Motor Control Design Question

[BobLfoot 302]

I have a dilema and could use some help from my firend at mrplc. Here is the situation. I have a ventilation fan driven by belt from a 3 phase AC motor. The motor is driven by a VFD and the VFD controlled by a PLC. We need to be able to know when air is moving. Have tried several "sail" and pressure switches. The variation in air pressure and speed between min speed and max speed have made it near impossible to find a "sail" switch or pressure switch which works. I spoke to the environmental engineer and he has agreed to the following. 1. We already have the following signals. A. Start signal to the VFD from PLC. B. At speed signal from VFD to PLC. C. Faulted signal from the VFD to PLC. D. 3 Phase Circuit Breaker Aux to PLC. E. Local 3 phase disconnect to PLC. 2. He has agreed that if we add a belt moving or belt not broken signal into the PLC this will suffice as confirming air is flowing. Anybody have ideas. I'd prefer a digital input and avoid encoders and encoder cards.

[P Daniil 0]

If possible, I would detect current, to the VFD or the load, as a measure of the action taking place. Most probably this is already available in the VFD as part of current monitoring for overloads/faults and can be used to signal the PLC for good/bad operation via a VFD bit output. If such a facility is not available in the VFD then you will have to measure current to the VFD with an external current transformer at the PLC (and unavoidably use additonal bugdet/space resources). An additional benefit of current detection is that you also detect power train problems such as a broken/loose belt in your case. Edit: Another less accurate alternative is to pick fan-blade movement with a proximity switch to estimate fan speed. Edited 9 Jun 2007 by P Daniil

[comeng 0]

I agree wholeheartedly with the above we use extensively Telemecanique ATV inverters on fans and can set the under and over current/power figures accurately enough to generate alarms to use them for belt break detection, bearing seizing and flow proving Edited 9 Jun 2007 by comeng

[TimWilborne 108]

How about a differential pressure switch. I'm not sure that is the technical term and don't have a part number handy but we used to use them to see how clean our filters were on our powder coating lines. Pretty much it is a pressure switch with two lines. One would go before the fan and one after. Wait, why don't we get real simple. Mount a prox that is triggered off of the fan blades or blade pulley. Timers for prox made and not made. If either times out then the fan is not moving. How fast is the fan itself moving at max speed?

[paulengr 44]

Assuming it's a centrifugal fan... motor amps are monotonically related to air flow. Monotonic is a big fancy word that means although it's not linear, an increase in amps does correspond to an increase in flow. If a belt breaks, something plugs up the fan, fan rotor flies apart...whatever, you will see a corresponding response in terms of load (HP) and hence amps on the fan. You could run the VFD up in steps and plot the curve if you really wanted to, or just monitor the amps at various conditions. It is reliable enough that you can even use a pitot tube and develop an amp-air flow conversion. You're not accounting for the zillions of inaccuracies and calibration problems but it's a very good approximation (within +/-10%) under most conditions. This does work. A certain large foundry with a 75 ton/hour cupola actually runs their primary induction fan system with this method. And I'm the engineer that gets called every time it has a hiccup. If you want better, consider a simple mass-air flow sensor. Omega sells a pretty decent one for about $1K. It will directly give SCFM as a readout. It works by running a small inductive heater on a thermal mass out in the air stream. The feedback is an RTD. The temperature drop due to the air moving across it is directly proportional to the mass flow of air. Very reliable as long as you don't have too much sand/dust/dirt which takes the sensor out. I've used them several times, both handheld and permanent mount. If you want the top-of-the-line model which is a bit more rugged, but much more expensive, check out "Kurz". Moving up the line, if you can tolerate putting an orifice plate in and you don't mind the really ugly equations, and you get a dP cell, this also works very reliably. The blast air (intake blower) for the previously mentioned cupola uses this technique. Finally the ultimate in air flow is an "averaging pitot tube". This is basically a "cross-bar" type arrangement where multiple pressure ports are drilled on the tubing. Again, dP cell is the sensor.

[Clay B. 1]

Differential Pressure switch is the most common I have seen. Add a piot tube in front of the fan and your in buiness. You do not even have to a port on the inlet side unless this is a closed loop system. Just leave the other side to atmosphere and go. If you use a piot tube with an analog Differential Pressure Switch you can measure flow and use it to set fan speed. Problem with using current and a prox is that if your not pulling hard enough the system can be fooled.

[andybr 5]

Endress and Hauser also do a good thermal mass flow meter similar to the one described by Paul but if you want a much simpler and cheaper sensor which uses the same principle the IFM Efector units work well. Andybr

[milldrone 0]

Square D makes a motion sensing prox that would give you a digital input without having to use a high speed counter I believe the part # is XSA V31373 Vaughn

[BobLfoot 302]

Thanks I'll check out SquareD Site. I'll also look at the diffP sensors, but with the size of these fans 3 to 5 ft dia and in some cases squirrell cage units and the speed variation 20 hz to 90 hz on the vfd the pressure band is so wide we have not found a good sensor.

[TConnolly 13]

Bob, in additon to the recommendaitons give already, check out these: http://www.ifmefector.com/ifmus/web/padv0&...2_10_10_10.html They make a thermal dispersion air flow switch that may work for you.

[paulengr 44]

That is quite surprising. My usual problem is that the pressure range is TOO wide. Ignoring roots blowers and centrifugal air compressors, most industrial fans never get above 40" wc. Above 20" or so, it is hard to realistically make a fan do that and be economical (very high tip speeds). At those kind of air flows, it tends to push you towards larger ducts and bigger fans, which reduces the air speed back down to something reasonable. It is very easy to find a 200" wc dP cell with sensitivities of a few tenths of an inch or less. It is much harder to find one with a narrow range. You might find something less expensive but check out www.omega.com. They sell a huge variety of ranges. Based on your post though, are you trying to find a "universal" sensor? That may be the mistake. If you have a variety of fans and blowers, you may need to have 2-3 sensors to cover all bases and not just a single more expensive one. Second, ignore the ones that give you specs on PSI for the same reason. Fan pressures are quoted in inches of water or mm of mercury for a reason! The pressure you are trying to measure is very small and not the sort of thing that can be measured with a typical 0-100 PSI gauge, or even a 0-20 PSI one. Second, are you mounting it downstream of the fan in the side of the duct, or is it remotely mounted via a bunch of tubing? Sometimes those remote mounts leak and you never get a decent reading until you find and eliminate the leak.

[Ken Moore 23]

I have a similar application and I'm using a Rosemount DP transmitter, it's the lowest ranged one they have -10"WC to +10"WC, I don't remember the accuracy off hand, but I respanned it to -1 to +1, and am able to detect 0.1" changes.

[BobLfoot 302]

I am curious Ken on what diameter ductwork did you mount your rosemount and what model. Everything I saw on their website was for fluid sensing and not air flow. Maybe I was at the wrong site?

[BobLfoot 302]

Looks like this might do the trick. We'll see.

[Ken Moore 23]

In my application the duct work is 36" square ducting, the transmitter is connected via 1/2" s/s tubing. There are 1/2" pipe couplings welded to the duct work. BTW, air is a fluid, and pressure is pressure.

[BobLfoot 302]

Correction to my earlier post - Dense Fluid Flow ie liquids and compressed gas not near atmosphere pressure air.

[mfkuss 0]

When using the AB powerflex series ac vfd's we have always used the analog output from the vfd to the plc to see the torque which will correspond to airflow

[geniusintraining 2]

Hi Bob, (and others) I would measure SCFM, I do this on several applications, a simple piot tube and a calculation in the plc would give you your SCFM, it's very accurate and dependable parts used: Dwyer 607-3 or Ashcroft XL-3-F02-42-ST-1IW <----- the Dwyer is made by Ashcroft Comptus track-mount barometric transmitter This is the code that I used, its on a 504, we tested it against several hand held calibrators and it was very good, I display it on a HMI and also use it to control fan speed EDIT: You will not need rungs 1-3, this is my HX, also you may need two transmitters to get your range that you are looking for or not... I run from 0-45"WC Edited 17 Jun 2007 by geniusintraining

[BobLfoot 302]

SOLUTION CHOSEN AND APPEARS TO BE WORKING WELL. Using IFM EFECTOR SLC 301 sensor. They tune in really quickly with screwdriver and 10 minutes. Mount quickly with a 1" hole saw and two self tapping sheet metal screws. Time will tell how they hold up. For a $125 sensor I am very pleased so far.