Techie question on lighting voltage.....?
- Thread starter Zerogee
- Start date
Zerogee
Clencher's Bogleman
Well, that would work, but then there would be no DCC control over the lights when operating on battery power.....
It would still require a 5v voltage stabiliser from the battery supply, so it really wouldn't save much.
Jon.
GAP
G Scale Model Trains, 1:1 Sugar Cane Trains
Well, that would work, but then there would be no DCC control over the lights when operating on battery power.....
It would still require a 5v voltage stabiliser from the battery supply, so it really wouldn't save much.
Jon.
Instead of focusing on the output have you considered looking at the input specifically the battery input.
Either get a bigger battery (probably not practical) or use a voltage booster for the battery input then you just treat the battery input the same as the DC input.
Simply connect it between the battery and the controlling device.
http://www.ebay.com.au/itm/DC-DC-Ad...252812?hash=item1c85f1120c:g:RWwAAOSw-3FZHqqM
Zerogee
Clencher's Bogleman
Well, that seems a perfectly reasonable way of doing it, at least in theory - maybe I'll try that for the Mk.II install (as I already have two Tam Valley DRS modules, I plan to convert two locos eventually), as I've already got the 5v regulators (ordered a second one which arrived today) I'll stick with Plan A for this one and see how well it all works. As you can't get something for nothing in this universe, I assume that using a voltage step-up unit must have the effect of drawing more amps from the battery, thus reducing the run time?
Jon.
Correct.. You don't get owt for nowt..
You could use a single regulator.. Then a relay to switch to front / rear lights..
You could use the 'Interior' light output of the decoder to feed the relay common, and the 'rear' light feed to fire the relay.. - Off (rear) would mean nc contact feeds front lights. On (rear) would feed to no contact, and rear light.
As 'interior' light feeds lighting power, the lighting key (DCC) would still turn the lights off.
Possibly saves you a bit of wiring, and a regulator board?
Zerogee
Clencher's Bogleman
Correct.. You don't get owt for nowt..
You could use a single regulator.. Then a relay to switch to front / rear lights..
You could use the 'Interior' light output of the decoder to feed the relay common, and the 'rear' light feed to fire the relay.. - Off (rear) would mean nc contact feeds front lights. On (rear) would feed to no contact, and rear light.
As 'interior' light feeds lighting power, the lighting key (DCC) would still turn the lights off.
Possibly saves you a bit of wiring, and a regulator board?
All these ideas would probably work perfectly well..... but it all seems to be getting 'orribly over-complicated..... the two cheap regulator boards simply spliced into the feeds for the front and rear lights will do the job for me just fine, I reckon!
Thanks for all the ideas, though....
Jon.
D
Deleted member 1986
Guest
the two cheap regulator boards simply spliced into the feeds for the front and rear lights will do the job for me just fine, I reckon!
Hmm....not quite, slight flaw in the application......... just tested the theory out in practice!
The two capacitors on the regulator board will at the very least destroy the Function Output and at worst could very well destroy the decoder.
Explanation, when the Function Output is turned off, the two on board capacitors will still be fully charged, they will discharge, and any residue voltage they are holding will flow straight back into the Function Output which has insufficient protection against voltage flowing into it!, the Function Output is designed primarily for voltage to flow out of it...............
Four 1N5819 schottky diodes will be needed, (the 1N40xx types which are already on the board cannot react fast enough and any further IN40xx types will each consume a further 0.7 volts adding another four will lose another 2.8 volts which will be a waste) on both the feed and the Function Output this will then protect and prevent "flyback" voltages from the capacitors on the voltage regulator discharging back to the decoder................
Just used one of these, to provide a 5v regulated output to drive a Relay from a Function Output, they are rated at 100mA, three 5V LGB Bulbs equals 99mA, the margin is a tad tight!
http://www.ebay.co.uk/itm/2-x-78L05...802256?hash=item1e99d3a110:g:CrEAAOSwgQ9VyDic
OR these..............bit of larger margin.........
http://www.ebay.co.uk/itm/5Pcs-Volt...597043?hash=item4d4dd9bcb3:g:u-EAAOSwEeFVJ5Z4
Last edited by a moderator:
Zerogee
Clencher's Bogleman
Thanks Arthur, that's the kind of odd little thing that I would only have found out by (expensive!) trial and error.......! Hmmm, the problem of Chinese electronics with no instructions, I guess.....
So the schottky pair on both wires from the decoder to the regulator board (one on each wire, in opposing polarities I assume?) should avoid any nasty capacitor-related problems?
Guess I will need to pick up four of them on Monday, the advantage of having the main warehouse of Rapid just a mile down the road!
Well, I did say this whole installation was a test-bed project.....
Jon.
So the schottky pair on both wires from the decoder to the regulator board (one on each wire, in opposing polarities I assume?) should avoid any nasty capacitor-related problems?
Guess I will need to pick up four of them on Monday, the advantage of having the main warehouse of Rapid just a mile down the road!
Well, I did say this whole installation was a test-bed project.....
Jon.
D
Deleted member 1986
Guest
Had to think for a minute, only ever had to fit on the POS feed previously, silver band on diode for input that is POS towards regulator board, for NEG band away from regulator board, allows voltage to flow from the feed and Function Output to the regulator, but NOT to flow back from the capacitors when they discharge.
Guess safest way is mock up a test rig to ensure they are the right way round, and does what it all should do, before connecting to the decoder!
EDIT: Mocked up a Test Rig............
Wouldn't be much of a challenge if you'd chosen a simpler project!
Guess safest way is mock up a test rig to ensure they are the right way round, and does what it all should do, before connecting to the decoder!
EDIT: Mocked up a Test Rig............
Wouldn't be much of a challenge if you'd chosen a simpler project!
Last edited by a moderator:
Neil Robinson
Registered
Sorry to hijack this thread, I tried but failed to send Arthur a message.
I've been following his contributions to this thread with interest, especially regarding Schottky diodes
I've used cheapo Chinese LM2596 based voltage regulator boards for various applications on analogue for a fair while now.
However I'm new to DCC and I read the advice regarding the use of Schottky diodes on the output of decoders with boards such as
http://www.ebay.co.uk/itm/272161880902?_trksid=p2055119.m1438.l2649&ssPageName=STRK:MEBIDX:IT
For LED coach lighting on analogue I use one on each coach from the track via a KBP 203G or similar bridge and adjust to a suitable voltage. Am I safe in assuming that this arrangement will be fine for DCC as the regulators would be connected via the rails to the decoder input? Hopefully this input should be protected from nasty spikes due to poor current pickup etc.
My other query concerns the Seuthe 5V smoke units as supplied as standard by LGB when used with ESU decoders.
The ESU decoders don't appear to have a "dimming" feature for the relatively high power decoder function outputs. On refection they may well have, but I've yet to master many of their features. I was hoping to retain the 5V smoke units for occasional modest speed analogue use, at least until they give up the ghost, and use a LM2596 board to drop the voltage. Would a single MBR 160 diode or similar be desirable in series with the input? Alternatively I could avoid a current inrush from the decoder by taking power from the track as with coach lighting (if this is O.K. with DCC) and using a relay with its coil on the decoder output for switching.
I've been following his contributions to this thread with interest, especially regarding Schottky diodes
I've used cheapo Chinese LM2596 based voltage regulator boards for various applications on analogue for a fair while now.
However I'm new to DCC and I read the advice regarding the use of Schottky diodes on the output of decoders with boards such as
http://www.ebay.co.uk/itm/272161880902?_trksid=p2055119.m1438.l2649&ssPageName=STRK:MEBIDX:IT
For LED coach lighting on analogue I use one on each coach from the track via a KBP 203G or similar bridge and adjust to a suitable voltage. Am I safe in assuming that this arrangement will be fine for DCC as the regulators would be connected via the rails to the decoder input? Hopefully this input should be protected from nasty spikes due to poor current pickup etc.
My other query concerns the Seuthe 5V smoke units as supplied as standard by LGB when used with ESU decoders.
The ESU decoders don't appear to have a "dimming" feature for the relatively high power decoder function outputs. On refection they may well have, but I've yet to master many of their features. I was hoping to retain the 5V smoke units for occasional modest speed analogue use, at least until they give up the ghost, and use a LM2596 board to drop the voltage. Would a single MBR 160 diode or similar be desirable in series with the input? Alternatively I could avoid a current inrush from the decoder by taking power from the track as with coach lighting (if this is O.K. with DCC) and using a relay with its coil on the decoder output for switching.
Last edited:
D
Deleted member 1986
Guest
In answer to Neil's post above, which is relevant to providing solutions to Jon's original post at the start of this thread.
Voltage regulator boards, picture of the four types I use regularly, should add I've never directly connected them to a Decoder!
Tend to take a very cautious approach with regards to using these, prefered option is to always use a Relay................
Using a 78L05 or 78L12 has never caused an issue when connected directly to a Function Output.
Of the four, whilst they will each in their own right if connected directly to a decoder Function Output, will destroy the Function Output unless some form of protection diode is used, preferably a 1N5819 schottky diode, as it responds instantaneously to prevent capacitor discharge flowing back into the Function Output.
The little one in the "middle", suffers from "Napoleon Syndrome" and will destroy a decoder in less time than it takes to blink!
That's not to say you can't use them. Perfectly safe to use as in this picture, two of them, set at different output voltages, going through a pair of relay contacts.
IN5819 protecting the + Output of the Bridge Rectifier, 4 way screw header, one pair for track INPUTS, other pair for connection to track inputs on a decoder.
Relay Boards, top one for low power Function Outputs, 5v Non Latching DPDT, bottom pair for conventional Function Outputs, 5v Latching DPDT Relays.
Top picture 5v regulator circuit to power an Arduino Decoder, middle DIY Stay Alive, bottom Supercapacitor Constant Coach Lighting.
With reference to the ESU Decoder, had a quick look at the manual, appears the configuration for Function Outputs is very similar to the LGB 55028 mfx/DCC decoder, snapshot of the decoder template I designed using the Massoth Service Tool program, perhaps better to create a new thread on this particular subject.
Voltage regulator boards, picture of the four types I use regularly, should add I've never directly connected them to a Decoder!
Tend to take a very cautious approach with regards to using these, prefered option is to always use a Relay................
Using a 78L05 or 78L12 has never caused an issue when connected directly to a Function Output.
Of the four, whilst they will each in their own right if connected directly to a decoder Function Output, will destroy the Function Output unless some form of protection diode is used, preferably a 1N5819 schottky diode, as it responds instantaneously to prevent capacitor discharge flowing back into the Function Output.
The little one in the "middle", suffers from "Napoleon Syndrome" and will destroy a decoder in less time than it takes to blink!
That's not to say you can't use them. Perfectly safe to use as in this picture, two of them, set at different output voltages, going through a pair of relay contacts.
IN5819 protecting the + Output of the Bridge Rectifier, 4 way screw header, one pair for track INPUTS, other pair for connection to track inputs on a decoder.
Relay Boards, top one for low power Function Outputs, 5v Non Latching DPDT, bottom pair for conventional Function Outputs, 5v Latching DPDT Relays.
Top picture 5v regulator circuit to power an Arduino Decoder, middle DIY Stay Alive, bottom Supercapacitor Constant Coach Lighting.
With reference to the ESU Decoder, had a quick look at the manual, appears the configuration for Function Outputs is very similar to the LGB 55028 mfx/DCC decoder, snapshot of the decoder template I designed using the Massoth Service Tool program, perhaps better to create a new thread on this particular subject.
Last edited by a moderator:
Zerogee
Clencher's Bogleman
Thanks for all the input, Arthur, even if some of the stuff in your last couple of posts went way over my head.....
Getting hold of the 1N5819 Schottkys tomorrow will be no problem, Rapid are showing plenty in stock and a pack of five of them will cost me just over a quid, plus a twenty-minute pop round to their trade counter!
Can you humour me and just clarify, in words of few syllables (or even a wiring sketch), exactly how I need to connect them between the XLS decoder light outputs (Li-V/Li-H and the +24v common) and the two (unlabelled) screw terminals on the input side of the voltage stabiliser board....?
Thanks in advance!
Jon.
Getting hold of the 1N5819 Schottkys tomorrow will be no problem, Rapid are showing plenty in stock and a pack of five of them will cost me just over a quid, plus a twenty-minute pop round to their trade counter!
Can you humour me and just clarify, in words of few syllables (or even a wiring sketch), exactly how I need to connect them between the XLS decoder light outputs (Li-V/Li-H and the +24v common) and the two (unlabelled) screw terminals on the input side of the voltage stabiliser board....?
Thanks in advance!
Jon.
D
Deleted member 1986
Guest
Tested and checked and then rechecked, when power is off, residue voltage dissipates instantaneously... (thank goodness!!!)
Note orientation of silver bands on diodes, POS, band towards regulator board, Function (LI-V) NEG band towards decoder.....
Repeat as above but for LI-H and +24...........
Continue to connect up the remaining bulb holders, if they are all grouped together, would make the connections as close as possible, then only case of running one of each POS and NEG back to the regulator board.
Bulb holders, they have a Flat Side, suggest this is used as a reference for Positive connections, the bulbs themselves don't care, but as you have to connect them all up, may save some head scratching in the future, plus if ever change to LED's half the job is already done!!!!!!
In order of preference, would first connect the bulb holders up, into the two groups (LI-V and LI-H), then connect to the regulator board(s), connect diodes, then connect to the decoder, check, then check again, test as the installation proceeds, makes finding a fault a darn sight easier!
CV 50.....this by default will be set to 32, may wish to lower this, when being on track power, conserving voltage is not really an issue, on battery power need to go Green!
The voltage regulators are linear in operation, any excess voltage over and above 7.5 volts will be turned into heat!
With two of them it's going to get warm , plus running time is going to be drastically reduced....
If I think of anything further...............
Zerogee
Clencher's Bogleman
Thanks Arthur, that's great! Very much appreciated.....
For the lighting connections I'll be using the existing LGB wiring for each group of 3 bulbs front and rear, so that shouldn't be a problem - there will be just two wires from each set of lamps to the voltage regulators. Very good suggestion about turning CV50 down a bit, if I drop it to (say) 18, that will deliver 18/32 of the full voltage from the outputs - so on track power (assuming 24v) that will be 13.5v, and on the battery it will be 8.3v - that will leave only a small amount of excess to dissipate when dropping down to 5v, if of course my maths is correct here?
Jon.
For the lighting connections I'll be using the existing LGB wiring for each group of 3 bulbs front and rear, so that shouldn't be a problem - there will be just two wires from each set of lamps to the voltage regulators. Very good suggestion about turning CV50 down a bit, if I drop it to (say) 18, that will deliver 18/32 of the full voltage from the outputs - so on track power (assuming 24v) that will be 13.5v, and on the battery it will be 8.3v - that will leave only a small amount of excess to dissipate when dropping down to 5v, if of course my maths is correct here?
Jon.
GAP
G Scale Model Trains, 1:1 Sugar Cane Trains
Thanks Arthur, that's great! Very much appreciated.....
For the lighting connections I'll be using the existing LGB wiring for each group of 3 bulbs front and rear, so that shouldn't be a problem - there will be just two wires from each set of lamps to the voltage regulators. Very good suggestion about turning CV50 down a bit, if I drop it to (say) 18, that will deliver 18/32 of the full voltage from the outputs - so on track power (assuming 24v) that will be 13.5v, and on the battery it will be 8.3v - that will leave only a small amount of excess to dissipate when dropping down to 5v, if of course my maths is correct here?
Jon.
So when running on battery you are going to accept degraded running speeds etc, 24V vs 14.8V?
Why not use a larger battery to increase voltage to close to 24V then all inputs are equal, much simpler that all that mucking about with regulators etc which are just power wasters anyway.
Zerogee
Clencher's Bogleman
Cost and size of the battery are the answers to that.
I don't think that lower running speeds will be a problem at all - I'd never be running at anything like the top speed that 24 volts could deliver, so on battery the only appreciable difference will be turning it up to a higher speed step to get a similar overall speed..... it may be that on battery I have to notch it up to (say) speed step 20 (of 28) to get the same actual speed that step 14 or 15 would give me on track power, but I don't see that as any sort of problem.
Jon.
GAP
G Scale Model Trains, 1:1 Sugar Cane Trains
Then maybe limit the input voltage to the same as the battery (suggested previously) and set the loco up to run like that, basically standardise the input to the electronics regardless in input mode.
Zerogee
Clencher's Bogleman
A very good point, Phil - which is why it's so useful to have these discussions here on the forum where different folks with different bits of knowledge and experience can all chip in!
With Arthur's timely and invaluable (well, about £165 worth of value if you want to put a figure to it!) help and advice about the protection diodes for the lighting voltage boards, I think that the theoretical part of the project is now pretty much settled - more news as I start to get everything assembled and then hopefully up and running!
Jon.
D
Deleted member 1986
Guest
Refer to post 31, first picture, bottom unit, at best it's a DC to DC Buck converter, voltage overhead of 5 volts to start, to supply an extra 3 volts at the top end, bit of a non starter, same rules apply about protection diodes if connecting to a decoder.
Correct.. You don't get owt for nowt..
You could use a single regulator.. Then a relay to switch to front / rear lights..
You could use the 'Interior' light output of the decoder to feed the relay common, and the 'rear' light feed to fire the relay.. - Off (rear) would mean nc contact feeds front lights. On (rear) would feed to no contact, and rear light.
As 'interior' light feeds lighting power, the lighting key (DCC) would still turn the lights off.
Possibly saves you a bit of wiring, and a regulator board?
Total and utter cobblers, drew this out on paper, most of the information is missing, for instance just how does the voltage regulator board fit into this scenario?, waste of space as a solution.
Refer to post 31, pictures two and three are working solutions!
So where does this mythical 24 volts emanate from?
Massoth 1200T provides 24.23 volts to the input of the 1210Z, at best 21.62 volts is available on the track.
Even a cursory measurement on a Massoth XL Decoder between GND and 22V on the decoder will show about 21.24 volts, the Function Outputs are slightly lower at 20.69 volts.
Using the Massoth Formula of 32 over 24 volts will provide an indication, for accuracy the 32 should be divided over the voltage provided from the Function Output.
Same applies to the formula of 255 for speed steps, to ensure accuracy, and more even distribution for motor control.
Nope, not a challenge at all, very simple implementation, not sure why this would be relevant or even needed for this particular scenario.
Diode strings 4 x 4 total of 16, 8 only ever conducting at any one time, 8 x 0.7 = 5.6 voltage drop.
Whilst a solution has been provided using the voltage regulator boards, my prefered option would always be a 78L05 type of voltage regulator, low voltage overhead, less of a heat issue, reduced risk of destroying the decoder, this one is available from Maplin, would do this particular job...
http://www.maplin.co.uk/p/ts2950ct-5v-150ma-ultra-low-dropout-positive-voltage-regulator-n69ca