Another flow sensor

Optical mouse.

John

use microwaves? You might have to modify your discharge tube / funnel, though,

Why can't you measure the system's dielectric strength (averaged over an appropriate time). Wouldn't that tell you "blocked" vs. "heavy flow"? (Static vs. moving) Or am I mising something here....?

Option 2. Use a clear window, a video (frame capture) and just do sequential frame compares. (??) If the same image appears in several adjacent frames, then tube =3D jammed. If image not the same frame-to-frame, then tube =3D flow OK. Use a solid color target, so that if tube is empty (i.e., no flow), then tube =3D not in use or empty.

Option 2 sounds expensive, though.

Disclaimer: I dont' do this for a living. For all I know, there's an off-the-shelf solution. Did you check with Banner Engineering?

Best of luck.

-mpm

Put a scale under the container and monitor the weight of the container. When a jam occurs, the weight stops increasing.

Since there's a drop through free air, shine an LED on a tangent to the flow, through the edge of the falling stream. The photoreceptor signal should be noisy, and go thru a high pass filter. When the flow stops, the noise stops and the filter output signal goes to zero.

Dangerous Bill

OK now I like John L.'s sound idea, can you put some blockage/ turn in the flow and listen for the noise of the particles hitting it?

George H.

The same thing I posted for the water flow would work here - product stream flowing by carries heat away, product stream stopped up allows heat to build up (even better with your dry product). Put in heat (constant power, or pulsed power) and measure temperature, or put in heat to maintain temperature and measure current/power input. Look for temperature to rise in the first case, or power input to drop in the second case. Good thermistors can give very fine resolution.

Why wont the simple IR-LED photo system detect a jam? When the light is detected above a threshold for a given amount of time, a jam has occured because no (or less) material is flowing through the light beam. I don't understand why that simple solution wont work. Properly calibrated and compared with known values in a microprocessor, a quantitative flow rate could even be sensed if necessary. Keep it simple!

What is to be done when it does jam? Some powder flow systems use a vibrator to constantly vibrate the pipes to keep material from jamming. Rather than trying to detect a jam why not do something like a vibrator or tap-hammer to insure a jam doesn't occur in the first place. Keep it simple!

This is called a "Loss in Weight" metering system. To add this would cost $1-5 thousand. We thought about this when we bought our hopper/ feeder and decided to go with a volumetric system.

Why not just make sure it doesn't jam? Eg a slowly rotating rough shaft or screw down through the middle?

Cheers Tony

Reminds me of the rotating paddle type granular/powdered material sensor. Usually these stall when material level reaches the sensor level, but perhaps the same torque-monitored rotating paddle could both sense material presence and help prevent or sense jamming.

Ultrasonic doppler.

-- Paul Hovnanian mailto: snipped-for-privacy@Hovnanian.com

------------------------------------------------------------------ Life would be so much easier if we could just look at the source code.

Thanks to everyone that's replied. Each idea gave inspiration. I just don't see how I can implement most of them, though. Here's a picture, perhaps this will make the situation a little clearer. The hopper is on the left. Twin augers feed horizontally. The vertical cylinder and funnel are where the blockages occur, the augers go almost all the way across cylinder. I can add on a very small piece of tubung below the funnel, perhaps 1/2 - 1". Everything is rigid plumbing, and because of the relationships between parts, just can't be moved much.

picture:

I don't know if a single sensor will work. At high feed rates, the free optical path (a clear shot across the tube) goes away. We would like to run at as high a feed rate as possible. At lower feed rates, where material is just trickled down, the free optical path is intermittant, so a photo-type sensor and a missing pulse detector circuit would work. For the higher feed rates (where the light path would be blocked but material is flowing freely), I might need to add a differential pressure sensor. As I mentioned in the original post, the system is pressurized by from 5 to 10 psi, by nitrogen flowing from the fitting at the top (with white teflon tape at the joint where it fits into the top of the vertical section).

That is some awful welding :) or might just be the picture.

I liked the idea of using an optical mouse. Can you put a small window in? The optical mouse has a very small video camera that looks at the texture to detect motion and output pulses for xy motion.

Cheap enough to try.

Tom

It's a little of both. 316 stainless isn't the easiest stuff to weld, and the machine shop we used had only a part-time welder. We've since found a shop with a much better, full-time welder.

I just did some googling on hacking laser mouses. I may be able to look up the data sheets if I find identifiable ICs inside. Any optical sensor (well, any I can think of anyway...) will need a window, go this is a given. I'll try this next week and report back how it worked (out of system - it's going to be a bit before we do a tear-down and I have access to this part of the system).

Does the torque on your existing augers increase when there is a jam? Can you monitor the power required to turn the augers (power used by drive motor)?

One source of datasheets:

Avago Technologies

Look under "Products -> Navigation Interface Devices"; you should probably start with the "LED-based" sensors first.

Enjoy...

Frank McKenney

-- Politeness is a signal of readiness to meet someone half-way; the question of whether politeness makes society cohere, or keeps other people safely at arm's length, is actually a false opposition. Politeness does both, and that is why it's so frightening to contemplate losing it. Suddenly, the world seems both alien and threatening -- and all because someone's mother never taught him to say, "Excuse me" or "Please". -- Lynne Truss / Talk to the Hand

-- Frank McKenney, McKenney Associates Richmond, Virginia / (804) 320-4887 Munged E-mail: frank uscore mckenney ayut mined spring dawt cahm (y'all)

I haven't been following this thread too much and I don't know what kind of material you're transporting in that funnel? We work at a wire and cable facility where of course, have many hopper feeding systems.

Some of our systems employ vibrators on the funnels to keep the compounds from sticking also, we use load cells to measure the container for mixing and contents underweight warnings..

We also have one system that uses a capacitor sensor with analog output. That not only measures the absence of material, it'll also show material movement, because the analog output isn't steady and thus is cap decoupled through an amp to generate a rectified DC reference as a movement indicator..

I guess you have your choices..

In spite of drawbacks i still think a capacitance or permittivity type sensor may work.