Sensing Sludge/Iron in Water

[TimWilborne 108]

I've got an application where I will be precipitating iron out of water. The iron will settle in the bottom of the tank. After so much iron is collected in the bottom of the tank it will need to be pumped out. Any thoughts on a way to sense the level of the iron without getting interference from the water? Plan B - Could I submerse a prox switch in the water to sense the iron at a specific level:? Thanks TW

[Ken Moore 23]

Well...... What's your budget? Guided radar level detection might be your best bet. With the higher end models you could measure the water and the iron at the same time. You have to "tune" the instrument to ignore the water and only read the iron. Any of the big name vendors can supply you with one. If there is the potential for multiple units, you may get them to let you trial one before buying. I had an application where I wanted to read the foam on top of a solution, I did not thing guided radar would work, but it did. Rosemount let me install and try the unit for 30 days. At the end of the trial period if I wasn't happy I could have returned the unit and payed nothing. Here is a link to Rosemounts offering: http://www.emersonprocess.com/rosemount/pr...vel/m3300b.html I do not work for or receive any compensation from Rosemount, but I do like thier stuff, and it's the unit I'm familiar with. Guided radar units are not cheap, I would guess somewhere in the 3-4k range ( I think I paid around $3400 two years ago). I could find out the name of your local Rosemount rep. if you like. You could also use Siemens, Krohne or perhaps Honeywell, I'm sure there are others, but we generally only deal with the big boys.

[andybr 5]

We use Endress and Hauser guided radar units and I am a big fan of these. Prices have dropped substantially over the past couple of years and you should be able to get a transmitter for quite a bit less than 3-4K$.

[kaiser_will 65]

I have had very good results with measuring tank level applications with Siemens Milltronics, Rosemount and Endress+Hauser guided radar systems. They are all pricey and all have very solid customer bases. Spend the extra time going through the application with your distributor to make sure their solution will work. I have found, in the past, that small system characteristics can cause the sensor systems to not be 100% reliable. Characteristics such as foam, turbidity, varying solution chemistry, ambient temperature change, etc. Do good to define the parameters of the application as much as you can. Another low-tech solution is (for blowdown applications, which this sounds like) is a flapper gate prox. As sediment builds up, the flapper gate arm starts to drop until the contact set drops out and the blowdown process is started. Downside with submerged contact devices is maintenance. This was a reliable, low-cost solution, but required us to do annual maintenance (not a problem since the whole tank was flushed and drained annually). We had experimented with venturi tubes for the blowdown and found good success with that non-contact approach (measure slight changes in the physical properties of the blowdown product...slow venturi meant high amount of sludge, while fast results meant clear solution). Some sampling logic was needed to store and calculate reference points.

[gbradley 0]

I wonder how much heavier the tank is with a layer of Iron precipitate on the bottom, compared to a clean tank? Probably not enough delta, but if it was enough, then perhaps you could just weigh it? Just thinking aloud. Edited 6 Jan 2009 by gbradley

[TConnolly 13]

Is this a static or batch reaction tank? Or is fluid being pumped through it during the precipitation process? If its static (fluid is pumped in and then batch reacted) then as the precipitate falls out the conductivity of the solution should change - so a conductivity meter may do the trick. If fluid is pumped through during a continuous precipitation process then as precipitation proceeds the tank weight will be a function of fluid level + accumulated amount of precipitate. In that case a fluid level sensor + tank load cell and a bit of simple math should give you the answer. Edited 6 Jan 2009 by Alaric

[TimWilborne 108]

It is pretty much a batch reaction. so the iron will settle to the bottom. The radar looks like it will definiately work, but I'm not sure if it would be justified. An operator will have to check on it occasionally, but I don't know whether this tank will fill wirh iron more often or less often than the periodic checks is yet to be seen. If it would eliminate extra check, it would pay for itself. If the operator nas to stop by 2 or 3 times between presses then it will be a waste. Did some testing and it appears I can actually use a prox on the outside of the polypropylene tank, only 2000 gallons so it is not very thick, and the iron will trigger the prox. While this doesn't give me an actual level, I will give me a point to send a message saying the tank needs to be pressed. I think this will work and is a down and dirty solution. but I need to do some more research. Keep the ideas coming though, I've got a few weeks Thanks TW

[panic mode 246]

in that case try analog proxy

[gbradley 0]

I don't know about Analog Proxy, but I'll bet that you can stack three or four standard proxys in a vertical line for a lot cheaper than that Radar detector. I have not had any luck with inductive proxys on Iron powder. I have used Capacitive prox switches with some success though. Were you able to pick up the Iron through the side of the tank with a standard Prox switch?

[paulengr 44]

I've had a lot of experience with thickeners/clarifiers before. Often it is difficult to tell where the "interface" is at. There's a lot of density type devices out there but I have yet to see a reliable one. One trick that I know does work is to look at impedance. Not sure how this will work with an "iron" sludge but essentially you need two cells. The actual devices are rings. Check the impedance on cell #1, and then impedance on cell #2 and compare the RATIO between them. I said impedance because this can be conductivity, or it could be capacitance (oscillator changes the response of the tank circuit and thus the frequency), or inductance (just a different shape for the probe). The reason for 2 cells is that often temperature/humidity/other factors will cause a shift in your circuit so you use the second cell as a calibration source (just like the "cold" junction for high accuracy thermocouples). The advantage of this approach is that you can easily start with your "iron sludge" and "plain water" in some pails in the lab and play around with different designs until you hit one that works. This same trick by the way also works excellent with detecting a sludge or foam interface vs. bulk liquid which is also notoriously hard to detect. IFM Efector by the way has a capacitance based level sensor that may work for you or may just give you ideas.

[TimWilborne 108]

Are you talking about point level measurement or do they make an analog version? They do have some nice capacitive point level measurement sensors. I think I'll email them

[paulengr 44]

Are you talking about point level measurement or do they make an analog version? I used a point-level version but I thought you could get an analog version of the same thing.

[DanW 41]

>They do have some nice capacitive point level measurement sensors. I think I'll email them I assumed you were looking for a continuous interface measurement, so I didn't chime in, but if you're looking for point level devices that can be incredibly sensitive to differences in aqueous composition, I can vouch for Siemens CLS 100 capacitive sensor. The situation was to determine when the resin beads in an industrial grade water softener dropped below a certain point. A bench top test amazed me. A large glass beaker was filled 1/3 with resin beads and then filled with water. I was extremely skeptical that any sensor could differentiate between water and water between resin beads. The CLS had no problem. We installed them and they worked flawlessly for the several years that company was still in business. It's an easy test to perform on a bench test water and iron sludge in whatever dilution is applicable. Dan