Getting two (or more) pulses per dial rotation

Forum is open to all questions, answers, and discussions related to water meters, ours or others.
Post Reply
medmunds
Posts: 2
Joined: Sun Jun 18, 2017 11:00 am

Getting two (or more) pulses per dial rotation

Post by medmunds »

I have a new EKM 3/4" water meter (SPWM-075) which I'm connecting to an Arduino, and I'm trying to figure out the pulse output.

It seems like each rotation of the 0.1 cu. ft. dial generates at least two pulses: a short pulse lasting ~5% of the rotation, then a gap of ~2%, then a longer pulse lasting 45% of the dial rotation. Sometimes the long pulse also has a very brief gap (~0.2%) about 5% of the cycle into it.

[EDIT: this turned out to be a bad multimeter -- not a problem with the water meter.]

Here's a picture of the pulses from two dial rotations (at a flow of ~43 cu. ft./hr):
ekm-spwm075-pulse.png
ekm-spwm075-pulse.png (3.42 KiB) Viewed 23244 times
Two questions:

1. Is this normal? I understand reed switches can be bouncy, but the gaps are well after the initial closure of the contacts.

2. Any suggestions for getting an accurate pulse count from this? I'm having trouble figuring out how to reliably ignore the short pulses or the gaps.

This is for a multi-zone irrigation system, so the flow rate won't be constant. At 5GPM, the short pulses are half a second long, and the gaps 0.15s. Also, it would be relatively easy for the water to shut off in the middle of a short pulse. (All of the de-bouncing approaches I can think of would get confused by that.)

Thanks for any info and assistance.
Mike

P.S., Here's an audio recording of my continuity meter hooked up to the water meter pulse output, covering several more rotations of the dial. The extra pulses and gaps are really easy to hear.
Last edited by medmunds on Thu Jun 22, 2017 3:50 pm, edited 1 time in total.
Jameson
Posts: 860
Joined: Fri Nov 04, 2011 7:42 pm
Location: Santa Cruz, CA
Contact:

Re: Getting two (or more) pulses per dial rotation

Post by Jameson »

Hello Mike,

Very interesting. Thanks for the audio clip. I have not heard of this before to be honest. It seems like more than normal signal bounce due to the reed switch, though I cant imagine why your meter would be any different. All it is inside our water meter is a magnet on one of the dials. The magnet pushes a reed switch closed for every rotation. I cant imagine why the magnet would push the reed switch closed twice for every rotation.

I tried to reproduce what you are seeing on a random water meter here, but I could not reproduce the signal gap using an oscilloscope. The pulses you see in the attached images are a few seconds wide.

All I can suggest is to swap your water meter for a different one, maybe your reed switch is too far from the magnet and gets some weird affect when it passes through the outskirts of the magnetic field.
Screen Shot 2017-06-21 at 12.47.23 PM.png
Screen Shot 2017-06-21 at 12.47.23 PM.png (374.39 KiB) Viewed 23230 times
Screen Shot 2017-06-21 at 12.47.03 PM.png
Screen Shot 2017-06-21 at 12.47.03 PM.png (278.84 KiB) Viewed 23230 times
Screen Shot 2017-06-21 at 12.48.03 PM.png
Screen Shot 2017-06-21 at 12.48.03 PM.png (329.61 KiB) Viewed 23230 times
Jameson
EKM METERING
http://www.ekmmetering.com
831.425.7371
medmunds
Posts: 2
Joined: Sun Jun 18, 2017 11:00 am

Re: Getting two (or more) pulses per dial rotation

Post by medmunds »

Hi Jameson,

I really appreciate your looking into the this. After seeing your response I wanted to test my meter again, and hooked it up directly to the Arduino (rather than using my multimeter as before). And it worked: perfectly formed pulses, with only a few msec of bounce where you'd expect at the transitions. No weird gaps mid-rotation.

Apparently the continuity beeper on my multimeter is, um, not so accurate.

I apologize for sending you on a wild goose chase. The water meter pulse output is just fine.

Mike
Post Reply