Flow sensor question

Does anyone here have any exeprience with mass flow sensors? We have a new application where we need to verify air flow in a 12” dia. pipe. The air flow is moving saw dust into a hopper where it is fed into our process. We sometimes have issues with the system clogging up want to have an alarm so we catch it sooner.

I have a found a few probes on the web that will produce a 0-5V or 4-20 mA signal that I can monitor in my PLC. Having never worked with any of them, I know nothing of their reliability.

If anyone has used anything like this, I’d appreciate any insight on which probes have worked well and which have proved to be garbage.

Thanks in advance.

your sensor will need to be rated class 2 div 1. and go through a safety barrier.

secondly, you will need some kine of shield to keep the sawdust from blowing into the sensor and destroying it.

i will have to talk to someone to see what we used in the past
for a similar project.

regards,
james

I've had VERY good success with Micro Motion and Endress & Hauser mass flows. If I had to chose, MM everytime. Most of my applications have been slurries, not fluidized powder, although there was one instance of using them in powdered limestone. I would suggest giving MM a call and discuss your application with one of their sales engineers.

We've had good results with Kurz and Endress & Hauser thermal mass flow meters on air service. FCI is good, but they have a slow response to flow changes - several seconds. FCI has specific flow switchtes as well as flow rate sensors. Endress & Hauser seems to be the best of the brands I've used.

If there is sawdust entrained in the air stream you may have significant problems. Most brands have a small (less than 1/4") gap between the two sensing elements, and some have a small window the air must pass through. I would be concerned about plugging. Putting a shield ahead of the sensor will distort the flow profile and may give bad readings. Definitely work with the flow meter manufacturer and make sure they fully understand the application.

It seems that you don't really want flow rate, just loss of flow. It might be possible to use a flow switch instead of a flow rate transmitter. It might also be possible to determine blockage with a pressure switch, possibly protected by a diaphragm isolator.

I agree with Tom. It sounds like you're just looking for flow, not necessarily flow rate. We use pressure switches or transducers at our blowers to monitor for blow-line blockage. If the pressure goes high, it indicates a blockage.

Try a negative pressure switch. Used to verify an exhaust fan is running. It's a diaphragm about 3" diameter, with a sensitive Microswitch attached.

Thanks all. That gives me plenty of decent leads. Tom is right that lack of flow is the main issue.

We were over at the plant yesterday and measured the pressure with a hand held Fluke 922. The measured pressure drop while running was only 0.8 inches water compared to reference atmospheric pressure, about 0.2%. I was worried that a pressure transducer might have trouble reliably picking up

In addition, management thinks that monitoring flow would be desireable for a couple other reasons.

It depends on how your system is arranged. When it's plugged up, do you have a positive pressure, or lack of negative? Is the fan pushing or pulling?

If all you need is flow or no-flow, then OZEE's solution is the ticket. I do this routinely on both blown sand and blown limestone processes. Sand is about as abrasive as they come (think sandblaster), so sawdust wouldn't be an issue, other than perhaps flammability issues and the need for a different Class sensor. You can even go further and use an analog style instead of discrete. High pressure indicates blockage, low or no pressure indicates something wrong with the air flow (broken compressor, blown seal, etc.)

Magnehelic makes pressure differential meters, switches and the like. We use them for filters and air conveyors. They just work on the difference of the up stream and downstream pressures.

Another way to go is sense motor current. It would be minimum with a clogged duct.
ASSUMING you are collecting dust from only one machine or source.
This may / would not work if the system is exhausting several machines and would possibly not sense if only one of sevaral ducts plugged.

Many of the sawdust systems I have seen have the fan downstream of the cyclone separator and teh fan discharging to atmosphere. You could place a sail switch on the fan discharge in this case and reduce (NOT eliminate problems with sawdust accumulation). Will still have the "erosion" from sawdust but that is just a fact of life with air transport of solid materials.

Dan Bentler

I have used Turck flow monitors in the past. I think it is better than a switch because I am always asking what is the setpoint, and we can never tell if the number is drifting over time - it's either tripped or not. No way to tell if it is getting 'worse' or whatever. My view is this is the same as a the pressure transducer as it correlates well with flow, if not indicating precise gas flow. However, it gives more info than a switch because we can see that the flow trends.