Post by: KPack on February 13, 2017, 08:48:36 PM
I purchased some 3.7V 700mah batteries from a hobby vendor online (banggood I think), that came in a 5-pack with charger for something aroud $20. I then bought a few Pololu voltage converter boards from Litchfield Station, and some 2-pin plugs from Evan Designs (modeltrainsoftware.com).
The hardest part of doing on this was figuring out where to fit the battery. I used a Kato SD40-2 and the only place a battery like this would fit was in the fuel tank. That meant milling off most of the metal from that area, which was a huge pain in the rear. My friend was kind enough to do it on his mill for me.
Once the milling was complete, I set the battery inside the fuel tank, routed wires through the frame and past the motor, then soldered it to one end of a 2-pin plug. The other part of the 2-pin plug was soldered to the voltage converter board. Then the power wires for the LM-2S were soldered to the converter board. And that was it, really. It wasn't much more complicated than that.
Here is an overview of the installation. The Pololu converter is mounted to the styrene board right in front of the LM-2S module.
Another view showing the wires coming up by the side of the motor
The battery inside the fuel tank
Look at how much room is above that speaker! I love how Kato's motor is so low.
Here is a quick video showing how the battery works:
-Kevin
Post by: William Brillinger on February 13, 2017, 09:37:02 PM
Post by: Alan on February 13, 2017, 09:57:39 PM
BTW Pololu makes all kinds of neat stuff for robotics. Their 2571 step-up regulator would allow you to get the normal RP 14.8V if you wish. It is slightly larger but should fit most places.
If you are thrifty like me, and can find the mounting space then check this out: http://www.ebay.com/itm/2pcs-MT3608-Step-Up-Power-Apply-Booster-Module-DC-DC-2V-24V-2A-NEW/171907688472?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D2%26asc%3D41451%26meid%3Dcfce865084c74c2f8361c6af72b22476%26pid%3D100005%26rk%3D5%26rkt%3D6%26mehot%3Dpp%26sd%3D191871557118 (http://www.ebay.com/itm/2pcs-MT3608-Step-Up-Power-Apply-Booster-Module-DC-DC-2V-24V-2A-NEW/171907688472?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D2%26asc%3D41451%26meid%3Dcfce865084c74c2f8361c6af72b22476%26pid%3D100005%26rk%3D5%26rkt%3D6%26mehot%3Dpp%26sd%3D191871557118)
Post by: nodcc4me on February 14, 2017, 01:01:11 PM
Being somewhat electronically challenged, I'm not sure how you are able to run it with sound on only 3.7v?
Post by: KPack on February 14, 2017, 01:19:44 PM
Al - I have a ton of space in this locomotive. Partly because Kato mounts their motor so low, and partly because the only things I'm adding into the shell are the very small Pololu board and the small 2-pin plug. I purposefully mounted the battery in the fuel tank because I wanted to keep the interior of the shell clean. The other thing I did (you can see it in the 1st picture if you look carefully) is I mounted a small strip of brass on the stryene board to use as the "common" for all the lights. All the lights have their positive wire soldered to that board....it really helped to keep things clean and tidy. And I always like to tuck as many wires as I can between the motor and the module so they aren't hanging all over the place like a rat's nest.
-Kevin
Post by: Alan on February 14, 2017, 01:23:23 PM
Quote
Being somewhat electronically challenged, I'm not sure how you are able to run it with sound on only 3.7v?
Kevin is using Pololu voltage converter boards from Litchfield Station. They convert the battery voltage from 3.7V up to 12V.
DC voltage up converters work very similar to the ignition system of old cars. In a car, points close allowing low voltage to charge the ignition coil. When the points open the coil discharges at a higher voltage. DC up converters function the same except at a much higher open/close frequency than car points and a much lower coil discharge voltage than an ignition coil. The coil discharge, instead of being sent to a spark plug, is temporarily stored in a capacitor. A regulator then drains the capacitor at 12V output. Before the capacitor runs out of juice the points have closed/opened again repeating the cycle.
EDIT Made an illustration showing the components:
Does that help?
We have to respect the laws of physics. Energy cannot be created, merely transferred. The wattage (power) needed to move the loco and play sound is the same on Kevin's system as it would be on powered rails. W=V*A. The DC up converter board is delivering the same wattage as the rails would be but it is doing so with 3.7V. Therefore, the amperage must be higher to make the equation balance. That is why I am so interested to see how Kevin's loco fares after the battery gets some life on it. New batteries always work good. Aged batteries, not so much.
Post by: KPack on February 14, 2017, 01:46:06 PM
Quote
That is why I am so interested to see how Kevin's loco fares after the battery gets some life on it. New batteries always work good. Aged batteries, not so much.
This will be interesting to find out. I don't run a ton unless I'm at someone's place on their layout. So I may not have much to report for a while. Worst case scenario is that the battery ends up going belly up much sooner than I anticipated. But if that happens, I just swap the battery for another one. These things only cost a few bucks a piece....maybe $5-6. Plus batteries will likely improve over time, allowing smaller packages with higher storage capacity (at least I hope).
Anyways, I still need to do some testing with this. I haven't hooked it up to normal Railpro locos yet to see how the start voltage/MU capability is affected. In the end, this locomotive will always be paired with another SD40-2 (ScaleTrains) that will also be battery powered, so any negative effect on the start voltage won't be an issue. I'm building a matched pair of SD40-2's that service my area....#1778 and #1999....they have been coupled together for years now serving the Central Washington area.
-Kevin
Post by: nodcc4me on February 14, 2017, 02:39:10 PM
Alan, your very cool diagram and narrative does help explain the required voltage gain. In another thread you mentioned that you were just an amateur at this stuff. LIES! ;D
BTW, what program did you use to make the diagram?
Post by: Alan on February 14, 2017, 02:46:57 PM
Quote
These things only cost a few bucks a piece....maybe $5-6.
On other forums I am on record as an anti-battery guy. Not because I love track wiring but for two primary reasons, one of which you mentioned:
First, my layout is all first gen 4 axle diesels. Batteries will not fit. Shucks, even an LM won't fit in my S-2s without shedding its plastic case and milling a flywheel! Doubtful batteries will ever be so small in my lifetime as to be practical in a little loco. This restriction is unique to my situation (or others with same). I get that. Not a direct fault of battery technology.
The second and more universally important reason, "These things only cost a few bucks a piece....maybe $5-6" There it is. My Fitbit made me aware I walk far more steps in a day than I would have ever guessed. I suspect the same may be true with hours on a loco. I like to play with trains and have a fair bit of track to do so on with quite a few locos. The life expectancy of a battery worries me greatly. If I had to install a new battery in each loco every year or even every other year, the operating cost would get ridiculous especially when compared to a one-time-purchase of power supply and wire. And I would still be left with locos that have to be charged regularly. Cell phone and power tool ownership has clearly demonstrated to me batteries are great when new. Two years of use later they peter out fast.
I dwell on this subject because way back in the LK&O design days I seriously considered battery power. You have to admit, there is a strong allure to having no track wiring. I am not trying to make my equipment run on other folk's crappy track nor am I too lazy to properly clean my track. Battery power makes no sense. Am I missing something? :-\
Post by: KPack on February 14, 2017, 03:03:22 PM
Would I ever convert all my locomotives to full battery? No. At most I will do this and the other SD40-2 I mentioned earlier. As it stands right now I don't feel that battery is practical for the long term, but I have to admit it is a blast to play around with. The fact that the locomotive will never stall, ever, on any track, is appealing. But I don't relish the idea that constant battery replacements may in order for the long term.
Arguably the best way forward to prevent the occasional power loss problems would be to do a keep-alive, as has been discussed in the other thread. I don't think that layout wiring is going away anytime soon, so might as well use it and keep-alives to maintain a constant source of power to the module. That is probably the route I will take on the remainder of my fleet. I have several locomotives that never lose power, but many others that seem to be more prone it. A bridge rectifier and KA would make easy work of that problem.
-Kevin
Post by: Alan on February 14, 2017, 03:03:26 PM
Quote
BTW, what program did you use to make the diagram?
Photoshop, started with the image from eBay.
Quote
A regulator then drains the capacitor at 12V output.
You may be wondering where is the regulator on the eBay cheapie board. There isn't one. That's why it is $0.58 from China. The output voltage on the eBay unit is controlled only by the switching frequency (potentiometer setting). Temperature, load, and driven circuit capacitance and inductance are just a few variables that will cause this unit's output voltage to fluctuate somewhat. But hey, what do you expect for 50 cents? :P
The real-deal DC up converters (like Kevin's Pololu unit) do use regulators. And they cost a good bit more than fifty cents. The question is... is that level of precision needed for model trains? I don't know. I tend to think it isn't.
Post by: TwinStar on February 14, 2017, 11:18:34 PM
http://www.dailymail.co.uk/sciencetech/article-4219518/New-battery-store-charge-TEN-years.html
Post by: Alan on February 15, 2017, 07:29:23 AM
Relative to my earlier comment... 'Lithium ion batteries don't even survive 1000 complete charge/discharge cycles,' said Dr Michael Aziz, a professor of Materials and Energy Technologies at Harvard University."
Post by: KPack on February 20, 2017, 11:12:28 AM
-Kevin
Post by: KPack on March 04, 2017, 09:29:27 PM
With the set up I did here I used a battery, the Railpro module, and very small Polulu converter board. Done.
Give the article a read if you can. It will make you ever more grateful for Railpro, even if you don't plan on doing deadrail.
-Kevin
Post by: Alan on March 05, 2017, 07:14:31 AM
Lest you think I am a RP fanboy, not so. RP will meet a similar fate. Look inside the HC and LM. Generic off-the-shelf components in both of them. AMD processor and flash memory in the HC, PIC processor and flash memory in the LM. The secret sauce for RP is the software. Hardware-wise RP is BlueRail is RP. Level of refinement is the only difference. Sooner or later genericification will come on the scene. Then software will be the sole differentiation. Perhaps why Ring is developing RP Assistant instead of additional gadgets.
While on my soap box.... the battery charging from rails idea Bruce speaks of is DOA too as far as I am concerned. Batteries in general don't respond well to high cycling rates especially LiPo. Tolerance, or lack thereof, of high cycling rates is what separates batteries from capacitors. If batteries cycled well we would have lot less need for capacitors in common devices. Anyone who has owned a laptop computer has learned batteries used in high current applications don't like cycling. In a charge-from-the-rails application the poor battery is subjected to hundreds of mini charge/discharge cycles every time the layout is operated. Good luck getting any life out of that battery. Case in point: Think about your smart phone battery. Certainly state-of-the-art in battery management. How well does it perform? Great in year one, okay in year two, horrible in year three. Now think about your watch battery. Runs 5-10 years without a problem. The difference? Your watch battery never gets cycled. A watch is the ideal battery environment - a single long slow burn.
It is truly a shame so many people have such an unfounded fear of wire and solder. The track may be a very poor network bus but it is an ideal power conveyance. I am amazed at how much time, money, and effort folks will invest just to avoid one-time soldering of a few more feeders or the four connections to power a frog. Bruce's article epitomizes this fact.
</rant>
Post by: G8B4Life on March 05, 2017, 07:38:20 AM
Alan, I'd be interested to know how you know it's an AMD processor in the HC.
- Tim
Post by: Alan on March 05, 2017, 09:04:46 AM
Post by: William Brillinger on March 05, 2017, 09:55:34 AM
http://model-railroad-hobbyist.com/magazine/mrh2017-03/dcc-impulses
Post by: Alan on March 05, 2017, 09:56:15 AM
Touch your finger to a piece of glass. Hold the glass to the light. Did you leave a fingerprint? Fingerprints are skin oils. Oil is an insulator Dust sticks in oil. Oil and sticky dust - two things everyone needs on their track. /S Will touching your track bring trains to a halt? Of course not. But it is just one more log on the poor connectivity fire. I wonder why people who flick switches bother to clean models before they paint them. also /S
Post by: G8B4Life on March 07, 2017, 05:18:37 AM
I mis-typed. Should be ARM, not AMD. The architecture, not the manufacturer
Thanks for the clarification Alan, thought you might have had some magic insight and could tell who made it and what it likely was just by looking at it and what's around it.
The article in question is available here:
Thanks Bill. For what it's worth I found out MRH's "next issue" clock is useless, It was available much much earlier that it said.
- Tim
Post by: William Brillinger on March 07, 2017, 06:56:16 AM
Quote
For what it's worth I found out MRH's "next issue" clock is useless, It was available much much earlier that it said.
The clock was counting down until the official release email. MRH is released early on the forums to let the regulars help find any errors that missed the editing process.
Post by: G8B4Life on March 07, 2017, 07:44:08 AM
The clock was counting down until the official release email. MRH is released early on the forums...
Thanks Bill, I was pretty sure subscribers (I am not one) were able to get it early, as you did, but thought the rest of had to wait for that clock to count down. It was less than half of the time remaining on the clock before it was available on the front page to me.
oh, and to keep this on topic; It was an interesting article to compare deadrail on DCC with deadrail on RailPro. I would hate to think of trying to cram all that extra stuff into my smaller than US prototype models!
- Tim
Post by: Alan on May 02, 2017, 10:40:15 AM
XP Power just released a new series of super compact DC-DC converters. JSM2512S15 is perfect for battery to RP power conversion. At 1" x 1" x 0.4" it should easily fit in HO models. It will provide steady, tightly regulated 15V at up to 1.67A until the battery drops below 9V.
http://www.xppower.com/Portals/0/pdfs/SF_JSM25.pdf (http://www.xppower.com/Portals/0/pdfs/SF_JSM25.pdf)