Level Measurement for a difficult application 7

[BizzyCon]

Hello everyone!

I've been racking my brain for a bit trying to figure out the best option for this problem.

I have an application with a very viscous resin and a powder catalyst being mixed in a small vessel (a very small one I will elaborate later). I would like to know the level on this vessel but there are a few problems:

1) The vessel can't be too big (I'm thinking maybe 4-5 inches of fluid maximum) because we want to minimize the residence time of catalyzed resin.

2) It would be best if the sensor is non-contact because catalyzed resin will build up on any type of contacting device

3) The catalyst is a powder and makes things very cloudy inside the vessel.

I have thought of the basics, laser (worry about particulates), ultrasonic (also worry about particulates, however maybe this isn't a big problem?), capacitance (rejected because of contact with fluid), Radar (this on seems the best but where can I find a small one?)

I guess I'm in between ultrasonic and radar but I'm not sure which is best. I'm a little worried about the dust with the ultrasonic but like the price and size (i've seen smallish ones). Is there a good small radar out there that will do better? Is this even the right direction in your opinion? Any help would be appreciated. Thanks!


- Bizzy

 

[IRstuff]

Is this powder so light that it floats in the air? If not, then you probably should reconsider a laser, since you really don't need to see through the mixture, and would most likely prefer not to.

A variant on the laser is a line line generator that's imaged by a camera. The image is processed to determine why the line image is distorted, and the result is what surface profile is. Since the laser profiles along the entire line, you can mitigate surface irregularity impact on the level measurement:

http://www.bytewisesensors.com/content/products/crosscheck/index.php

There are also 2-D laser profilometers:

http://www.melsensor.de/laser-scanner-m2ilan2-en/m2-ilan-product-overview.html

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[BizzyCon]

@IRstuff,

Yes the powder floats in the air. When I said it makes things cloudy I meant that the inside of the vessel has powder floating all over the place. We have to keep the vessel covered so that powder doesn't get all over the place.

Now that I think about it, the resin becomes cloudy in appearance as well..... that may have been the source of confusion



- Bizzy

 

[IRstuff]

A pulsed laser might still do the job, since you can often set those types of rangefinders for last return only. Also, obscurant obscuration is highly dependent on wavelength and particle size, so some of the longer wavelength lasers like 1571 nm, might still work OK. I'm not sure, however, how available those lasers are.

I wouldn't necessarily rule out capacitive sensors. Seems to me that there ought to be a way to protect the sensor head and still get it to work for your application.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[BizzyCon]

@IRstuff,

Thanks for your suggestions so far. Any ideas who I could talk to (vendor?) that would be able to work with me through this problem? I don't have much experience with lasers.

If anyone else still has ideas I am open to hear them. I'm still trying to find a solution. Thanks!!

- Bizzy

 

[danw2]

My two experiences with industrial laser level didn't work due to dust. The laser beam is too 'narrow' and couldn't get through the dust. One was done by ultrasonics with a much wider beam, the other by radar that wasn't as sensitive to dust. But these were 20' talls tanks.

Banner engineering has some small ultrasonic transducers that might work better for a short shot on your small vessel than the Milltronics/Siemens industrial monsters intended for long shots on silos and large tanks.

http://www.bannerengineering.com/en-US/products/8/Sensors/30/Ultrasonic-Sensors

 

[ScottyUK]

Erwin Sick make some of the best sensors in the business. Definitely worth a call to their applications engineers.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[controlnovice]

Is level the concern to prevent overfill?

Or would measuring the weight of the vessel and contents suffice?

With a weigh scale, there is no intrusion/entrance into the vessel to determine its contents. Everything is external.

______________________________________________________________________________
This is normally the space where people post something insightful.

 

[BizzyCon]

@controlnovice,

Level is the concern for overfill. I though about weight too but this hopper would be attached to a ~300 pound mixer and there would probably be less than a pound of resin in the hopper. I worry about being able to get good data from it.

@danw2,

Thanks for the ultrasonic suggestion. I will take a look. Do I need to worry about and ultrasonic transducer with the particulates in the air? At what point do the particulates interfere with the reading? Thanks!

- Bizzy

 

[danw2]

>At what point do the particulates interfere with the reading?

I don't know the answer for that. Operation in dust environment with the large tank/silo instrumentation was/is purely empirical. Getting out of a fill stream was a must. other dust could be handled in some amount. Radar tends to be better at shooting through dust, but I don't know who makes a 'short shot' radar.

I suppose there's some suspended-solids-in-the-air measurement, but the vendors don't supply such a factor, probably because the material makes a difference, the distance makes a difference and most people don't have whatever instrument it is that can take the measurement, anyhow. It's more practical and easier to plop in a transducer, power it up and see whether it works.

I haven't priced them, but I suspect that those Banner units aren't beyond the price where one can be bought just to try it.

 

[glowing]

the laser will not likely work...has not in my experience.
The scale might, but I am guessing there will be piping that will scew that measurement.

I would consider a Nuclear gauge. I have had to use this in similar applications in the past.

 

[DanStewart]

The first consideration should be what sort of resolution and repeatability are you looking for?

If required resolution needs to be extremely tight, a load cell will probably be the best option here. Good ones have accuracy to within 0.03% of full range.

Radar does not accurately read below 6 inches of liquid as it pierces the material and reflects back the bottom of the tank resulting in a constant zero level reading. That eliminates it entirely.

Ultrasonic that resides on the lower end of the frequency range is capable of handling dusty conditions, but finding one is difficult. In general, ultrasonic resolution isn't very accurate for such a small span, so count it out if you need high resolution.

You also mentioned this tank is mounted on a mixer. Does the level need to be measured during agitation? Yes? Then resolution of laser and ultrasonic will be highly diminished.

This brings us back to the load cell. Highly accurate and widely used in batching operations just like this one.

 

[Medeski]

I'm in no way an expert in measurement devices, but Balluff makes a programmable electronic eye that seems pretty versatile (it caught my attention when I was looking for information on a Balluff rotary cam encoder we use). I'm not sure if it fits your application, but it might be worth a look.

http://www.balluff.com/NR/rdonlyres...FA486/0/881076_212855_BVSbrochure2010_WEB.pdf

Best,
Medeski

 

[bjoh]

Soembody might have already posted something like this, I didn't read the whole thing.

Endress and Hauser has been good to me in the past for an ultrasonic indicator. I've used this one before on a couple of powder dilution systems where the vessel gets pretty cloudy and it never seemed to be an issue. The only thing I would ask them about is if it would be accurate in such a short range.

http://www.ca.endress.com/#product/FMU40

 

[danw2]

You can do this with load cells. The former GWT, now Sartorius, German origin load cells are truly in class by themselves. They will provide fast, non-averaged readings of a fraction of a pound; even with 1 pound load on a 300 pound mixer.

Check out the first 90 seconds of the video here

http://www.lesman.com/train/lrvids/sartorius/sartorius_load_cells_lq.htm

that shows the instantaneous response to the addition of a 500 gram weight to a 1500 Kg mixer on load cells. It is not your run-of-the-mill load cell.

claimer: this product is sold by my company.

 

[ornerynorsk]

Can you meter the inflow of product instead?

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[lacajun]

I can read jmw now but.........

If you have the time and inclination, look into Coriolis meters to measure the mass flow of catalyst and resin into your vessel. I would only use Micro Motion for this application. But, you can look into E+H, Krohne, Yokogawa, Foxboro, etc.

The dust created by the catalyst and the size of your mixing tank are going to prohibit the use of many technologies such as ultrasonic and radar. I put ultrasonic on some "non-dusty" product tanks many years ago and the dust was a huge problem. I didn't think there would be enough dust but there was. Operations was unaware of the dust created by the finished product. Radar wouldn't have worked either. I've heard recent stories of radar signals being effected by too much dust. They didn't quantify how much was "too" much. I've also heard stories of vapor effecting radar but, there again, don't have quantifiable information.

The way I understand your setup, small hopper attached to ~300# mixing tank, you'll need to have an excellent mechanical installation for any load cells to prevent measurement interference. Then again, you may be able to overcome this with experience and trending rather than precision measurement. Some processes run this way quite successfully.

Do you need to account for the angle of repose of the catalyst? Don't know your process but it may be something that needs consideration to measure the true quantity of catalyst in your vessel, with distance type instruments.

You could use a Coriolis meter on the resin and maybe an auger on the catalyst. Some paint manufacturers use augers to meter pigment into mixing tanks. I've never reviewed them and cannot comment on accuracy, size, flow rates, etc. My plant produced smaller volumes and we couldn't economically justify the equipment used in the larger plants.

You may be able to convince your suppliers to provide the quantities needed prepackaged thereby eliminating any measurement needs. Unless you want/have to buy in bulk. I would review the economics of those two situations.

Those are some thoughts that popped into my mind. Yours thoughts may vary.

 

[jmw]

Lacajun, taking my name in vain?

Why do you want to know the level?
Overfill protection?
As I understand it, this is a batch reaction and from what you say, after each batch you presumably have to clean the vessel.
So if you start with an empty vessel and charge it with pre-measured amounts of each component, how can you overfill the vessel?
Once charged you seal the vessel and then agitate.
When you agitate you won't have a stable level anyway.
With a lid, nothing will escape during mixing.

On that basis the problem is really how to accurately charge the vessel with each component. You could then use load cells, meters or whatever.
And, presumably, once charged and before you start the agitator or fit the cover, someone can eyeball the contents?

But, if what you are trying to do is evaluate how the behaviour in the vessel during mixing/agitation, then that's something else again.

JMW

http://www.ViscoAnalyser.com

 

[1gibson]

Ping it with a hammer and "listen" with an instrument of your choice.

Close to full = thud
Empty = ring

I would think a piezo transducer/microphone bonded to the outside of the reservoir would be isolated enough to mostly pick up response of the vessel wall and not ambient noise.

Developing a control system for this may be more than you want to get yourself into, unless this type of product is already commercially available (no idea.)

 

[jonbuchanan]

If overfill protection is your concern, I would use a nuclear point level switch. Completely non-contact (outside the vessel) and it would be unaffected by the dust. The only concern would be expense (costs a lot more than an ultrasonic - probably $4K-$6K), and licensing is an issue. However, most states allow you to purchase the small radioactive source under a General License (free of charge) to avoid having to apply for an expensive Specific License. The only real drawback to the General License is that you have to hire the vendor to startup the system. You'd also need to hire them anytime you need to move or dispose of the system.

Nuclear switches are sometimes considered the last resort because of the issues mentioned above. But they work very well in the toughest applications. Once they are installed, they are maintenance free. If you decide to go that route, I recommend you call Berthold Technologies.

http://www.berthold-us.com/industrial/contact.htm