How to run engine on battery but control by DCS

GGG posted:

Why limit your self to the DSCRC.  When you can get the full DCS installed.  G

Likely the $200 price tag for a TIU... But it does work.  

I'm wondering if it might be possible to more or less hack the output signal of a TIU to transmit the needed code.  Perhaps someone with a scope could see what frequency the data is sent at, and how it is formed.  It may be possible to push over radio with fairly inexpensive parts.  Without knowing what the track signal actually looks like it's hard to say for sure.  

PaulB posted:

 

I have a question on your battery voltage. Will smaller size batteries work with this set-up. I currently have 14 and 17 installed. Is the 22.v a better choice to power the TIU and an MTH Big Boy?

It's not the voltage per se.  What's important is the amount of stored Energy which is measured in Watt-Hours.  Casey Jones2 is using 22.2V battery with 10 Amp-Hr capacity.  That's 222 Watt-Hours.  So if the engine is consuming 10 Watts, it can run for ~22 Hours.  If consuming 100 Watts, it can run for ~2.2 Hours.  Etc.   A DCS engine at idle might draw 10 Watts.  A G-gauge engine pulling a bunch of cars going up a grade can hit 100 Watts.  Turning on smoke can add ~10 Watts in G-gauge.  And so on.

I figure his 22.2V battery is a stack of six 3.7V LiPo batteries (3.7V x 6 = 22.2V).  Since the market for battery-powered G-gauge trains is a drop in the ocean compared to laptop or similar battery high-volume consumer applications, I think you must resign yourself to what's readily available.  For example, it seems many battery-powered tools use 18V, 20V or whatever...and come with a mating battery charger - makes charging/swapping battery packs that much easier.

So whatever battery voltage you use, I'd suggest the highest Amp-Hour rating.  Practically speaking, which means what you'll find readily available, you should choose the highest DC voltage that the DCS engine will tolerate (I would NOT go much more than 22.2V) and the highest Amp-Hour rating that will fit in your space available. 

Don't go too low in voltage because this will limit your max speed (independent of the Amp-Hour rating).  A 14V battery pack might limit your top speed to less than 50 sMPH (depending on load).

Thanks for the info. It is very helpful. 

Paul

JohnGaltLine posted:

...

I'm wondering if it might be possible to more or less hack the output signal of a TIU to transmit the needed code.  Perhaps someone with a scope could see what frequency the data is sent at, and how it is formed.  It may be possible to push over radio with fairly inexpensive parts.  Without knowing what the track signal actually looks like it's hard to say for sure.  

I'd say member Adrian! has done the most hands-on reverse-engineering of the DCS track-signal.  He's published extensively such as:

https://ogrforum.com/...-signal-not-constant

https://ogrforum.com/...ck-signalling-format

The bandwidth of the DCS on-track signal is measured in the Megabit rate vs. the Kilobit rate of the IR DCSRC handheld -> receiver or the RF DCS remote -> TIU.  This high data rate makes it tricky to co-opt low-cost RF datalink modules.  But to your point, I'd think it more productive to "hack" the relatively slow-speed (kilobit) data link between the DCS handheld-remote and the TIU.  I believe this is at 900 MHz RF for which there are low-cost data modules on eBay.  Obviously building such a decoder would then eliminate the need for the TIU in the moving engine.  The data processing could probably be handled with an Arduino.  Good idea but to quote GRJ...

Nothing is so easy as the job you imagine someone else doing.

stan2004 posted:

A G-gauge engine pulling a bunch of cars going up a grade can hit 100 Watts.

Turning on smoke can add ~10 Watts in G-gauge. ….

 (I would NOT go much more than 22.2V) 

I saw that an average MTH one gauge engine's average draw is fairly low, between 1-2 amps on my Bridgewerks meter, closer to 1. With more engines added the draw doesn't go up fully the way I had expected it to. I ran seven diesels for example and the draw was far less than 4 amps. The meter varied as the train went around the yard (up and down hills). When I turn on the smoke, especially with more than one engine, the amps go up pretty well. Not double but more than you posted. With the seven engine example the total amps went up maybe to between 5 and 6? Still less than I expected.

 I guess it's dependent on what voltage we are talking about? I set my track at an optimum of 21+ volts (about 22 on the meter). That allows fast running, lower amp draws on that meter, and covers any of the power dips in areas. I would guess at lower voltages the draw could be higher?

 Almost all of the battery guys who post, don't use smoke. That maybe a combination of consumption, and the equipment they use. Some more basic older decoders don't do smoke well. It would be interesting to read how many MTH people use battery and also which of those use the smoke units. With the features built into the MTH decoders and engines, it seemed an obvious choice to me to use their system and track power. The economy multiplied when I acquired many more engines. I read of others having to buy many wireless type boards and other equipment in each engine, let alone having to replace the batteries that have limited charging cycles. My expense was more upfront, and is really small since. I can run smoke on many engines and only need to stop to add smoke oil.

 All that said and I still look at battery, as well as DCC equipment constantly as I know the technology advances. I have many cordless tools and enjoy the freedom. I don't enjoy replacing them every few years as they won't perform anymore. I think many battery advocates, maybe don't run the bigger trains for longer hours regularly, or maybe just don't post of their real expense?

 How many out there have had to replace a cordless tool because the battery wouldn't hold a charge and just wasn't worth replacing?

stan2004 posted:

I'd say member Adrian! has done the most hands-on reverse-engineering of the DCS track-signal.  He's published extensively such as:

https://ogrforum.com/...-signal-not-constant

https://ogrforum.com/...ck-signalling-format

The bandwidth of the DCS on-track signal is measured in the Megabit rate vs. the Kilobit rate of the IR DCSRC handheld -> receiver or the RF DCS remote -> TIU.  This high data rate makes it tricky to co-opt low-cost RF datalink modules.  But to your point, I'd think it more productive to "hack" the relatively slow-speed (kilobit) data link between the DCS handheld-remote and the TIU.  I believe this is at 900 MHz RF for which there are low-cost data modules on eBay.  Obviously building such a decoder would then eliminate the need for the TIU in the moving engine.  The data processing could probably be handled with an Arduino.  Good idea but to quote GRJ...

Nothing is so easy as the job you imagine someone else doing.

 

I've read Adrian's posts last year, and I like his work a lot.  

I don't have enough knowledge on the technical specs to know for sure, but I think that high frequency track signal may be required to retain all the features of a full DCS system.  You could easily read the signal from he remote, but then you need to have a way to speak those commands to the engine... essentially building a TIU from scratch.  at that point you may as well just use the radio that's already in a TIU.  

FWIW, 2.4GHz modules and BLE modules are now very inexpensive as well, though they do take a bit more to make them work than the 433MHz ones.  

In this particular case I don't think it's easy at all, no matter who is doing it, but I do think it might be possible.

Look at Mark S. post about running the trains on his computer. He also developed a wireless control.

Engineer-Joe posted:

stan2004 posted:

A G-gauge engine pulling a bunch of cars going up a grade can hit 100 Watts.

Turning on smoke can add ~10 Watts in G-gauge. ….

 (I would NOT go much more than 22.2V) 

I saw that an average MTH one gauge engine's average draw is fairly low, between 1-2 amps on my Bridgewerks meter, closer to 1. With more engines added the draw doesn't go up fully the way I had expected it to. I ran seven diesels for example and the draw was far less than 4 amps. The meter varied as the train went around the yard (up and down hills). When I turn on the smoke, especially with more than one engine, the amps go up pretty well. Not double but more than you posted. With the seven engine example the total amps went up maybe to between 5 and 6? Still less than I expected.

 I guess it's dependent on what voltage we are talking about? I set my track at an optimum of 21+ volts (about 22 on the meter). That allows fast running, lower amp draws on that meter, and covers any of the power dips in areas. I would guess at lower voltages the draw could be higher?

 ...

Power = Voltage x Amps.  If you're running G-gauge engines with only 2 Amps at 22 Volts (44 Watts) doing whatever you have to do - pulling heavy loads, going up grades, around curves, etc.. then so be it.  I suggest 44 Watts is a bit low in power to budget for a G engine, but it really is a your-mileage-may-vary and of course you'll get more than double the battery life if you only use 44 Watts vs. 100 Watts!   BTW, it may surprise some but the audio power is relatively small...even blasting the horn at full DCS volume you incrementally use less than 5 Watts.

But to your question.  Yes, ignoring some arcane technical nuances, the DCS electronics is somewhat agnostic as to the voltage.  It's the power or product of Voltage x Amps that's important.  So if the engine is pulling, say, 50 Watts this could be 20V at 2.5 Amps.  But if the voltage was only 16V the engine would draw about 3 Amps.

Again, it's all about power (in Watts).  And for battery operation, it's about Watt-Hours (or stored energy).  Perhaps obvious but the LiPo phenomenon has truly raised/lowered the bar for battery operation of trains.  

A very common LiPo rechargeable in battery packs today is the so-called 18650.  Of course one must take some of the claimed Amp-Hr ratings with a grain of salt but note in this example that even though the LiPo battery is a bit larger, the stored energy is ~5 times greater than the NiMH.  

Perhaps obvious, but with the Lithium cells, it's that much more important to consider the charging method.  The "old-school" overnight trickle charger method is just that - old school.  Lithium cells essentially have a different charging profile so best to pair a standard battery pack and mating charger.  My 2 cents.

JohnGaltLine posted:

GGG posted:

Why limit your self to the DSCRC.  When you can get the full DCS installed.  G

Likely the $200 price tag for a TIU... But it does work.  

I'm wondering if it might be possible to more or less hack the output signal of a TIU to transmit the needed code.  Perhaps someone with a scope could see what frequency the data is sent at, and how it is formed.  It may be possible to push over radio with fairly inexpensive parts.  Without knowing what the track signal actually looks like it's hard to say for sure.  

You realize the capability difference between the two?  I do not think the $200 comes into play with the large outdoor G gauge guys.  Would they like it if it was less sure, but they are not going to run a railroad with all engines on address 1 and limited to the basic features of the DSCRC.

As far as doing this in O, ok.....  What is your market size, 10-15 people maybe doing one engine.  This would turn thier rail road into a LC engine with one remote telling all engines what to do simultaneously.  Other than being a science project, which could be moved over to the electrical section.

Most O gauge engines have more metal than the #1 gauge, are not run outside, and if you really have to have a battery, you can run it through a tiu.  But if you have some one who wants battery and also wants no track power.  Go with a system designed for it.  Like Airwire.  Or convert LC+ to battery.

You don't buy a Dodge Challenger when you want a European sports car like a Porshe, then try to make it handle like one

GGG posted:

JohnGaltLine posted:

GGG posted:

Why limit your self to the DSCRC.  When you can get the full DCS installed.  G

Likely the $200 price tag for a TIU... But it does work.  

I'm wondering if it might be possible to more or less hack the output signal of a TIU to transmit the needed code.  Perhaps someone with a scope could see what frequency the data is sent at, and how it is formed.  It may be possible to push over radio with fairly inexpensive parts.  Without knowing what the track signal actually looks like it's hard to say for sure.  

You realize the capability difference between the two?  I do not think the $200 comes into play with the large outdoor G gauge guys.  Would they like it if it was less sure, but they are not going to run a railroad with all engines on address 1 and limited to the basic features of the DSCRC.

As far as doing this in O, ok.....  What is your market size, 10-15 people maybe doing one engine.  This would turn thier rail road into a LC engine with one remote telling all engines what to do simultaneously.  Other than being a science project, which could be moved over to the electrical section.

Most O gauge engines have more metal than the #1 gauge, are not run outside, and if you really have to have a battery, you can run it through a tiu.  But if you have some one who wants battery and also wants no track power.  Go with a system designed for it.  Like Airwire.  Or convert LC+ to battery.

You don't buy a Dodge Challenger when you want a European sports car like a Porshe, then try to make it handle like one

I think you missed the point of my post.  You asked why someone would choose the DCSRC.  I expect the cost difference is the answer, if that product offers the features they want.  if it doesn't, the TIU works as well.  When compared to a command upgrade for O scale, $200 isn't really a bad price. on the other hand, if all you want is to make the engine move, and blow the whistle now and then, why not do it for 1/4 the price?  You don't buy a Porshe when you want to run on a 1/4 mile straight track, you buy the Charger, (or a fox body Mustang if you know what you're doing), if you want to go in a straight line fast and cheap. 

I never mentioned anything for the O market, and most of my post was on an entirely different subject, trying to transmit the full DCS signal from a TIU to a locomotive.  I understand it's not quite the hobby level sort of thing, but it might be worth while for guys like you to investigate.  After all there would be plenty of green to be had in offering installs of a "wireless DCS" board for the outdoor guys if such a thing were created.  

In any case, I'm out of the thread.  I should have known better than to reply in the DSC forum.  

With the current state of battery technology, running model trains at any gauge on battery power is just a matter of choosing the correct battery voltage and capacity. I have been running several locos on battery power for over three years without a single problem. In my opinion, a wireless direct to loco system, either bluetooth/wi-fi, 900MHZ, or even 27/49MHZ,  is the only way to take full advantage of the specific benefits of battery powered operation.

John, You missed the point I think and your mixing apples and oranges.  The engine is already PS-2 or 3 and they want to run DCS with Battery outdoor.  That is where this all started.  It is not the engine control system being modified, it is the power source and a mobile TIU not limited to stationary and track power.

Having said all that and heading in every other direction which is usual in these post ; I agree with Bob Walker.  Simpler methods to get battery train control AND LAYOUT CONTROL in O.  By layout control I mean multiple engines running together off one remote or device. G

G, this works pretty well

Blocked Post

I have simultaneously controlled up to five locos with BlueRail and up to three with AirWire. BlueRail can handle consisting as well. When I built my first 49MHZ battery powered wireless system in 2011, I could only control speed and direction of one loco at a time. The currently available systems are capable of so much more.

How to run engine on battery but control by DCS

It seems to me the OP posted here in this MTH DCS sub-forum because he specifically wants to control a battery-powered MTH DCS engine.  We are NOT talking about battery-powered engine control in general - there are plenty of other discussion venues.  In my opinion, the discussion needs to focus on DCS with a Protosound 2/3 equipped engine as a given.  Again, my opinion only but suggestions about using Airwire, LC+, BlueRail, 49 MHz, etc. are what is confusing apples and oranges. 

I still maintain there are only 3 candidate solutions to the topic as titled: 1) TIU, 2) Explorer, 3) DCSRC.  By stating he wanted to control using the TIU's remote handheld, this ruled out the Explorer (uses WiFi smartphone) and the DCSRC (uses proprietary IR remote).  As I understand it, the OP already has a TIU so he has a solution.  Maybe the thread should be closed.

But if the purpose of the OGR forum is to discuss ideas, I believe this topic will come up again with the increasing availability of LiPo battery systems affording run-times in hours rather than minutes.  Several recent posts dwell on the one-engine limitation of the DCSRC.  But all 3 methods have limitations since they are not being used AS INTENDED; these are all work-arounds.  For example, it seems to me the TIU method is also limited since I believe a TIU remote can only talk to (up to) 5 TIUs?  So by dedicating 1 TIU per engine, it seems you'd be limited to 5 engines.  And if the idea is LAYOUT control which means activating switches/accessories using a stationary TIU/AIU, then it would be 4 engines.  

I think the value added to continuing the discussion is to document issues specific to the topic.  To that end, I powered up a DCSRC and a TIU for the specific purpose of measuring idle power requirements.  The DCSRC was 1/2 Watt.  The TIU was 2 Watts.  I do not have an Explorer.  With battery-powered operation a key parameter is available Watt-Hours...a foreign concept for track-powered systems.

The pie-in-the-sky solution is for MTH to develop an optional module that resides in any DCS engine.  The module would have a postage-stamp sized WiFi radio and translate smartphone commands to the magic DCS signal.  Simple hookup: 2-wires in (from battery) and 2-wires out (to red/black PS2/3 track power connections).  In principle this is like the tiny Airwire CNVRTR modules except those only talk to DCC-equipped engine.

stan2004 posted:

PaulB posted:

 

I have a question on your battery voltage. Will smaller size batteries work with this set-up. I currently have 14 and 17 installed. Is the 22.v a better choice to power the TIU and an MTH Big Boy?

It's not the voltage per se.  What's important is the amount of stored Energy which is measured in Watt-Hours.  Casey Jones2 is using 22.2V battery with 10 Amp-Hr capacity.  That's 222 Watt-Hours.  So if the engine is consuming 10 Watts, it can run for ~22 Hours.  If consuming 100 Watts, it can run for ~2.2 Hours.  Etc.   A DCS engine at idle might draw 10 Watts.  A G-gauge engine pulling a bunch of cars going up a grade can hit 100 Watts.  Turning on smoke can add ~10 Watts in G-gauge.  And so on.

I figure his 22.2V battery is a stack of six 3.7V LiPo batteries (3.7V x 6 = 22.2V).  Since the market for battery-powered G-gauge trains is a drop in the ocean compared to laptop or similar battery high-volume consumer applications, I think you must resign yourself to what's readily available.  For example, it seems many battery-powered tools use 18V, 20V or whatever...and come with a mating battery charger - makes charging/swapping battery packs that much easier.

So whatever battery voltage you use, I'd suggest the highest Amp-Hour rating.  Practically speaking, which means what you'll find readily available, you should choose the highest DC voltage that the DCS engine will tolerate (I would NOT go much more than 22.2V) and the highest Amp-Hour rating that will fit in your space available. 

Don't go too low in voltage because this will limit your max speed (independent of the Amp-Hour rating).  A 14V battery pack might limit your top speed to less than 50 sMPH (depending on load).

Stan2004, Thanks as was just going to write all that up myself but you beat me to it!

I buy battery packs already made from eBay. The 22.2V LiPo's are for large RC helicopters I believe and can be bought for like $90. There's a few larger 12 & 14,000mAh ones out there but that thing is like $350.00.

We have used the MTH Explorer in a conversion but as Barry posted no soft key function which I don't like and uses a smart device only which I don't like esp. for use outdoors in the sun....I prefer the one-handed DCS Remote. 

Oh and can go up to 30V and we run on track power that high with DCS engines.

 When posters here were first describing the Explorer, I didn't realize it was for apps only. I would have guessed it would work with the DCS remote. After releasing more info, I saw that it didn't list the DCS remote for control. I suggested it as most people have an older cell phone, current one, or even a pad around for control.

 So then the OP stated he had a full DCS system for using to control just this engine. That made sense to use what he had. He already paid for it.

 I have found that DCS engines are stable below like 24 volts. I personally would not take them up to 30. Maybe they can handle it but I don't believe they operate reliably and stay cool. I don't really know fully, for sure. I did have some issues at 26 to 27 volts during a test and went back to my current established 21 volts at the board. I figure that's what MTH designed them for. I also noted that at 18 volts some converted engines could not reach full speeds.

 So I operate between 21 and 24 volts outside. I understand that some have converted the USA BigBoy for example and run higher voltage to obtain full speed. That's not necessary with MTH engines.

stan2004 posted:

How to run engine on battery but control by DCS

It seems to me the OP posted here in this MTH DCS sub-forum because he specifically wants to control a battery-powered MTH DCS engine.  We are NOT talking about battery-powered engine control in general - there are plenty of other discussion venues.  In my opinion, the discussion needs to focus on DCS with a Protosound 2/3 equipped engine as a given.  Again, my opinion only but suggestions about using Airwire, LC+, BlueRail, 49 MHz, etc. are what is confusing apples and oranges. 

I still maintain there are only 3 candidate solutions to the topic as titled: 1) TIU, 2) Explorer, 3) DCSRC.  By stating he wanted to control using the TIU's remote handheld, this ruled out the Explorer (uses WiFi smartphone) and the DCSRC (uses proprietary IR remote).  As I understand it, the OP already has a TIU so he has a solution.  Maybe the thread should be closed.

But if the purpose of the OGR forum is to discuss ideas, I believe this topic will come up again with the increasing availability of LiPo battery systems affording run-times in hours rather than minutes.  Several recent posts dwell on the one-engine limitation of the DCSRC.  But all 3 methods have limitations since they are not being used AS INTENDED; these are all work-arounds.  For example, it seems to me the TIU method is also limited since I believe a TIU remote can only talk to (up to) 5 TIUs?  So by dedicating 1 TIU per engine, it seems you'd be limited to 5 engines.  And if the idea is LAYOUT control which means activating switches/accessories using a stationary TIU/AIU, then it would be 4 engines.  

I think the value added to continuing the discussion is to document issues specific to the topic.  To that end, I powered up a DCSRC and a TIU for the specific purpose of measuring idle power requirements.  The DCSRC was 1/2 Watt.  The TIU was 2 Watts.  I do not have an Explorer.  With battery-powered operation a key parameter is available Watt-Hours...a foreign concept for track-powered systems.

The pie-in-the-sky solution is for MTH to develop an optional module that resides in any DCS engine.  The module would have a postage-stamp sized WiFi radio and translate smartphone commands to the magic DCS signal.  Simple hookup: 2-wires in (from battery) and 2-wires out (to red/black PS2/3 track power connections).  In principle this is like the tiny Airwire CNVRTR modules except those only talk to DCC-equipped engine.

 

Actually he stated #1 gauge as the very first requirement.  You jumped it over to O, which lead to the spiral.  Will see how long this last as an O gauge solution that folks require.   G

Casey Jones2 posted

Stan2004, Thanks as was just going to write all that up myself but you beat me to it!

I buy battery packs already made from eBay. The 22.2V LiPo's are for large RC helicopters I believe and can be bought for like $90. There's a few larger 12 & 14,000mAh ones out there but that thing is like $350.00.

We have used the MTH Explorer in a conversion but as Barry posted no soft key function which I don't like and uses a smart device only which I don't like esp. for use outdoors in the sun....I prefer the one-handed DCS Remote. 

Oh and can go up to 30V and we run on track power that high with DCS engines.

This is what I'm planning to do as my next project - DCS Explorer w/ Battery.     

Based on what I can tell, the only limitations of the DCS explorer are having to use a smart phone outdoors (which I do everyday) and  losing soft key functions (Which I'd never use anyway). 

It sounds like steam will eat some battery but 22.2 V Lipo's would be ideal?   Ideal Mah?  I'm not likely to run a train much over an hour / hour and a half and could easily swap batteries if I want to run longer.   

Double-heading is a question for me - Would dcs explorer and battery fit in say a Hudson Tender... or must there be a trailing car?  If they do fit in tender, can multiple DCS explorers be run from the same APP at the same time?

can multiple DCS explorers be run from the same APP at the same time?

Yes, they can, however, the number of engines that the DCS App can control is still limited to a total of 3, regardless of which engines are controlled by which explorer.

Further, the DCS App has no lashup function available when operating in DCS Explorer mode. Why not just tether the engines together to one DCS Explorer?

This and a whole lot more about DCS WiFi is all in MTH’s “The DCS WiFi Companion 2nd Edition!" This book is available from many fine OGR advertisers and forum sponsors, or as an eBook or a printed book at MTH's web store!

Barry Broskowitz posted:

can multiple DCS explorers be run from the same APP at the same time?

Yes, they can, however, the number of engines that the DCS App can control is still limited to a total of 3, regardless of which engines are controlled by which explorer.

Further, the DCS App has no lashup function available when operating in DCS Explorer mode. Why not just tether the engines together to one DCS Explorer?

That's interesting.  So 3 engines total per app, regardless of explorers.   That's too bad.  That means trailing car.  My Logic was that having each engine with it's own independent explorer would allow for maximum usability.  Having to lash engines up to one explorer / battery limits what one can do.   I might have to re-evaluate my plans.   Weighing the cost and usability, it might be better for me to just gut PS2 and install Railpro.