I just installed a Bluerail board in a Chattanooga, putting all the parts in the tender.
The Bluerail board is somewhat limited in the amp draw, but it works fine in a big hauler.  Pulling the tender and 4 passenger cars, on a 5' diameter circle, on the flat, I measured .36 amps at 1/2 speed, and .5 amps at WOT, which is well below the bluerail max amps.  This is using a 1300 mah 3 cell LIPO.  Have yet to try on an incline nor the bluetooth range.

For low voltage cutoff I made a small board which utilizes a relay which cuts out if the voltage drops below 9.6 volts.  I also put in a 2 amp fuse although the Bluerail board has overdraw protection up to a limit.

For sound I put a small Bluetooth speaker in the tender along with the LVC board, Bluerail board, and battery.  I found where some folks have robbed the BT receiver part out of a cheap BT speaker so as to use the existing tender speaker and also replaced the BT speaker battery with a voltage regulator so it is powered off the main battery.  The bluerail app includes a way to sync the chuffing sound, which with some fiddling works fairly well. 

If anyone is interested in how to make the LVC protection, I can send a schematic of the circuit although it is dirt simple.   I use a momentary "start"  button to initially trigger the relay, and then it is held with the battery power going through an appropriately sized resister such that when the input voltage drops below 9.6 (or whatever is desired for a cut off) the voltage drop from the resister puts the voltage below the minimum hold voltage for the relay being used.  I also power the "drop" circuit through the relay contacts so when it reaches LVC the hold circuit is also disabled.  Hence there is no draw at all if accidentally leaving a connected battery in the train.

I charge outside of the train.

I'm not pushing this compared to the RC solutions - just wanted to experiment with the Bluerail board and other than a potential range issue, it looks like a nice solution.

Now after further examination, finding a substitute motor truck seems like the best option.
Running the unit upside down with the lower "lid" off, the noise/vibration is coming from the worm gear/spur gear junction.  Given that use of a straight cut spur gear (as opposed to spiral or helical) is prone to noise and vibration I don't think much can easily be done to improve this.  Add to this various parts that are far from precision.
Why the trolley on the track didn't respond well to DCC speed steps is a bit of a mystery.  Running upside down it responded correctly to speed steps.  It wouldn't start until step 4 but this could be rectified with Vstart CV.

It appears folks have had success using either a USA or Aristocraft motor truck.   Interested in anyone knowing where I might find one.

Thanks, Bob

G scale.
Since my 1st post I noticed it looked like the wheel gears may have been rubbing slightly on the bottom half of the case   I dremeled the plastic a bit.  This cut down on the noise but only slightly.  The worm gear lash looks about the same on both ends so I don't think it is binding at that point.  BTW, I think this unit was either NIB or very slightly used.

Meanwhile, with more testing it appears the motor doesn't respond well to incremental DCC speed increases.  I'm running on a DCC controlled point to point track with a "middle" stop (TamValley unit).  No problems with an ART railbus or a "Clemetine", both with the same decoder (NCE D13SR).  The start up and slow down is controlled by the decoder CV 3 and 4 values.
I then ran the united traction back and both using my normal DCC base/throttle. The united traction will run fairly well if kept at the same speed, but as with the TamValley shuttle unit, it doesn't gradually pick up speed very well.

All this might not matter except for my intended use on a point to point track.  Crawling out of a "station" and then suddenly going to the desired speed just doesn't look right.  One reason for going to a DCC shuttle instead of a DC shuttle with diodes was to get the incremental start/stop as well as consistent light intensity.

Still looking for suggestions but meanwhile found an article where someone gave up on the original motor brick and converted to an Aristocraft brick that is a close fit. 

Thanks,  Bob

Any ideas on adjustments etc. to eliminate grinding noise with a united traction trolley?
While converting to DCC I didn't see anything wrong with the gears and they are well lubed.
Thanks, Bob

Got the answer for the variable stopping points.
The unit deals with any 2 digit loco address on the track (additional device available for 4 digit addresses) by cycling through stop or start commands for all the loco numbers.  Hence, the loco on the track could be close to the next command to be issued, or if just missed in the sequence you wait for it to circle around.
Not that big a deal unless one wants to be fairly precise on where the loco stops.

I looked at the dimension product but it is designed to go between a receiver and speed controller.  It's designed for airplane usage - slow the motor down, let the servos still work, etc.  The product I'm looking at using is a Bluetooth receiver and speed controller in one board.  So, what I need to do is have an LVC between the battery and the power to the board.

Found one so far by mcmelectronics but it is way over kill.

Suggestion from another forum was to make my own with a relay and properly sized resistor to disable the coil current at a known voltage.  That said I've never made an electronic gizmo that turned out as simple as first thought.

I may be over worrying this.  The cheap low voltage detectors that flash a light and/or sound a buzzer may be just fine.   

I will need some safety mechanism. Going to play with various end of track complete power shut off mechanisms.   Easy solution is just an isolated section of track on each end but this would waste a lot of run length especially if allowing for a train with cars running backwards.
Speaking of which when testing with a loco with cars, I had a knuckle coupler jiggle loose. I suspect the system might get confused if a detached lighted car goes into an end or middle zone.
Note regards sound:  The TamValley doc notes that you can't activate sounds, which is true but if sound board is by default active you will get chuffing etc.  Just no way to turn it off without having an on board switch or move to a track hooked to your DCC cab.

Anyone know where I can find a Low Voltage Cutoff device (LVC) for a 3S LIPO?

I want to use a LIPO for a battery conversion but want to make sure I have an LVC included.  I'll probably just use a fuse for overdraw and will be taking the battery out for charging.

I was told elsewhere to look for a PCM or PCB but the descriptions of these just talk about overdraw and overcharge protection.

BTW, I have found some LVCs but they are meant to go between a receiver and speed controller, like in electric model airplanes that don't have the LVC built into the ESC.  I will have a receiver and speed controller (considering the Bluerail unit) but the receiver and speed controller are all in one.

Thanks, Bob

Ton I can't see how sensors could be used with the TamValley unit.  It works by sensing current drawn from the wires connected to the end sections (or middle).  I supposed one could rig up a way that a sensor would put a load on the end section wires but I've since discovered the sensing of crossing the gap isn't the problem.

I took slo motion videos such that the image shows the loco wheels relative to the gap and at the same being able to see the red light on the shuttle unit that comes on when the load is introduced on the end zone wire.  My videos show that the sensing time occurs between 1.5 and 2 inches after the first set of wheels with pickups cross that gap.  What this tells me is the delay in sending commands to the decoder is the variable that causes the inconsistent stopping points.  The doc says there can be up to a 1/2 delay in sending the stop command so i'm speculating the delay is not consistent.

I'm still talking to TamValley about this but if nothing can be done about the stopping variance I'm seeing, at least for me the unit is still a good solution. 

Any one converted a United Traction open Streetcar to DCC?
Thanks,  Bob

Ton, I sent you a PM but haven't heard back to see if you have had the same results.
I have been testing the Tam Valley DCC shuttle unit.  It works as advertised by simply putting a DCC trolley/loco on the track after the appropriate wiring.
The only issue I have so far is inconsistent stopping points.   I've tried this with no deceleration and with CV 3 and 4 set to 3.
In both cases testing a Clemetine (0-4-0) the stop points vary by 6 inches.  Tam Valley told me to check that the track, wheels, and contacts are clean, which I did.  My test track is ART and although not screwed down it is fairly flat.  No light flicker so at least one set of wheels is making contact throughout.
Given there is essentially no suspension on this 0-4-0 I can imagine the first set of wheels crossing the gap may not always make contact. 
A Bachmann big hauler had a worse variance - 15" although I did not clean the wheels on this test.
I discovered it doesn't take much current draw from the end zones to trigger the unit.  Backing a caboose with only 3 lights was enough to trigger the unit so the position of the gaps will be important based on what is being run.  I'm already planning on multiple gaps with switches to determine the "real" ones so this isn't an issue. 
If the variable stop position is indeed caused by dirty track, wheels, and/or contacts then some extra length needs to be included in the end zones.  The stopping points could start at one distance but get farther as contact worsens.

Because the documentation notes that powering off in an end zone could cause a loco to start in the wrong direction when re-powering, I think it will be prudent to install some sort of emergency shut downs at the ends of the runs. Another solution could be to include an isolated section of track on each end but this could use up a lot of space for the rare instance where it might be needed.  Not bad for just a short trolley but quite long if pushing a train with cars backwards.

A DCC system using sensors instead of current sensing would be interesting to look at.  If anyone knows of DCC systems other than Tam Valley, or has experiences with the Tam Valley unit, I would appreciate any input.
Thanks, Bob

I've since discovered a slow down method for the shuttle unit I have which is to introduce an additional gap on each end spanned with diodes facing the other way which drop the voltage hence slowing the trolley/train down before coming to an abrupt halt after passing the end zone diode.   Number of diodes in series determines the voltage drop.  Minor point is needing at this point one or more diodes in parallel to the others facing the other way to let the trolley proceed after the voltage is reversed.  Bonus is these diodes will also limit the initial acceleration.

I'm going to try this although for a new point to point I'm making I think I'll go with the TamValley DCC shuttle unit.  It is a new product but someone local has one coming in so I can hopefully find out the nuances before I actually buy one.
Next trick will be how difficult it is to put decoders in some new trolleys I have coming.  One is an ART railbus that although a bit of a pain, someone has documented how to do it.   No clue about sneaking one into an open Bachmann United Traction.

Thanks again for ideas, Bob

Thanks  Bach-man.  I did look at the unit Dallee has and even though the DCC at first appealed to me I discovered an issue.   The one I'm looking at can start in the wrong direction on a power toggle if left in one of the end zones.  I have 5 gscale locos I put NCE D13SR decoders in, which are only 1.3 amp/2 amp peak.  I've never had a problem with these running a variety of locos on a flat track even pulling 9 or 10 cars.

But, if using the DCC shuttle unit I've been looking at I accidentally hit a bumper I know I my stall current will be too much for the decoders I'm using.  I suppose one could put enough extra isolated track beyond the end zones but this could really increase the overall length if want to shuttle a train with cars.   Another possibility might be toggle switches disguised as barriers to kill the power if the train runs into them.  All this might be moot given the Tamvalleydepot unit is only good for 1.5 amps anyway - if it shuts down rather than blowing up.

The DC solution regards lights is becoming almost moot.  Besides the idea of using a switching VR, I see from some of the web sites that ready to go LED units that do this are readily available.

I'm also introducing another very silly possible requirement.  A unit that comes active from just applying the transformer power would allow me to interface a power switch with Alexa (amazon gizmo).  "Alexa, turn on train".  I like doing things for visiting kids and this would be great.

Thanks again, Bob

Thanks for the input.
I did find some DC units that don't use diodes and will slow to a stop.  I also am going to experiment with a capacitor with the unit I currently have.

However, I am willing to convert my trolley to DCC and located the Tamvalleydepot DCC shuttle unit. Besides slowing to a stop I like the DCC solution because my interior lights will stay the same regardless of speed.  As Digitalgriffin notes, the deceleration is controlled by a CV setting which allows more control of the slowdown in comparison to some of the DC units.  Also, I can use the same track for testing/exercising my DCC locos as I don't always have a circular track setup.

Researching the units I've found, there are a number of differences worth considering.  For instance, I would like to put different trolleys/trains on a point to point track, and being able to easily change the end points for different length trains would be nice.  Some units use sensors with no track gaps which would make it easy to move the sensors for different end point lengths.  On the other hand, even with those that utilize track gaps (like the DCC unit I found), I can envision putting in extra gaps where all but one on each end are connected to allow a variance in the end point length.

The options to have midpoint stops, sidings and even up to a 5 part switchback are also interesting.

Thanks Hunt for leading me to Circuitron.  Besides having a DC unit that does soft stops, they have many other interesting items.

In case others are interested in the options, I'll add to this after I try out some of the products.  Regards dim lights on DC running a slow speeds, I suspect there are many light oriented discussions on this topic but I'm also going to experiment with a switching voltage regulator to run lights designed for 3V.  Some years ago I did this in some cars using a linear VR but the resultant heat dropping from 12V to 3V is huge.

Thanks again for the input,
Bob

I have an auto reversing unit that works fine but because it is the type utilizing diodes across a track gap, the trolley comes to an abrupt stop.  Anyone know of a system that allows the trolley to slow to a stop?

Thanks, Bob