Neptune Water Meter Software

[Zee Petty]

Longtrusted by thousands of utilities across North America, Neptune T-10 water meters are time-proven for accuracy and dependability even at low flow rates and provide a wide effective flow range for maximum revenue. The T-10 water meter is manufactured right here in America in our own foundry using proprietary technology.

The 3-wire connection between the two seems to consist of common, data (from meter to transmitter), and clock/power (from transmitter to meter). The transmitter apparently polls the meter periodically by sending a string of clock pulses to it. The meter responds by putting (usage) data on the data line, clearly clocked out by the clock pulses from the transmitter. I can see both the clocks and the data; now I just need to decode the data so my home monitoring system can use it.

Back when I hoped to be really non-invasive (never disconnecting the original wires), I made up something with spring contacts to connect to the screw heads of the 3-wire interface. That gave me a nice wire coming out with all 3 connections I could play with.

The only Schottkys I had were SMT, so I sliced two insulating slots in a bit of scrap copper clad, soldered the diodes on, and set it up so it would be easy to insert inline in the black (clock/power) and red (data) lines. (Green is common and will remain connected.)

I built an optical telescope looking at the spinner, and even got it to work and calibrated it in gallons used. It ran for a while, failed for a while, got fixed and ran a little more before final failure. Details are here.

Just thought you might be interested in the above website (not mine).

It describes a method using a hall effect sensor to bypass the meter mechanism and simply measure the rotating magnetic fields from inside the Neptune meter.

This provides a much higher resolution than the default interface.

Plus it has the benefit of not interfering with, or even connecting to, the wireless transmitter.

Jim,

Your hack may be incomplete. From my understanding, the account number (meter serial number(?)) should also be included in the read. This to tie the reading to a specific meter. I wonder if there is a command sequence from the uP that is necessary to fully implement the protocol? Incidentally, I understand the Sensus protocol is also known as UI 1203. This is some form of (unpublished) standard.

Also plausible would be that the commercial account mapping is between the account and the R900. That device has a 10-digit serial number, and could very easily include it in every transmission. And I saw reference in some Neptune doc to a 10-digit number in every transmission from an MIU (Meter Interface Unit). Maybe the same number?

If monitoring and/or recording water flow is your end-goal, you can do anything you want on your (the customer) side of the meter at most water utilities. You have indicated that you have a basement with the incoming water line, this would be an easy access point to the water line. There are a variety of flowmeter devices and sensors here: You could then get resolution suitable for your purposes.

Hi Jim,

I have implemented data capture with an ethernet Arduino for both the ProRead and the E-Coder versions of Neptune water meters. Your information was invaluable for the ProRead.

I would be happy to send you the design data and setup if you would host the files.

Jim: I have a Neptune T10 with a R900 V2 transmitter. The steplight group (website above) out of Australia has a monitor that would fit my needs yet it is not compatible with the Neptune. They told me I would need to put another water meter in series with the T10. There are a variety of e-monitors on the Chinese merchant websites yet the Aquamonitor has the best functionality for me. I am not a tech savvy as most of your followers. Has your research discovered any other devices similar to the AMRUSB-1 or can match the Aquamonitor ?

Dear Jim

Your work is very very interesting

Since your code allow Arduino generate the clock..so R900 should not be connected to the water meter. Please confirm.

Could you be more precise the electrical connexions betwwen Arduino and the meter?

Regards

I cannot figure out how to physically interface with the Neptune 3 wire meter when there is no R900, just the meter? Does the Arduino energize the meter and read the values or is a R900 needed to make the meter send data?

Hi Jim,

I love to automate just as you do and my next project was going to be to be able to read my water meter. My city also uses a Neptune meter but it is the older Neptune Auto H65N 5/8 T-10 that uses the Absolute Encoder. Unlike your older setup, there is no Neptune R900 transmitter. The city seems to use an induction system on the top lid that connects to only 2 wires of the meter (Red and Black are used, Green is NOT used). My guess is that they pass a coil over the top lid that energizes the meter and reads back the reading from the Absolute Encoder of the meter.

Were you able to wire directly to the meter and read the Absolute Encoder bypassing the Neptune R900? I want to be able to read the meter one routine intervals (Say every 30 seconds) but not let that stop the once a month visit from the meter reader that I assume passes a wand over the top of the lid that covers and it wired to the meter.

I would love to be able to discuss this with you if you can give me some pointers.

Sincerely,

Jean-Marie Vaneskahian

je...@vaneskahian.com

I see that you use Watercop as a valve connected to another ESP IC. It is a z-wave device that works with Samsung Smartthings. There is a wonderful lib that integrates the ESP8266 and ESP32 with Smartthings by Daniel Ogorchock ( _Anything) that you may want to explore You may also use it to create your own devices based on the same infrastructure.

The WaterCop is connected to a regular Arduino. Its Z-wave stuff is unused; I just wanted the valve. There are much better choices available, including this one Jean-Marie is using: =ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1.

The only reason I bothered with doing the Neptune decode in the interrupt routine was that I had other timing constraints in the main loop (serial interaction with the system that monitors the house).

I actually had a mistake. Looking again the T-10 has exactly the same protocol that you found including the 20037 in the beginning. The clock cycle is 854 μs and the signal charges the meter for about 107ms before it starts receiving back the data.

Similar hardware and needed to find the least disruptive method of disconnecting the transmitter to move both it and the meter during a remodel. It appears one only needs to detect voltage across the red and black wires at meter, then immediately disconnect, complete the move and get rewired within an hour.

This makes me wonder what would happen if the hour interval is exceeded. Nothing? Or maybe it triggers an error broadcast until reconnected? (Then, philosophically, if there were a meter distress call but no utility vehicle nearby to receive it, did the customer really tamper with anything to begin with?)

?? ?? lsb\ /msb (7 bits) \/parity bit (1 = odd number of 1s) [CRC? ] [CRC? ]

0010000011101000110111000001100110000011001101100101011011101011110111010001100000110011000001100110000011001100000110011011101000110000011001100000110011011001100110100111001100001110111000011101110101011001101010110011010011100110111011011101110100011000000101110111010001100000010111000000101110000001011101110100011000011101110001011011101100000011000000000110000001001100010100111

[ STX ] [ 1 ] [ 0 ] [ 0 ] [ S ] [ W ] [ ETB ] [ 0 ] [ 0 ] [ 0 ] [ 0 ] [ ETB ] [ 0 ] [ 0 ] [ 3 ] [ 9 ] [ 8 ] [ 8 ] [ 5 ] [ 5 ] [ 9 ] [ 7 ] [ ETB ] [ ] [ ETB ] [ ] [ ] [ ] [ ETB ] [ p ] [ 4 ] [ ETX ] [ ] [ ] [ ]

Our utility company provides an account I can log into to see my water usage in the same detail the meter provides (hourly). They might provide email with a less-granular summary. Have you checked with yours?

For Neptun E-coder and R, G and B pins; can somebody provide mapping of what the GND, clock and data pins are? On my meter, green wire is connected to R, black wire to G and red wire to B so it is all mixed up and confusing.

I was trying to capture clock coming from the reader, left it connected overnight and I was not able to detect any clock. Most likely, what I think is GND and clock is not correct.

quote; For a period of a few years the average quarterly usage was 875,000 gallons per quarter

You mean you actually paid the bill for that amount of water for "a few years"? And what happened to lower the recorded usage?

Edited 1 times.

this meter is in one building on a site with many buildings. the water bills and all bundled in their system and then paid. They DID pay this.

is there a way on this site to send a message to a particular person?

Had Town water installed a couple of years ago, never got a bill then got an outrageous water bill which I come to find out that the meter was not working properly. Town guy came over and just explained how a silent leak is probably the cause and went on his merry way. never performing any test. after his leaving I performed a very simple test. I filled a 5 gallon bucket and watched the meter. it said I used 20 gallons.

Got them to come back replace and take my meter back bor a bench test.

Now I have a proper reasonable water bill.

Good luck.

I too have a Neptune T10 5/8" Model. I began having VERY HIGH readings when they replaced our meter approximately 6 months ago - claiming they were having troubles with reading from the truck. Our Monthly bill had been $42-$48 for 18+ years. After installing this Model, our bill jumped to $140-$180 /month. They have replaced the meter but still indicating our water usage is 120,000-160,000 gallons/month and that we have a leak. IMO with that much water leaking we'd have a lake on our property. Compounding the problem is that the per gal rate increased the more water you are "using". I have had a Plumber/Handyman here and he/we cannot find any sign of a leak. I have tried one of the "tests" in another reply... filled a 2 gal bucket 3 times... Meter indicated that I had used 9.5 Gallons. All toilets have had "tank guts" replaced - NONE are running, hot water heaters are relatively new, in the middle of this have bought a HE Washer and Dishwasher... Getting very frustrated... wondering if the meter is defective or ??? By coincidence they had read the meter 2x in 2 days - in less than a 8 hour time span and it showed we had used 7,000 gallons over night. We do not have a sprinkler system and I am the only person here. Any ideas???

3a8082e126