voltage to the command station and booster.

[Bishdurham]

I can understand why my mate plays with Hornby Dublo and Triang!  Then again he can  turn the air blue when talking about Super 4 Track!

[MikeParkin65]

14 hours ago, ATME said:

So does this mean that all decoders should work in the range 8.5v to 22v despite being informed by Accurascale that ESU v5 decoders require a minimum of 14v?

The decoders dont need 14V but for the stay alive to be active they need at least that. Personally I think the pragmatic answer would be to remove the stay alive  - with live frogs a large heavy multi wheeled loco doesnt need it. In fact with well laid and clean track only the smallest of locos needs it. 

[McC]

3 minutes ago, MikeParkin65 said:

The decoders dont need 14V but for the stay alive to be active they need at least that. Personally I think the pragmatic answer would be to remove the stay alive  - with live frogs a large heavy multi wheeled loco doesnt need it. In fact with well laid and clean track only the smallest of locos needs it. 

 

If there is insufficient voltage, the stay alive wont' charge, but it also doesn't impact running or 'get in the way' and we'd not recommend cutting capacitors off the main board which -will- void your warranty :) 

[MikeParkin65]

12 minutes ago, McC said:

 

If there is insufficient voltage, the stay alive wont' charge, but it also doesn't impact running or 'get in the way' and we'd not recommend cutting capacitors off the main board which -will- void your warranty :) 

Thanks - haven't looked at doing this but was assuming the stay alive was a plug in as per SLW. Appreciate if its soldered to the board thats a different proposition. 

[57xx]

4 hours ago, McC said:

 

If there is insufficient voltage, the stay alive wont' charge, but it also doesn't impact running or 'get in the way' and we'd not recommend cutting capacitors off the main board which -will- void your warranty :) 

 

Out of curiosity (I don't do DCC) what is in circuit to stop the capacitor charging at lower voltages? Is there some funky switching circuit across it with voltage detection? A capacitor will normally  charge to what ever voltage is present across it's terminals.

[SHMD]

1 hour ago, 57xx said:

 

Out of curiosity (I don't do DCC) what is in circuit to stop the capacitor charging at lower voltages? Is there some funky switching circuit across it with voltage detection? A capacitor will normally  charge to what ever voltage is present across it's terminals.

 

The "Stay Alive" Capacitor will always be at a lower voltage than the "track" voltage.

 

Here is a "simple" circuit for a "Stay Alive" circuit for a decoder...

 

As can be seen, the "Stay Alive" Capacitor, C1, has two diodes involved in its' charging, through R1. (Either D1/D4 or D2/D3 depending upon the DCC polarity that instant.)

Then, there is a further diode, D5, involved in the discharging/supplying of jolts if the (DCC) track supply is interrupted.

Each diode will "drop" between 0.6 and 1 volt - depending upon diode selection/specification.

 

So, if your track DCC voltage is 12.4v, then the very best the "Stay Alive" capacitor can "replace" is...

12.4 - 0.6 - 0.6 - 0.6 = 10.6 volts!

(Of course, in this scenario, the decoder is only getting 12.4 - 0.6 - 0.6 = 11.2 volts with perfect track/wheels!)

 

Of course, other "Stay Alive" charging/discharging circuits can be used (MOSFETS?) with much lower resistances but they quickly become very complicated/expensive and (can) suffer other "unforeseen" problems/behaviour(s).

 

Conclusion - Pump up the volume/I mean voltage!

 

 

Kev.

[NIK]

Hi,

 

There are some stay alive circuits that charge the capacitors to a lower voltage than the typical track voltage and then step up the voltage to that the decoder requires when there is a pickup problem between track and loco.

 

This could be done so that large capacity supercapacitors can be used which inherently have a low maximum voltage. These type of circuits tend to be more expensive.

 

The Hornby HM7000 series power pack is probably of this type (the step up is probably done on the decoder).

 

If loco decoders have Back EMF feedback operating and working well it is possible for low voltages at the track to be disguised until a DCC speed step is requested by the operator that is high enough that the motor cannot make the loco/train go faster as it simply isn't being supplied with enough voltage.

 

Maybe I will make a DCC track voltage monitoring vehicle (with radio link for the voltage data to be towed behind a loco).

 

Easier would be a power resistor connected by wires to crocodile clips clipped to the track and then a DCC voltage meter in parallel.

The resistance value would be that to take the max current the power district is rated for without tripping. If the voltage is <7V then that part of the layout may be out of spec for HO. If the voltage is <13.4 Volts then locos will not be able to achieve the max speed they would do on 12V DC (pre decoder fitting).

 

Regards

 

Nik