Home-Brew Macerator Pump

I’m taking a little break from the lithium battery project today.  Usually I don’t stray from hookups long enough to worry about emptying my waste tanks–it helps that I can handle 75 gallons down the toilet, and another 150 gallons of grey water–but for the first time in 10 years, I needed to empty my tanks and didn’t want to move the motorhome.

Where I’m parked, I’m about 150 feet away from a septic access.  I could get close moving the RV, but the tank outlet would be a little bit below the septic access.  So moving the RV doesn’t buy much, and the nearest RV park or dump station is about 20 miles away.

I’ve seen others with 12V macerator pumps and coils of hose, but was never impressed by them–especially considering the cost.  But if I wanted to empty my tanks without a drive, I’d need to pump.  I knew a garbage disposal could pump, and could grind up digested food just as well as undigested food, so I thought I’d give it a try.  A little searching showed that I wasn’t the first with the idea though, which just accelerated the project.

Note that I wanted to end up with bayonet fittings on the disposal, but where I’m at there isn’t one for sale within 30 miles.  Everything except the discharge hose was purchased at Lowe’s.

Here’s what’s needed:

Setting it up is pretty simple.

  1. Pull the sink mounting stuff off of the top of the disposal and push the cleanout adapter on.  While I haven’t done it yet, it’ll be held in place permanently with PVC cement.
  2. Connect power cord to the leads on the bottom of the disposal with wire nuts
  3. Install tailpiece, bracket, and gasket to disposal outlet.
  4. Install coupling on the end of the tailpiece
  5. Connect discharge and sewer hoses.
  6. Start disposal, open grey water valve momentarily, and check for leaks.
  7. Empty black tank and rinse with grey.

In my case, I was pumping through 150′ of hose, with a few feet of elevation gain.  By the time I had walked the length of hose and checked for leaks, the black tank–which had over a month’s worth of stuff in it–was empty.  Rinsing with grey water, the inside of the discharge hoses looked cleaner than a 3″ sewer hose ever does, and laid flat as soon as the water drained out.

I expect to make a few tweaks still to make things even easier.  First, I’ll set it up to directly hook up to a short flexible sewer hose, and mount the disposal in the wet bay.  I’ll also add a shutoff valve on the end of the discharge hose, so that when I’m pumping uphill it’s easy to disconnect without draining it at the RV end of things.  And I might also connect it to the tank level monitor I’ve been working on, just so that I can push a button to start pumping and walk away, knowing the disposal will shut off when the tank is empty.

Quick update:  It’s almost over, but I just saw that Amazon has a more powerful (1 hp instead of 1/3 hp) disposal on sale for the same price I paid.  It’s physically a little bigger, but has over 4,000 largely very positive reviews: Waste King L-8000 Legend Series 1.0-Horsepower Continuous-Feed Garbage Disposal, $87.97

A Few Observations on Battery Performance

This post is going to deviate a little bit from my typically long-winded format and present a list of observations and measurements so far.  In short, I’m very pleased:

  • The charger charges at a constant rate until it reaches the upper voltage shutoff
  • After shutting off the charger, voltage drops by less than 0.2V  (in 48) over the next several hours with no load.  This suggests that the surface charge effect is small.
  • When the charger is on, there’s no immediate spike in battery voltage.  Turning the charger on/off only influences the voltage by about 0.1V, meaning that even if it is on, we can still use voltage as an indicator of state of charge.
  • At each of several measurements, all cell voltages stayed within 0.01V of each other.
  • The self-discharge rate is (for all practical purposes) zero.  After 2.5 months in storage, the voltage was unchanged.
  • Charging or discharging a 4kW module at 1500W results in no noticeable heating of the battery pack, cables, or connectors.

Coming soon (not necessarily in this order):

  • Overall system objectives
  • Discussion of battery management/monitoring systems
  • High and low voltage disconnects
  • Over- and under-temperature disconnects

Charging a 48V Volt Battery Section

Now’s where it starts to get interesting.  We’ve got a bunch of 48V batteries that need charged.  We’re sitting at 43.8V, after who knows how long between the donor Volt’s death plus 2 and a half months of storage.  Do we really have a good battery?  I’m going to save the details of how we’re charging the battery for later, and instead focus on some of the battery’s characteristics, and what we’re looking for.

WARNING: Just like a lead-acid battery, there’s a lot of stored energy that can be released from this battery pack.  Unlike a lead-acid battery though, this battery pack can release its energy much, much, faster.  Extreme care is necessary when exposing the battery terminals.  Stray metal shavings, a dropped tool, nut, or bolt could easily result in an explosion from the rapid heating of the shavings, tool, etc.

Before we connect power to the batteries, a check of individual cell voltages is prudent.  For each 48V section, there are 12 cell packs (technically, each cell has 3 internal cells, wired in parallel).  We’d expect to see 3.65V on each, but more importantly, we want to look for any deviation.  This would seem to imply lifting the top plastic cover, as shown, but don’t go there just yet.

48V Battery module with cover removed.
48V Battery module with cover removed.

Removing that cover exposes the tabs from each cell pack, and the U-shaped connectors between them.  You can also see at the lower left and lower right the 10mm nuts for making the external electrical connections.  But in this configuration, it would be all too easy to drop something across the raised tabs, and very likely destroy the pack and severely injure yourself.  Fortunately, there’s an easier way to check each of the cells.  See the multi-pin connector on the right side?  Each terminal on the pack is accessible on that connector, and it’s a far better place to probe with a multi-meter.  It can be done with the cover on, minimizing the risk of nastiness.

When we do that on this pack, each cell measures exactly 3.65V, as we’d expect.  You can also see that there’s another 48V module physically attached in the picture–I want to charge both at the same time.  While the voltages for the two modules measure almost identically, prior to connecting them together I still want to make sure they’re at exactly the same voltage–even slight differences would result in a lot of current flowing between modules.  So the first connection is done with a large (9-ohm, 100W rated) resistor in series with the battery leads between packs.  Left for a minute or so, this is just extra insurance that there won’t be any surprises when connecting the modules in parallel.

The next step is to get the battery modules connected so that they can be charged, and to collect some data.  I’m not going to talk about the charging equipment yet (later, I promise!), but I do want to mention that–particularly on an inverter/charger–there’s going to be a fairly large inrush current to charge various capacitors inside.  To prevent sparks, and melting of terminals while hooking things up, I’ve connected the batteries through a disconnect, and used the same resistor from above across the disconnect terminals for a few seconds to charge those capacitors.

We’re going to be charging with a constant-current source.  It’s not a “smart” charger in the sense that it knows nothing about this battery pack.  It will have to be monitored, and shut off manually.  Technically, it’s a regular 3-stage charger, that will be finished before it gets out of the “bulk” phase of charging.  I’m going to be monitoring pack voltage constantly, and using a camera to log voltages with time stamps.  At several points along the way, I’m also going to be checking individual cell voltages, just to make sure everything operates as expected.

Volt SOC

The charging process runs just as expected, and we see in the chart above that there’s a very nice, linear relationship between voltage and time.  Since we’re charging at a constant current, we also know that voltage and state of charge are likewise related, with just a little bit of algebra.

With this particular charger and inverter, and sticking within reasonable bounds for battery pack voltages, the inverter would charge this 4kWh (3.2kWh usable) pack in about 5 hours.  What’s not recorded is the actual charger output (DC), so we only know state of charge in a relative sense.

So far, everything is going according to plan, and supports the tentative system design.  We’ll start to talk about that next time.

Taking Apart a Volt Battery

Ok, I know we’ve put a lot of text up with very few pictures.  This is where we start actually doing stuff and seeing what we’ve got.  This particular battery was purchased from a large recycler near Dallas, Texas, back in October.  I was in the area ahead of my first triathlon, and thought that bringing one of these home would be a good way to relax prior to the race.

Continue reading Taking Apart a Volt Battery

Why a Volt?

If you’ve been following along, you’ve seen why we want to make use of a 48V battery pack.  Now the question is a matter of where or how to get and/or build one.  There are three pretty obvious choices here:

  • Buy prismatic lithium iron phosphate (LFP) cells, connecting 16 cells in series.  This would be the route that normal people would probably consider first.
  • Buy a Chevy Volt battery out of a junk yard (or an “auto recycler” in politically-correct terms), open it up, and re-configure it.
  • Buy a used battery from an EV or hybrid other than the Volt

Continue reading Why a Volt?

Why does a 48-volt battery bank make sense?

Before deciding to use a Chevy Volt as a battery source, the plan was already going to make the switch to 48 volts.  We’ve talked about why the RV industry settled in on 12-volt house systems; now we’re going to look at what might be better.

What are some of the considerations when looking at battery voltage?  The higher the voltage, the easier it will be to transfer power between the source (batteries) and loads.  But higher voltage also carries greater risks–arc distances are longer, and higher voltages can drive more current through you if you happen to complete a circuit.  We also have various electrical codes that outline the practices for making connections and running wire (whether or not they’re mandatory in an RV).  Is there a sweet spot?

Continue reading Why does a 48-volt battery bank make sense?

A few words on Wire Sizing and Voltage Drop Calculations

Wire sizing guides leave out a part of the calculation important for figuring out the right size wire to use, particularly when dealing with low-voltage circuits where voltage drop can be a significant fraction of the total.   Standard practice is to use Ohm’s Law to calculate the current needed by a device, and look up that current in an ampacity table.  But that’s only part of the picture.

Continue reading A few words on Wire Sizing and Voltage Drop Calculations

How did 12-volt battery systems become the norm?

Perhaps the most significant change in this project is the use of a 48-volt battery bank.  There are a number of reasons for this, and we’re going to discuss them before looking at a new configuration.  Here though, we’re going to start with why we have what we have: an electrical system designed around 6 flooded lead acid cells in series–commonly referred to as a 12-volt battery.

Continue reading How did 12-volt battery systems become the norm?

New Project: Installation of a New Lithium-Ion House Battery Bank

Here goes nothing!  After a lot of planning, and several revisions, it’s happening.  I’ve started the process of replacing an aged house battery bank and inverter/charger with a new system based around lithium battery technology.

This isn’t going to be a simple remove-and-replace project though.  While there are a number of commercial offerings already available using prismatic lithium iron phosphate cells (LiFePO4, sometimes abbreviated LFP), often packaged with a battery monitoring system of some sort, we’re going to look to a much cheaper source of batteries, and a system design that should yield far better performance in the long run.

Typical LFP batteries run in the neighborhood of $1 per usable Watt-hour.  We’re going to be using a battery pack from a wrecked 2012 Chevrolet Volt as the starting point for this system–prices vary, but I’d estimate they’re out there for an average of $2100 or so.  That gets a pack with 12kWh of usable energy, or about $0.18 per Watt-hour.  That means that if we’re intending to use the full Volt battery pack that we’re saving about $10,000 just on cell cost.

This is a big project, and I’m going to be breaking down the design process into a bunch of separate posts.  But we’ll end up with a power system with energy storage equivalent to 24 golf cart batteries with about 80% weight savings, and less total cost than even a conventional lead acid setup (at 50% depth-of-discharge, we’re looking at 0.5kWh each).  We’re also going to be able to run some big loads (like air conditioners) without straining the system.  Stay tuned!