False High Flow readings with Everydrop 1004-EX Wired Flow Meter (Possibly applies to all Flow Meters?)

Mine isn’t shared. It’s a separate run from the valves.

I thought about this and the theory doesn’t seem to hold water as the issue is only with a handful of the same zones out of 16 that experience the problem.

Yes. the slope is the problem. After your zones go off the lateral pipe drains out. Can be fixed by installing check valves in the low heads to stop that. Depending on how many heads need it can be done by looking at them after it shuts off.

Still, though, the flow ought to stabilize after a couple of minutes (“Pressure time” in the settings), slope or not.

Fwiw when calibrate I mine, which also has slopes, my flows stabilize in <1min, I have my “Pressure time” set to 3, and still occasionally get false readings (2 today: 1 high and 1 low). Really if you want to debug thoroughly, I think you need to “catch it in action” by observing the reading on the meter itself at the time the “high flow” triggered. Then you’ll at least know if the meter is giving a false reading or the Rachio is somehow perceiving a false reading.

I’m tempted to point a video camera at mine, but it’s not repeatable enough vs the cameras I have available to be worth it (i.e., I’d need to invest in a camera with better features). One might also be able to get a Raspberry Pi to record the pulse timing and monitor at both ends by moving the Pi, but my Pi skills aren’t sufficient at this point to make it an easy project.

I missed replies, better late than never?
One thing you could try, is connect a temporary wire from flow meter to controller just on top of the ground. See if the problem goes away.

Twisting and shielding are both designed to minimize the effects of signal noise. Twisting cancels out noise, and shielding prevents it from getting to the wires (shield has to be grounded at 1 end). In industry, you would actually use both, called shielded twisted pair(STP). Whatever cable you use, it should be rated for outdoor/wet locations, typically indicated as UF-B. You can’t use any old ethernet cable or wire you have laying around if you want it to last more than a couple years. If you wanted to test with it, I would use 2 pairs as the wire is very thin 23awg for CAT6 and potentially significant voltage drop.

There are ways to bore under a sidewalk. In future, its wise to install a pipe under sidewalks called a “chase” so you can add/replace wires in future.

1 Like

Btw the problem seems to have gotten worse for me. I haven’t done full analytics yet, but it seems like I’ve got a 5% false alert rate, roughly 1/2 high-flow and 1/2 low-flow. If I was the person writing the software (and I am a software engineer and am familiar with issues with time-series monitoring data), I’d alert off of a moving average requiring a minimum number of samples. I suspect they are getting a pair of bad readings (high followed by low or low followed by high) and taking it as gospel, which is a classic formula for causing false alerts.

Also go home, Rachio, you’re drunk:

1 Like

0.7 gpm seems like it might be a drip station… I’ve seen this behavior with the 1004-EX meter. Below 1 gpm is generally real close to the lower threshold of sensitivity - especially if there’s insufficient back-pressure from the station leaving air in the vicinity of the meter. (My case, the meter is ABOVE the valve.)

Thus, if you’ve calibrated the station at a low flow number, ANY slight variation or anomaly in flow (ie: air) through the meter can appear as a drop or surge in signal. It IS ridiculous that Rachio reports an error AT the high limit, but you can manually set the high flow limit to a value that you might expect with a real leak (several gpm?). I decided to plumb in a 1" PVC gate valve about a foot downstream from the meter and very gradually tweaked it closed a bit increasing back-pressure until meter readings (at all flows) smoothed out.

I second the advice here about wiring for the 1004-EX. Use the right wire - shielded twisted pair for direct burial. It’s sometime called “sensor cable” or “Toro cable”. I got mine from SiteOne. Cat 5 or 6 isn’t usually made for burial.

I agree with JBTexas that 0.7GPM is a very low rate to try to read. The spec is >1.0GPM and only useful for leaks below 1.0GPM. Especially since this is set as your high limit, meaning your actual reading is less than that. I would probably set the high limit at something like like 2GPM or whatever flow rate you would get with a broken line somewhere near the end (the lowest flow rate broken/disconnected parts would cause).

I don’t know how many samples are needed to trigger, but I think this unit is around 1 pulse per 0.1GPM at the low end. Also, you typically need 10-30x the pipe diameter upstream and half that downstream from the meter for least amount of noise from water turbulence, however the spec sheet says it has it built in.

1 Like

I am tempted to pick up some shielded/twisted wire from Siteone and lay it on the lawn to test if it clears up the false readings. If it does I’ll bury it. What do you think about this stuff?

@MeterManSays - what are your thoughts and experience?


Every situation needs to be examined individually because there are multiple things that can manifest itself with variations of these particular symptoms. (which is why this thread is going to confuse way more people than it is going to help) I would reiterate my advice from June 13 up-thread which was to contact the ED support line. Also, all troubleshooting begins with watching the meter display during pressurization time to see what the meter is actually seeing.

Let me rephrase the question @MeterManSays … which wire is recommended for the average ED installation? Thanks

The question of the quality of the wire, I would say its appropriate wire. Its direct burial rated (good), its foil shielded (ok), its copper wire (good), its 18awg (good).

Given its $1.20/ft, whereas Ethernet runs much cheaper, IF you wanted to test for a week or month above ground to confirm, you COULD use ethernet, then buy this cable for the permanent. Save a buck if this doesn’t solve the issue.

As an industrial controls engineer that doesn’t do sprinklers professionally, replacing the cable to test is the next step I would do after the advice of MeterManSays of watching the display on the flow meter to confirm it does not match what Rachio says during a Calibration run, accounting for some delay/averaging that Rachio is going to do to that signal.

This is a great idea. I’ll see if I can round up a long piece of cat5/6 and just set it on the ground for now. Is it also shielded with foil or similar?

They make shielded CAT, but its much less common, and not necessary for testing. Its twisted pairs though. Being on the ground, not near another wire, its far less likely to get interference anyway. Ideally, you use a set of pairs for your wires, but since its very small wire (24awg for CAT5, 23 for CAT6), you may need to use 2 pairs (Blue + Green for AC+, and Blue/Wt + GreenWt for Signal).

Can’t you also make the argument that regular thermostat wire is also in/on the ground, not near another wire and less likely for any interference? Especially at short distances, 40’ or less?

I don’t know what the original wire was run with. Most likely culprit of wire interference is that the insulation is somehow damaged and is shorting.

I hear about interference all the time in my job, but never actually witnessed it in 10years until last week actually. A 24VDC unshielded/untwitsted signal wire ziptied to a cable with 208VAC both running to a chiller about 75’ and the signal wire intermittently got enough induced voltage to register a shutdown signal, happened about once every 6minutes.

I’ll repeat my recommendation, after more than a year with the installation and testing of 2 different 1004-EX’s. Get cable that’s made for the application - sensor cable (which you’ve already found). Thermostat cable won’t cut it, non-buriable CAT-5 or CAT-6 won’t cut it, non-twisted (like sprinkler valve leads) won’t cut it. You’re going to bury this cable in your yard - probably at a depth of 6" to 12", and you NEVER want to have to dig it up or repair it. At about a dollar a foot (about a dollar and a half for the armored version), I would have saved many hours if I’d installed separately buried sensor cable from the start.