The recent thread about carbon pile regulators got me
to thinking it might be a good idea to ask everyone here
a few questions regarding a 32 volt battery charging circuit
I am working on.

It is a long story with many complex details, so I'll attempt
to be brief and explain the overall picture as best that I can.

The East Troy Electric Railroad has several Chicago, South Shore
and South Bend Railroad cars operating. These various cars
have several different methods for charging their "32 volt"
batteries. I say 32 volt in quotes because that is only
the nominal voltage. It varies somewhat from car to car
because some have the original Edison cells from CSS&SBRR,
while others have series strings of lead acid truck batteries.
The lead acid battery strings have a nominal voltage of
36 volts instead of the original 32 volts.

These cars were originally powered by 1500 volts DC on the
South Shore. At the museum, they are operating on 600 volts DC.

The air compressors have been refit with 600 volt armatures.

The motor-generator sets which were originally used to
keep the 32 volt battery string charged have not been
modified to operate from 600 volts. That is for most of
the cars. A few cars such as the diners do have proper
motor-generator sets working.

The remainder of the cars have a charging system set up
which uses a resistance in series with the 600 volts to
keep the 32 volt battery string charged. This crude system
works, but leaves much to be desired.

I've come up with an experimental regulator that is ready
for testing. It is a somewhat crude experiment, but I feel
that it works well enough on the bench that it is ready
for testing on one of the cars. All it does is it senses
the voltage and either turns on the charging system
or turns it off. It works by using zener diodes to create
a sensing circuit which has adjustable set-points for
turn on and turn off. That very low current sensing
circuit is optically isolated from everything else. The final
output operates a heavy-duty industrial relay which
has contacts suitable for operating the 32-volt coil
on the 600 volt contactor which energizes the resistive
battery charging.

This outfit has been bench tested and is
ready for live testing on a CSS&SB car.

However, I am also interested in learning all that I can
about different methods of regulation for a 32 volt battery
charging system.

I have heard that in some cases, some trolleys and interurbans
simply had the air compressor in series with the batteries.
That then charges the batteries at around 5 amps when the
compressor is running.

I am looking to start a general discussion thread here
about regulation circuits of all various types and applications.

By far the best option for the South Shore cars would be
to have proper MG sets. But since that does not appear
to be viable mainly for cost reasons, I've got to look
at other options.

Right now the zener diode operated prototype is all I've
been able to come up with so far. I've got several more
ideas, but figured it's time to put this out there
and see what others might have to say about it.

How many amps required to charge the batteries? Obviously it varies, but in a reasonable timeframe. eg Would something like this https://www.spower.com.au/r100-series-high-600v-input-isolated-converter-100w but with 32V output work (this one is 24V).

I'm starting to think about a cheap device to replace the batteries and allow us to turn the lights on for display purposes in our carriage. 24V and 55A if all the lights were on (not likely). It may be cheaper for us to just get some truck batteries and a charger connected to the mains for the long periods when the car isn't moving.

I'm sure these are not cheap, but NO need to reinvent the wheel:

https://schaeferpower.com/content/products/dcdc-converter/

34 volts, 40A, 400-700 volts in, $500, in stock:

https://www.mouser.com/ProductDetail/Vicor/BCM4414VG0F4440C02?qs=sGAEpiMZZMvGsmoEFRKS8Koqt8Pjkl395UH5y7vpjkRmt5B2nKmGcg%3d%3d


Actually, looking at google, the market is full of DC-DC converters that work in the 600v range. Getting your output voltage exact might not be easy, but if needed you could feed an adjustable DC-DC converter.

Also, may be able to hack into a switching power supply. I have one made for European voltage that rectifies it's input into 350 DC, it would be nothing to feed it 350DC directly, I see ones made for 480 VAC which would be a 700 volt internal DC rail if it follows the same pattern.

If you are looking for longevity and simplicity, I must recommend AbsoPulse Electronics. Their units are designed for 600v railway service and have extremely durable and long lasting designs. We have been using their units for a lot of different presupposes and have not had one fail yet on our cars.
https://absopulse.com/dc-dc-power-conve ... s/#railway
The 600v units are toward the bottom of the page, if you email or call them they can build just about anything you need as well.

Yes, grounding the compressor through the batteries is one method, but not necessarily the most effective ...... If your batteries are of questionable integrity or the car doesn’t see regular service, just trying to keep lights lit for an extended period will be a major PITA. This is what has been done to Cars 34 and 40 at IRM, and it only works well because the cars do not see nighttime operation..... I believe car 19 which Has fluorescent lights got an MG set from a 6000 series CTA car..... certainly something could be fashioned, or try looking at 24 and 25 and replicating theirs which came out of the Paul Averdung era..... good luck

Thank you for all of the interesting responses so far!
Thirdrailcookbook:

In my time working in the barn at East Troy, WI. over
the past couple of years, I've slowly been uncovering a few more
bits and pieces of evidence and artifacts from the Paul Averdung
era there. I was so far unaware that the diners 24 and 25 came
from that era, but that is not at all surprising. In fact, it
does make sense and it fits right in with everything else.

Just to give a little background:
I did a stint there at that railroad back in 1982, through 1984
just before that group got booted out. Since then, there have
been several regimes, including the Averdung era. The present
incarnation has inherited everything cumulatively from previous
regimes. So, for me, it's much like an archaeology expedition
uncovering evidence and artifacts as I piece things together
to make sense of why and how everything has ended up in
its current state.

As for cars 9, 30, and 33, they all have a resistive charging system.
Those use some 4200 watt baseboard heater strips wired in series
as the dropping resistor. (I think it's 4200 watts each, if not exact,
it's a close number). That works the same as having the compressor
in series.

Presently, these are configured so that on a normal operating
day with headlight on bright, car lights on, the charging is
quite weak. Maybe nonexistent while running, especially since
the switchgroup contactor coils use a fair amount of current.

This crude system has proven to be viable because the one-way
run takes about 25 minutes, with a half-hour layover. This allows
headlight to be shut off and maybe car lights too, depending upon
what the voltmeter says. Keeping an eye on the voltmeter
does allow the more aware operators to function as the
badly needed regulator for this charging circuit. Of course this
is definitely not idiot-proof in any way, so as one might imagine,
there are lots of dead batteries and stranded cars from time to
time because not all operators keep an eye on things.

There is an idea floating around about installing some
regulation so that the charge current either diminishes
or completely shuts off when a full charge condition is
detected. This should prevent the boiling and extreme
heat and loss of water which the unregulated system
always produces.

Meanwhile, car 13 has the fluorescent lights, the Edison cells,
and a proper MG set.

By the way, that quote from Henry Cordell is one of my
all-time favorites. Indeed. "Who vas da motorman?"

Figured I'd bump this just to see if anyone has
any other ideas. Thanks.

You've just replicated, with modern solid-state components, what was a very common system.

The historically accurate system is a small "pilot relay" sensing the battery voltage, and that
operates the "line contactor" to switch the line voltage to the battery via a large ballast resistor.

Except that what is "accurate" for these cars is the M-G set.

A 1500V set is going to run at approximately half-speed on 600. You won't be able to do much
to change that unless you rewind the motor armature. The set turning at half-speed means it
will want to put out approximately half nominal voltage on the generator side. You might be able
to inexpensively get the output up high enough (36-38V to float charge a 32V battery) on the
generator side by reconnecting the shunt field coils. There are probably 4 coils connected in
series that come out to the "GF" terminal and frame ground. If you reconnect the field coils
internally to be in series-parallel, that will double the field excitation for a given voltage applied
to GF. However, because of magnetic saturation, the result will be less than double the flux.
Still, there might be enough margin there to get the thing to put out. Of course, the capacity
will be greatly derated both because of the reduction of input power and the reduced airflow
meaning reduced heat removal ability.

bump.

Resistance charging is only going to work if you run the control cicuit and headlight only. Thats why cars like 9 and 30 have their interior car lights rewired to work in series off the 600v. Even the air compressor governer was in a 32v relay circuit operating a contactor. 9 and 30 have been rewired to place the governor in series with the compressor, eliminating the 32v contactor circuit. It was done that way to always have the compressor working, in case battery volts dropped too low to operate the contactor. 13 was still on the 32v contactor circuit when i left ET. those s-16 governors were fine when in the low volt circuit, but they really arent rated high enough to operate a D3F compressor directly.

You really need a motor generator or a solid state inverter with a charger and low voltage power supply if you want to be running cars with all the xtra 32v stuff beyond control circuit and headlight.

i once contacted a company about building an inverter, they laughed when i only wanted a quote for 6 units. They wanted a huge order.

In the basment of the garage, was a crate of mostly GE PCC MG armatures. There may have been a couple Westinghouse ones mixed in the lot. There may hav even been a GE motor generator down there as well. An mg or inverter is only the real solution.

In the relay compartment of the cars is a device that looks like a clock. Im told it was old school battery regulation. it was set to a certain amp hour, and once that charge was reached, it cut battery out of the MG circuit. Not sure if its that, or just an watt meter. Dont really know.

Have to do what was done with the diners. Adapt a CTA 6000 PCC MG. 9 and 30 did have the adapted 6000 MGs, but they were taken down becasue they used to burn out frequently.When the blower cages are taken off, it lightens the load on the MGs, and they tend to overspeed, shortening their life.and they were needed for on the diners,so they were taken off and subbed with the resistance charger.

Most streetcars of the da had the dropping resistors for the battery charging controlled by a relay placed in the air compressor circuit so at least the battery wasnt always being charged, but even this really isnt providing much real regulation.

when the CTA made the 4000s into married pairs, they elimintated the charger from one car. as long as the 32 trainline is good, both batteries will charge from one charger. we used to run the 4000s by having one charger on, one off for one trip. then reverse the settings for the other trip.

i also saw your note on the batteries. when i was there and we dropped more newer batteries, we eliminated cells to get the voltage down closer to 32v. the diners may intentionally have their batteries higher than 32v to account for any sag when running all the extra 32v coffe pots etc that we used to use. not sure what you guys do for dinner runs.

bump

What is wrong with a DC-DC converter? Simple, fast, compact, reliable, clean power, and really they are cheap.

I have concerns with the idea of using a contractor and resistor to drop 600v to 32 or whatever voltage. If there is a break in the circuit(open cell on the battery), you will now have 600 volts on the entire 32vdc line. An appropriate DC-DC converter could actually work without a battery at all.

Also, your resistor will have to be HUGE. 600v-32v is 578 volts, so that's 578 watts of heat per amp of the desired charging rate. Typically I would double the watt rating when selecting the resistor. 4 amps of charge would be 2300 watts of heat to shed, also 2300 watts of heat wasted!

IMHO, if you can keep it original, do so, if you have to go with a more modern solution, why not just go all the way?

Pegasuspinto:

The present battery charging circuit being used on
South Shore cars 9, 30, and 33 is of a type and a
design that they had in place before I came back
in 2016.

I was there back in the 1980s, then took a long break
from it, and came back in 2016. Since I've been back,
I've learned of the unregulated resistive charging system
that's in place and being used now.

In the above posts in this same thread, I've already
explained in detail that the dropping resistors are
(as far as I know) two baseboard heater strips with
a power dissipation of around 4200 watts each.

These are wired in series so as to produce approximately
a 5 amp charge on the 36 volt lead-acid truck battery
string. The complete battery set consists of (2) series strings of
(3) batteries each. The two 36-volt series strings are wired
in parallel.

You are correct, I shudder to contemplate the result
from the whole battery ever going open! All of the 32 volt
stuff would suddenly have about 625 volts on it.
Worst case scenario: Lots of ruined equipment.

My guess is that (so far at least), the cost of replacing
these sets of (6) batteries rather frequently due to the
excessive heat and boiling which takes place due to the
nature of a totally unregulated charging system, is still
less than the cost of a proper commercial solid-state unit
such as a regular transit company would use.

Cost has been the limiting factor in all of this.
As far as what I can glean from all info gathered
on this charging circuit, the main reason it is being used
is that it's a fall-back that works when all other methods
fail. (Such as undersized MG sets with no fans nor cooling which
burn out rather quickly).

This unregulated resistive battery charging scheme has kept
these cars running and hauling paying revenue passengers
when all else fails.

A couple years ago I jumped in and decided to see
if there is any way that I can help to improve this
situation.

I am now looking at any and all ideas, across the entire
price spectrum. That cost range presently spans from zero
(i.e. do nothing and just keep boiling those batteries
and adding gallons of distilled water), clear up to
and including a proper off-the-shelf modern solid state
unit made for this application.

As far as "keep it original", that's not quite possible
due to the fact that these cars operated on 1500 volts DC
on the South Shore. Now operating on 600 volts at ETER.
The original 1500 volt to 32 volt MG sets were never
altered to operate on 600 volts. Again, cost has been
the determining factor.

Tear off a page from BART. The apocryphal story is that BART engineers were taking a tour of CTA faciities and found a gigantic storeroom of 600VDC motors awaiting overhaul. The motors were for compressors, A/C units, traction motor blowers, car ventilation etc. The BART engineers said "Oh, not us". They specced an inverter that made 480VAC from traction voltage. Then they specced commercial off-the-shelf 480V, 240V or 120V AC motors for every accessory from A/C packs to fluorescent lights. No such storeroom exists at BART, motors go directly to any local rebuilder who does AC motors all day all night.

So that's what I would do.

1) Get a 600V to 120/240VAC inverter, commercial off the shelf.
2) Get a common COTS 120VAC or 240VAC battery charger for the batteries you are actually using, preferably one with 3-stage charging.

Feed the inverter's power into an outdoor rated Siemens/Murray main-lug subpanel, with this interlock kit. The inverter feeds one breaker side, to the other side you connect a plain inlet so you can run it on shore power. Other accessories could get hung off this too.

By the way, if you want to achieve actual 32V via lead-acid batteries, some golf cart batteries are 8V, and can be had dirt cheap in the spring; golf cart batteries are unwanted once they can no longer make it 18 holes. A 3-stage charger *actually appropriate for 32V batteries* can be had from Quick-Charge Corp who supports 32V-actual (and 64V) in most of their models.

If you want to do it with Edison nickel-iron's, those are still made for the off-grid solar community, but they're pricey.

And here is a link to an inverter vendor product page that references using their product for "restored trolleys" ...