Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Creating a Homebrew Industrial Process Monitor?

Cliff posted more than 7 years ago | from the doing-it-yourself dept.

Hardware Hacking 97

pionzypherm asks: "I work at a glass plant for a major beer company. My job entails monitoring the furnaces that melt the glass. I have been working on a project on the side, collecting data from various sources and compiling it into an easily used form for the higher ups. I've finished two of our three furnaces, but one remains. This furnace uses technology from the early nineties. There is no networking, the hardware is completely closed and unavailable for any screen scraping. Two of the items I'm looking to monitor (and would appear to be the easiest starting point) are two valves for a gas and oxygen line which will provide data on a portion of our energy usage. I was thinking of a microcontroller board or something similar tied in to monitor the positions of the valves. I'm unsure where to begin though. What books, microcontroller boards or alternatives would you recommend for someone new to this? What suggestions would you have for such a project, and what pitfalls might I run into?"

cancel ×


Sorry! There are no comments related to the filter you selected.

What I need (-1, Troll)

Anonymous Coward | more than 7 years ago | (#18992629)

Is an industrial strength one that resists jis.
HA! youve been cyber boned.!

Open resources (3, Informative)

zappepcs (820751) | more than 7 years ago | (#18992729)

You might try to ask this question on some of the robotics yahoo groups. They are filled with people that do this kind of thing for a hobby and spend a great deal of time thinking about such things not to mention that they do their work with small home brew or cheap microprocessor systems.

People that make their own CNC machines know a LOT about monitoring position of things etc. This might be your best bet for initial and longer term answers and help about how to accomplish what you wish to do.

One piece of advice though is think through what you want to ask. When you ask, explain the system in some detail, your thoughts on what might be monitored, how, and what your end goal is with your monitoring. They may have suggestions that go beyond your knowledge scope if you explain more about the system so they can think about the problem with all the requisite information.

Re:Open resources (3, Insightful)

LWATCDR (28044) | more than 7 years ago | (#18993399)

Most home-brew CNC machines are open loop so that tech wouldn't apply to this problem.

Re:Open resources (1)

Neil Jansen (955182) | more than 7 years ago | (#19001715)

No, most of them are closed loop. A small percentage of them use stepper motors but most use servos with PID controllers (which is exactly what is inside a process control). Either way, I wouldn't recommend bothering the guys on Yahoo Groups.

For a working example of a PID controller, try the OpenServo project: [] There are a few revisions to the board, each with a different Atmel chip powering them. The best part is, it's written in WinAVR GCC C/C++.

Re:Open resources (1)

40ohms (528261) | more than 7 years ago | (#19006363)

What your describing is not difficult to do, however does mean some PLC related programming. Planning will need to be done. Since it may be critical to the operation of the place I would recommend making plans to simulate the real world field wiring and make things work before actually making the commitment to put that equipment on line. From past experience with many custom control projects I can attest that modifications usually need to be made several times before and after such a project goes on line. Beside much of the off the shelf PLC hardware there is the MAT PLC [] project on Source Forge. One of the more interesting small PLCs I have run across (and been very happy with) is made by Tri Logi [] . If you have not dealt with PLCs or control systems I would highly recommend finding someone that has some experience with them to help. They are not generally difficult, however out of the 15 - 20 various flavors I have worked with in the past all have had some differences that make writing the programs a bit of an art. Different manufacturers have different ways that the logic is processed. Sometimes those minor differences can make what initially looks like a simple task a challenge to debug.

Easy (3, Interesting)

Anonymous Coward | more than 7 years ago | (#18992739)

If all you need is monitor the position of a valve, you can homebrew the position sensor yourself using a plastic disc, a magic marker, and an LED. The sensor sends pulses as the valve turns, these get picked up by a small microcontroller board. Get an experiment board that supports RS-232 serial. Then, you write some minimal code on the microcontroller to process the events into some reasonable data format, and send them over serial to a PC. Do your real logging and data processing there, instead of on the micro board.

Re:Easy (1)

Steve-o-192.168 (1096403) | more than 7 years ago | (#18993145)

Great idea. Another way would be to use a magnetic pickup off of a bicycle speed sensor. Either way, be sure to save your state to flash or something so when power is lost, you dont end up not knowing where your valve is. Steve

Re:Easy (1)

Phreakiture (547094) | more than 7 years ago | (#18993375)

If all you need is monitor the position of a valve, you can homebrew the position sensor yourself using a plastic disc, a magic marker, and an LED. The sensor sends pulses as the valve turns, these get picked up by a small microcontroller board.

You could also dismantle a mechanical mouse for this task, or use the sensor from an optical mouse to read the movement of a disc like the one you described in your solution. Another option is to mechanically tie a pot to the valve movement, and read the wiper position with an ADC, which would have the advantage of not needing to keep state (i.e. it could dead-reckon the valve position).

Re:Easy (0)

Anonymous Coward | more than 7 years ago | (#18995433)

(just nitpicking, but dead-reckoning would describe the former solution)

Re:Easy (2, Interesting)

Doctor Memory (6336) | more than 7 years ago | (#18993977)

Interesting idea, but it won't monitor the actual position of the valve, just changes in its position. So you won't know if it's closed or not, only if it's more open or more closed than it was the last time it was at rest. Actually, with just a single LED, I'm not sure you could even tell which direction it was moving.

It's hard to know what to suggest without knowing what form the valve takes. Does it have a round handle, like an outdoor faucet, or a lever, like many natural gas connections have? How many degrees does it move through from fully closed to fully open?

Instead of trying to measure the valve position (which isn't a value you really want), you might want to think about trying to tap into the lines the valves control and read the actual flow. If the valve has a threaded connection, it might actually be easier to attach a flowmeter than to instrument the valve.

Re:Easy (1)

Watson Ladd (955755) | more than 7 years ago | (#18997531)

That's what a rotary encoder [] is for.

Re:Easy (1)

Doctor Memory (6336) | more than 7 years ago | (#18998389)

Right, that's what I understood the parent to be suggesting. However, the description only included one LED, which means you can't tell which way the encoder is turning, just that it's moving. If you had access to a decent print shop, you might be able to make an encoder wheel that "faded" from one state to another: e.g., instead of just an opaque block on the track, you'd have a gradient. Then, depending on whether you had an abrupt (clear to opaque) or ramped (clear to translucent to opaque) transition, you could determine the direction of rotation. Easier said than done, I suppose, or maybe this is a well-known technique to people in the encoder field. Or maybe I should have kept my mouth shut and just applied for the patent — d'oh!

Regardless, the parent post just recommended punching some evenly-spaced holes in a disk, which obviously can't give you a direction indication.

Re:Easy (1)

sumdumass (711423) | more than 7 years ago | (#18998439)

I'm not sure if he is attempting to show when the furnace is on or how much fuel each unit is using and when. I'm not sure a valve position is anything close ot what he want. When turning a valve open or shut, the amount of whatever flows isn't always consistent with the amount the valve it open or shut. In other words, 3/4 open might be letting 5/8 of the maximum flow though.

Instead, I suggest going with a professional product like a mass flow meter for the gas supply connected before the control valve. Actually, I think that one controls the valve too but there are [] ones like this [] that just measure the volume of gas passing though it. There are ultrasonic meters that don't require fitting pipes and such.

I think if he is looking to measure how much gases were used, when(what time) and the temperature of the process, I would look at a commercial sensor and then homebrew the backplane together to get the information from it. I think I would work on getting accurate measurement and then worry about collecting them into a usable format.

Re:Easy (1)

Doctor Memory (6336) | more than 7 years ago | (#18998725)

I'm not sure if he is attempting to show when the furnace is on or how much fuel each unit is using and when.
I got the impression all he was looking for was some way to measure the natural gas and oxygen flows to the furnace, to get an idea (or maybe a better one) of furnace efficiency and to provide some operation cost figures to the bean-counters. He seemed pretty aware that process control was something he didn't want to get in to.

There are ultrasonic meters that don't require fitting pipes and such.
That would probably be ideal, as a later response indicated that these furnaces were intended to run continuously until they were EOL'd. That would make fitting an in-line sensor problematic at best. I wasn't aware that any of these existed for gases, though; every one I've seen has been for liquid flow measurements. I'm not in the industrial control and measurement field, though, so I'm sure they have lots of toys I don't know about...

Re:Easy (1)

sumdumass (711423) | more than 7 years ago | (#18999813)

Those links I offered were just the first hit form a google search for "gas volume meter". There are tons of them. The company I bought a air quality analyzer from has shut down but they sold quite a bit of these meters too.

I'm pretty sure they work for gases as well a liquids. I think the air going in would be forced air and it probably be more efficient to check the air pressure and calculate the size of the ducting asuming it is a forced air intake.

I'm not in that field either. I used to be in the environmental services industry and we needed to have some of these things on site/in our response bag to detect leaks and such(I work on computers now). I didn't get too much into that end of the operations. I was more into the Contamination containment and material response part of it. I was more involved in protecting and containing the spill and identifying the hazards (not in that order of course) and then coordinating the response. Usually I ran a pump truck and did whatever was necessary. Most sites already had a response team working before I would get there. But we had to know how to use everything in case someone else had to leave the scene, we could fill the gap.

You would be amazed at some of the stuff they have available. Even for the recreational person the technology has soured. Look at the images at the bottom Of this page. It is a side scan sonar I have on my wish list. Of course it will sit there for a while.

Re:Easy(er?) (1)

go$$amer (218906) | more than 7 years ago | (#18997313)

Is there a visible moving part that you could just take digital images of at intervals and use some visible, machine discernable marking point to calibrate a monitoring program?

Re:Easy (1)

smellsofbikes (890263) | more than 7 years ago | (#18997351)

As other people have said, there are many ways to measure rotation. Optical encoders with absolute position are cheap and easy to interface. If you have to make it yourself, fire up any CAD program, even the cheapest, and draw a big circle -- 30 cm in diameter -- and subdivide it with 360 or 720 or whatever radial lines, then plot it on a laser printer using overhead transparency film. You can get superb resolution. Then read it with an opto pair from an old mouse -- or two opto pairs in quadrature, for rotation direction. But for $20 or so you can buy an optical encoder that does the same thing: check MSC or Grainger or even Digikey, I believe. There are newer semiconductors that do the same thing: you glue a bar magnet onto the end of the shaft and stick the semiconductor onto a pad nearby and it determines the rotation to 1/1000 of a rotation, as I recall. (I haven't used them, just seen them in EDN and the like.) But for quick&dirty, the overhead-transparency route is really easy. I built one that could show a half-degree variance between two shafts with a cogged belt connecting them (because one pulley wasn't concentric.) It took me 30 minutes to plot, cut out, and glue onto the shafts.

Look into GE Fanuc or Allen Bradly (3, Informative)

p!ssa (660270) | more than 7 years ago | (#18992743)

There are tons of options out there (I havent worked in the field for ~ 10 years). Assuming you can access something to get the readings off you could get a 90-30 PLC to pull the data points. The Cimplicity MMI is a great software package for monitoring, alerting, reporting etc. Try calling GE Fanuc and just tell them what you are trying to do and the can give you plenty of options.

Re:Look into GE Fanuc or Allen Bradly (1)

Rogue974 (657982) | more than 7 years ago | (#18993195)

p!ssa and some others that posted here have it right. Look into something like Allen Bradley os GE or something. If it is important enough to put a monitor on it, spend a little extra money and do it right. I am a Controls Engineer and have spec'd out stuff like this all of the time. There are cheaper solution, but failure rates of homebrewed methods have too high a chance of failure. Like p!ssa said, call a vendor that sells the stuff, or if you don't have the know how to spec it, call a system integrator (i.e. process controls engineers for hire, that is what I do for a living) and have them assit you in specing it out. There are way to many potential problems to homebrew controls equipment and the results of failure can be disaterous, espically if it is monitoring used to make controls decisions.

Re:Look into GE Fanuc or Allen Bradly (2, Insightful)

AB3A (192265) | more than 7 years ago | (#18993649)

I'll second what Rogue974 said. I'm a registered control systems engineer. Before you go "monitoring" a furnace you need to consider several things:

1) Where is your data going? Who might use it and how.

2) What instrumentation are you going to use and how will it interfere with the process?

3) What are the safety and reliability issues?

4) Are there any legal ramifications?

These systems may be independent for a very good reason. I can't tell you how many data geeks have salivated over the SCADA and plant control systems I design and manage. All these folks mean well. However, they see this information as a wonderful fountain of pure lovely unadulterated data. Yeah, right, and the Brooklyn Bridge is for sale too. This data is subject to all sorts of problems including calibration errors, maintainance, gaps in the real time data stream, and so forth. That's why you have operators.

There are serious and very complex safety issues with using this data. It may not be up to standards for use as a database of record. It may even confuse things. It may also be subject to hacking. Note to all: Yes, it's true, process control systems are highly subject to hacking because patches are not applied in as timely a fashion. We have to make sure the patches work and are ultra stable before we apply them to a working industrial control system. That's why we try to stay off of the IT intranets. It's not that we don't want to share the data. It's that we don't want to spread incorrect data.

The operators are there to act as a primary source of data for you. Let them do their jobs. If you think that you can simplify their jobs, work with them. Don't foist some cheesy embedded system on them to do the job of a hardened industrial controller. You won't keep their trust that way.

In the long run, it's usually worth the expense to purchase a hardened controller to do this job. Could you do it yourself? Sure. Would it work well enough for anyone to use on the very first try? Not likely.

If you choose to forge ahead anyway, keep your eyes and ears wide open for experienced hands to give you suggestions. Part of the plant mentality is that we'll say our peice just once. If the outsider is any good, they'll listen to the wise suggestions. If not, their stuff will just break and we'll throw it out at the first opportunity. Our trash heap of such well meant projects is embarrassingly large. Be prepared to listen and learn.

Re:Look into GE Fanuc or Allen Bradly (1)

billdar (595311) | more than 7 years ago | (#18993875)

OT, but what defines a "registered" controls system engineer? Is it a US thing, or some state issued title? I only ask because I've never heard it applied before here in California.

Re:Look into GE Fanuc or Allen Bradly (1)

Graff (532189) | more than 7 years ago | (#18994199)

Take a look at the ISA CSE License [] . There might be other certifications and licenses but the CSE License is a major one for sure.

(ISA is the Instrument Society of America, CSE is Control Systems Engineers)

Re:Look into GE Fanuc or Allen Bradly (1)

Doctor Memory (6336) | more than 7 years ago | (#18994941)

That's funny, because I think California is the only state the does register control systems engineers. At least, everyone whose resume I've seen it mentioned on always states "Registered control systems engineer in the state of California".

Re:Look into GE Fanuc or Allen Bradly (1)

alienw (585907) | more than 7 years ago | (#18996263)

Each state has its own engineering licensing system. But basically, if you offer engineering services directly to the public (as in, working for an engineering firm), you need a license. This mainly applies to civil/architectural/industrial engineering.

Registered Control Systems Engineer (1)

AB3A (192265) | more than 7 years ago | (#19001349)

At last count I think there were something like 38 states that offer Control Systems Engineering as a practice of engineering one can register for. This is basically the ISA's Control Systems Engineer certification.

For those who are clueless about the whys and wherefores of registration of professional engineers: All states in the US offer tests for which you can become a registered professional engineer. It's basically a way to put your name on the line. You stamp the drawings and documents with a seal that says you take responsibility for the design. Note that this stamp works both ways. If you consult and the customer changes your stamped design before or during construction, you can not be held liable. However, if they adhere to your design and it fails to perform as expected, you can be held liable.

It also entitles you to act as expert witness in a court of law. Many municipalities insist that a senior operations or design office have a registered professional engineer in charge.

Control Systems Engineering is an extremely broad field of study. It includes practical applications of thermodynamics, fluids, valve types and applications, instrumentation, real time network design, Laplace transforms of various process behaviors, Understanding P&I Diagrams, and so on. There is a lot to study and a lot to know. I came at my practice via an electrical engineering degree. However, there are nearly as many chemical, mechanical, and civil engineers among our ranks.

Most schools don't teach controls engineering. The ISA is hoping to slowly change that. However, there aren't many who choose to get in to engineering in the first place. Engineering doesn't present a clear promotion path to the executive halls as one might see with the sales, accounting, or legal professions. So most of our students are in it for the love of the subject, not because they seek to become top dawg of the company.

On the positive side, Professional Engineers often have opportunities to join standards committees. And it is through the standards processes that one can really make a difference in the world. I wish there were more people interested in this sort of thing, but sadly, very few students want to think that far ahead...

Re:Registered Control Systems Engineer (1)

billdar (595311) | more than 7 years ago | (#19021751)

Good information, thanks.

So, if I am understanding correctly, it is a focus or discipline of the PE accreditation. (IE, the whole testing/mentoring process).

Re:Registered Control Systems Engineer (1)

AB3A (192265) | more than 7 years ago | (#19027331)

Yes. Controls Engineering is recognized by NCEES [] .

Re:Registered Control Systems Engineer (1)

billdar (595311) | more than 7 years ago | (#19037745)

Thanks again. Been working in the field for several years now and never heard of this.

Might be looking at this wrong... (3, Insightful)

fineghal (989689) | more than 7 years ago | (#18992795)

Are they not willing to fund it/hire someone to do it?
You might be making this too complicated.
Let's say you misjudge the tolerances and your fancy little project gets turned into cinders/melts inside the furnace?

Why can't you monitor the volume of gas flow and then calculate the energy? I assume these gases are stored in a tank or something like that. It should be comparatively easy to attatch some type of flow sensor upstream of the furnace.

Re:Might be looking at this wrong... (2, Insightful)

pionzypher (886253) | more than 7 years ago | (#18994779)

This monitor won't be used to control the furnace in any way. There are control systems in place that work as advertised. The issue is that the system is completely closed. Numerous other (and more experienced) entities have attempted to figure out a way to gather the data without success via software. The system itself allows trends with ten minute increments up to 24 hours previous(on screen, no way to get a hard copy besides writing each data point down) This data needs to be collected for the purpose of monitoring our energy usage. With the increase in gas and electricity prices, the bean counters want this data (in addition to its value to us as a metric on furnace efficiency, burner adjustments and the like).

Invasive modifications to the gas or oxygen lines are definitely out of the question. We're talking approximately 40k cubic feet of gasflow per hour. Though we do already have two existing pressure readings, your post makes me think I might be able to replace those with industrial grade electronic meters to do some sort of venturi style measurement.

Re:Might be looking at this wrong... (1)

Hektor_Troy (262592) | more than 7 years ago | (#19003749)

Hrmm ... sounds like it's able to output values to a monitor?

How about putting a capture unit in between the controller and the monitor?

That'd allow you to do all kinds of manipulations to the data you get, without altering the controller, and without affecting the direct output to the monitors.

You'd be working on graphics, but you already know where all the interesting numbers are, so you "simply" do OCR on the interesting parts of the picture. Then dump the numbers into an appropriate database and do your datamining from there.

Would be doable for the price of a capture unit, small computer and a few days (maybe a week) programming I'd think.

Too easy (1, Informative)

Anonymous Coward | more than 7 years ago | (#18992811)

These folk: []

are in the business of making small microcontroller projects easy, quick and fun. Something like a member of their Basic-stamp family would be pleasantly overkill for your needs. They can convert your temperature readings, valve closures, infra-red readings and such to a time-stamped serial data stream that your computers can collect. There's a large number of good books on how to make the Basic Stamps do all sorts of cool stuff. So if you spend the money there, you'll be building a working system from cookbook instructions.

With externally actuated valves my favorite sensor is the hall-effect sensor (available at the site above). It's corrosion resistant, doesn't wear and is not angle sensitive. You glue the cute little magnet where it will fit, and the sensor where it will be close to the magnet in the on-position and Viola!

When you want a low cost solution, and to make thousands of them it's hard to beat the Atmel ATtiny series of microcontrollers for low power/cost and high reliability.

Re:Too easy (1)

ReKleSS (749007) | more than 7 years ago | (#18996959)

I haven't looked at Parallax for a while, but the Atmel chips seem like a better alternative. If you want the convenience of a Stamp, take a look at Arduino [] . Open source, based on an Atmel, and fairly cheap.

Don't want to be done on the cheap (3, Informative)

billdar (595311) | more than 7 years ago | (#18992953)

If this is an industrial application, you really don't want to homebrew it.

Spending the money up front for a reliable, standard solution will save a ton later when your homebrew breaks or some other poor bastard has to support it. There's been too many times I've opened a a panel where my first words are "WTF?".

Especially if you're working with oxygen. Get yourself a nice little flow meter (micromotion makes a good one). Then you can get both volume, and (presumably) valve position. If the valve is electrically actuated, you can use the information for a host of alarms.

Either way, if the information is valuable enough to record, its worth the money up front.

Re:Don't want to be done on the cheap (2, Insightful)

boristdog (133725) | more than 7 years ago | (#18994321)

"If this is an industrial application, you really don't want to homebrew it....if the information is valuable enough to record, its worth the money up front"

HAHAHAHA! Oh, if you only knew how most factories operate. I work with several leading edge semiconductor fabs and you'd be amazed at the amount of homebrew/seat-of-the-pants solutions abound.

I am currently working on a system to track production on about $300 MILLION dollars worth of equipment. My equipment budget? I was lucky to get $30,000 and most of that was for data storage hardware.

Face it, the suits don't think that factories make data. They think factories make widgets.

Re:Don't want to be done on the cheap (1)

billdar (595311) | more than 7 years ago | (#19022175)

I have worked in a ton of factories, and know exactly what you're saying. When a +/- of $0.02 on the shelf is enough for someone to buy the competitors product, I can see why this pressure exists.

However, I'm sure you've seen factories that are best un-documented and at worst death traps because someone didn't put the $$ or effort up front when doing something.

Using your example, $30k sucks ass for process data collection. You can do it cheap as hell, say an old dell box with fix32 sending OBDC data to a *nix box running MySQL and 1TB of cheapy IDE drives. Add in some custom VB code to glue it all together, and you've got a system that I've seen running one of your favorite breakfast food brands.

You think the suits upstairs will like to hear it will cost another $30k + downtime to reverse engineer your own program 5 years later when a changes need to be made and you can't remember whats there? You can't tell me that doesn't happen.

Re:Don't want to be done on the cheap (2, Interesting)

tlhIngan (30335) | more than 7 years ago | (#18995313)

If this is an industrial application, you really don't want to homebrew it.

Well, I thought that way too, until I re-read his requirements. It really depends. If it's automation to control stuff, I'd go with off-the-shelf professional type things. One, there's a lot of CYA - if anything goes wrong, you can blame the hardware, rather than it being your fault (even if it really isn't). Secondly, it's control - things go wrong, and unless there are tons of failsafes and alarms redundantly connected, well... (and hopefully a profession would know how to avoid having a single point of failure for all the redundancy).

But in this case, it's a simple monitoring thing. The guy wants some values to report to higher ups. The machine has its own controller (can't be screen-scraped), and all that's really happening is just a way to say "valve X is open Y%", without worrying about alarms and crap (presumably the machine has alarms and everything built in). As long as the monitoring won't be doing anything safety critical (it shouldn't interfere with the existing controller), a homebrew solution is perfectly fine.

Heck, a different way is to simply find a way to capture the display output - a camera works. Do it properly and it should be possible to analyze the image and generate the data that way, too.

Re:Don't want to be done on the cheap (1)

pla (258480) | more than 7 years ago | (#18996043)

If this is an industrial application, you really don't want to homebrew it.

Depends on why he wants to monitor this machine.

Working in a similar environment myself, I've found that "management wants to know how often we go out-of-spec" means a whole world of difference from "one mistake will halt production ".

"Real" hardware to do these tasks, if even available, costs a bundle. The homebrew solution, if just a matter of someone having accidentally uttered a meaningful phrase at a meeting, usually comes in thousands or tens of thousands cheaper.

Not to mention, things often start as "a critical measurement" that a year and a half after successfully implementing a homebrew counter, not a single person has ever even looked at the counts (Hmm, does that sound too much like a real example?).

Use nature (2, Funny)

Timesprout (579035) | more than 7 years ago | (#18992977)

Put a toad in the furnace and then ignite it. If the toad does not jump out then clearly the furnace is not heating quickly enough. For day to day management use a dragon. They are very good with high temperatures and will be able to help out with your energy bills by giving your furnace the odd blast. One safety tip with the dragon though is that if your name is George its probably best not to let the dragon know that.

Re:Use nature (1)

glwtta (532858) | more than 7 years ago | (#18993337)

For day to day management use a dragon.

If a whole dragon seems overkill, remember that hypercontiguating two minus-dragons produces 0.6 dragon.

Re:Use nature (1)

SnarfQuest (469614) | more than 7 years ago | (#18994409)

Make sure you don't make any sudden moves, or your dragon is likely to explode.

This is Diskworld, inn't it?

where to begin (1)

timias1 (1063832) | more than 7 years ago | (#18992989)

Are you trying to monitor energy usage or provide more advance controls to the furnace? For control: first I would find out the specs on those valves (mfg, inputs, outputs) find out if the mfg has a control board and what interface it uses. Otherwise a PLC is one way to go, and usually can you can get i/o that will work in most applications. Start at one piece at a time. An industrial PC could work too. For Energy usage monitoring I would start by sub-metering the furnace's gas and electrical feeds. They have whole product line around this. Call some vendors and get a demo

Not asking Slashdot... (2, Informative)

Hijacked Public (999535) | more than 7 years ago | (#18993009)

I would suggest asking this same question at [] rather than here.

If you just want the position of two discrete valves I would suggest finding a used PLC on Ebay. Single box types (like an Allen Bradley SLC 150) that work with discrete IO only can be had for a little bit of nothing. Your biggest concern with costs would be the programming software so I would stick with brands offer it free of charge.

Without knowing what kind of budgetary firgure you are working with to implement this it is hard to get much more specific.

Industrial automation is a business (3, Informative)

Kohath (38547) | more than 7 years ago | (#18993033)

There are people that do industrial automation. There are large companies that make every type of sensor you can imagine to monitor anything you want. There are industrial controllers to control automatic assembly lines. This stuff is all off-the-shelf. It's not "homebrew".

If you're asking this on Slashdot, you're looking in the wrong place.

Do it like a professional would do it. It's a furnace. Stuff can go wrong. Monitoring it with a half-assed homebrew approach is probably worse than simply observing it carefully and worrying about it all the time.

Re:Industrial automation is a business (3, Insightful)

RGRistroph (86936) | more than 7 years ago | (#18993153)

If all humans were like you, we'd still be hunting with spears, because "hunting is for hunters, doing it with your half-assed bent stick and string that throws a small spear is dangerous."

Look dumbass, how do you think the "professionals" do it ? They just "homebrew" it and slap some fancy decals on it.

Re:Industrial automation is a business (0)

Anonymous Coward | more than 7 years ago | (#18993489)

You have absolutely no idea, do you ?

Couple that with the attitude, and that post really isn't doing you any favours.

Amen, brother! (4, Interesting)

pestie (141370) | more than 7 years ago | (#18995081)

Every time someone asks Slashdot a question like this, the hysteria crowd comes out of the woodwork to scream about how it's absolutely impossible for an "amateur" to do it, and you absolutely must hire a "professional," lest something tragic happen, ranging from the ever-popular "you'll lose your job!" to a bucket of dead puppies or something.

Yes, I realize that professionals are sometimes necessary, especially in situations where life is clearly at stake (pilots, medical, law, etc.) I'm sure some jackass will show up to tell me how this is an industrial furnace and that clearly means that a professional is warranted, but we have no idea what the particulars of this situation are. Just stick to the freakin' question, people.

It used to be the case that "professional" implied not only a degree of competence, but also a certain amount of integrity and experience. But that's just not true any more. All it means now is that someone gets a paycheck for doing something. Often it means that they're experts in nothing more than doing something as cheaply as possible.

For what it's worth, I'm personally fond of the Atmel AVR [] microcontrollers. Many, many people are also fond of Microchip [] 's offerings in the PIC line [] . But for rapid development, something like the Parallax BASIC Stamp [] is probably the way to go. They're cheap and easy (like a good woman) and let you focus on the task at hand rather than the bit-level details of how to read sensors, etc.

Re:Amen, brother! (4, Insightful)

moosesocks (264553) | more than 7 years ago | (#18996693)

I feel like you're missing the point here.

What the original poster is trying to do is not innovative in the least. In fact, it's pretty well-worn territory (which he may have been unaware of by no fault of his own).

If he had said "I've looked at all of the commerical options, and nothing fits quite right", I'd be in agreement with you that he should go out and try to create his own solution. However, this is not the case, and a lot of time and experience has gone into developing products that fit his needs very well.

And even at that, there will likely be a good deal of 'hacking' involved in getting these valves to do what he wants them to, given that they're industrial components. Any EE on the planet knows that it's preferable to use a commercially-available IC instead of constructing an equivalent circuit out of components as long as the IC fits the job. The same goes for industrial components.

Re:Amen, brother! (1)

Mike1024 (184871) | more than 7 years ago | (#19004311)

Every time someone asks Slashdot a question like this, the hysteria crowd comes out of the woodwork to scream about how it's absolutely impossible for an "amateur" to do it, and you absolutely must hire a "professional," lest something tragic happen

I can't speak for other posters, but when I see stories like this I can't help but think "this could be done by an experienced amateur, but if you're posting on slashdot asking how to go about it, you probably aren't an experienced amateur".

To put it another way, he's better off learning to program microcontrollers by making himself a thingy to adjust his computer's fan speeds (and/or a similar basic project to learn with) and leaving the industrial process control alone until he has some more experience.

Just my $0.02.

Re:Amen, brother! (0)

Anonymous Coward | more than 7 years ago | (#19009111)

Professional is just a fancy word for "insured."

Re:Industrial automation is a business (1, Funny)

Anonymous Coward | more than 7 years ago | (#18996499)

If all humans were like you, we'd still be hunting with spears, because "hunting is for hunters, doing it with your half-assed bent stick and string that throws a small spear is dangerous."
Dear Slashdot:

OOg want to catch buffalo. OOg never catch buffalo before. OOg only seen buffalo once, from far away. But OOg has good idea to catch buffalo with coconut. Can anyone tell OOg good books to read about coconuts?

Also, OOg live in North America. OOg only have acorns, and no coconuts. Can anyone tell OOg how many acorns it takes to equal one coconut? OOg got finger bit off by dingo, so OOg only able to count to nine. Will nine be enough acorns for OOg to catch buffalo?

Please tell OOg soon. OOg real excited about catching buffalo. OOg going to ride buffalo like pony.

Re:Industrial automation is a business (0)

Anonymous Coward | more than 7 years ago | (#19003777)

UNG the caveman advise OOg to consult with cave-guru OOG_THE_CAVEMAN [] . OOG_THE_CAVEMAN once chased by buffalo for mile ! That make him big expert, him only charge 400 clams per hour but Gardner Report says it is worth it.

Re:Industrial automation is a business (1)

Kohath (38547) | more than 7 years ago | (#19002761) do you think the "professionals" do it...

They spend the money to buy the pieces of equipment that are designed to do the job. They hook them up the way they are designed to be hooked up. They put fail-safes into the system. It's (somewhat) expensive and time-consuming. It's not a hack.

It can't really be considered "homebrew" any more than buying a Cisco Router is a "homebrew" routing solution.

Asking programmers and PC hardware hackers how to do industrial automation is like asking a plumber how to write a web app. Except unreliable industrial automation systems can kill people, where unreliable web apps are simply annoying.

A "homebrew" furnace monitor just seems like a bad idea.

Re:Industrial automation is a business (0)

Anonymous Coward | more than 7 years ago | (#19009075)

If all humans were like you, we'd still be hunting with spears, because "hunting is for hunters, doing it with your half-assed bent stick and string that throws a small spear is dangerous."

This makes me so tired. If all people were like you people wouldn't have been hunting with spears in the first place, because they would all be busy trying to reinvent them, and doing a terribly job at it.

He's not out to "go where no man has gone" with some innovative solution, he's out to solve a problem that has been adequately solved a million times over. He just want to do the same thing the professionals do, except he'll take more time, do a suckier job, and probably even waste more money. The best thing that can happen is that he reads up enough on it to actually realize that the solutions that exist are perfectly fine, and implements something similar. But in that case he could have just hired professionals right away.

Look, the only reason you take this idiotic romantizised attitude towards this is because you know nothing of the target domain. If this was a guy who wanted help writing a text editor in assembly (which he had never programmed in before), not because he thought it was cool, but because he had no idea there are more higher level languages out there or even complete stable text editors already, you'd just tell him to google it instead of wasting his time by producing something inferior anyhow. Furthermore, if it could potentially cost the company a lot of money should this text editor be riddled with bugs, you'd start to see the unreasonableness of it.

Look dumbass

What did he do to deserve your grade school insults, disagree? How old are you? People like you are an embarrassment to this whole community -- not only talking out of your ass, but being rude about it too.

how do you think the "professionals" do it ? They just "homebrew" it and slap some fancy decals on it.

If you think there's no difference between "homebrew" and what professionals do, that word doesn't have any meaning at all. If that's what you think, fine, but you don't live in the same universe the rest of us do. And you probably use another dictionary.

Re:Industrial automation is a business (1)

pipingguy (566974) | more than 7 years ago | (#18998235)

He's not modifying the (potentially dangerous) process, he just wants to monitor data in a way that doesn't involve physical risk.

Suggested way to do this (3, Informative)

Steve-o-192.168 (1096403) | more than 7 years ago | (#18993045)

You should probably install an electronic flow control valve with a flow sensor. Use a microcontroller, PLC, or some such thing to monitor your sensor & control your valve. Monitoring it visually (via camera hooked to an embedded computer) is doable, but way harder than what I just described above. You also run into problems with high temperatures getting to your camera. Take a look at some of the solutions Freescale has to offer. You can order a development board, so you can breadboard something together. Should cost around $200. jsp?nodeId=02430Z [] For a SR Design project for Electrical & Computer engineering, we were tasked to do exactly this with a freescale microcontroller. We needed to precisely monitor the amount of certain gasses put into an oven we use to bake chips. Hardest part about all of this is getting to know the specific PLC or microcontroller you're using. PLC's are easier (generally) to program than most microcontrollers, but not quite as versatile in number/type of interfaces. You need to have a very good understanding of Assembly, C, & how sensors work. You also need to be able to read & understand the mind numbing manuals & technical documents describing the sensors & microcontrollers you choose to use. -Steve

Re:Suggested way to do this (1)

pipingguy (566974) | more than 7 years ago | (#18998333)

You should probably install an electronic flow control valve with a flow sensor. Use a microcontroller, PLC, or some such thing to monitor your sensor & control your valve.

Installing any kind of inline valve/sensor into a process system almost always requires a shutdown. Shutdowns tend to be very expensive. That's why he wants to do "non-invasive" monitoring and data collection. It may be possible to install valve positioners while the facility is still operating, but be careful.

There is a way to do what you suggest; look up "stopple bypass" or see .pdf []

This doesn't work with all installations (for a whole bunch of reasons) and you DEFINITELY don't do it without professional help.

Fuji PXG (0)

Anonymous Coward | more than 7 years ago | (#18993059)

I own a small glass studio and we use a number of different types of process controllers to run the furnaces and annealers. May I suggest you look at the Fuji PXG series of controllers; they are relatively cheap and support MODBUS/RS-485 which allows easy serial interface to PC. Fuji also provides a free basic application (PXR-Lite) to program and monitor the controllers. You can send me email jon[at]eastfallsglass[dot]com for more information.

In Industry Avoid Homebrew If Possible (2, Informative)

Graff (532189) | more than 7 years ago | (#18993139)

If you are doing serious industrial work do not attempt to roll your own unless you absolutely have to. The money you save on professional instrumentation will be wasted in downtime and glitches.

There are plenty of professional solutions out there, from gas flow monitors to automated valve systems to integrated industrial process monitoring and control systems. If you are looking to control fuel and oxygen supplies then you need to get stuff that is blastproof so that a stray spark can't set anything off.

Start off with a major supplier like Grainger Industrial Supply [] . There are tons of components there that might suit your situation. Particularly look at their process monitoring [] section.

use gas flow monitors instead. (1, Interesting)

Anonymous Coward | more than 7 years ago | (#18993175)

Why don't you use gas flow monitors instead. Most of the modern ones offer rs232 and some even have 10baseT interfaces. These devices are temperature compensated and can be set to match the gas being measured. This would be much more accurate than monitoring the valve position.

Judging from your question, you sound like a programmer thrust into a hardware problem or a plant engineer thrust into an IT problem. In either case, check with your contacts, suppliers or company engineering group to get suggestions for implementing and monitoring the monitor systems.

Perps the best place to start is OMEGA.

A cheap, easy, powerful board (0)

Anonymous Coward | more than 7 years ago | (#18993199)

Check out the Arduino board. []
It is completely open source and talks to a computer via usb. It is designed as a multimedia interface. There are lots of tutorials that tell how to connect almost anything to it. There are connections to all kinds of web enabled and multimedia programs.

If you want ethernet connectivity, check out the Make controller. / []
It isn't quite as easy to deal with as the Arduino but it does have ethernet and has all the other advantages of the Arduino.

PicBasic Pro (1)

bergeron76 (176351) | more than 7 years ago | (#18993333)

Check out Pic Micro controllers. You can even use something like PicBasic Pro ( to program the PIC's these days. You basically give the chip a regulated power supply (or buy a prototyping board) and use code to work with the IO.

They even make blank prototyping boards that are already silkscreened for what components to use.

You'll need a pic programmer (I use this one [] , some blank PIC chips (a few dollars each at, and some software to program them in.

Another approach is to use Parallax BasicStamps. They cost a bit more for each chip (like $40 I think), but you can get a kit with everything you need for a hundred bucks or so. They can be programmed in Basic, which makes them great for getting started in electronics.

Reading without integrating sensors (2, Interesting)

RGRistroph (86936) | more than 7 years ago | (#18993343)

One technique you might think about, coming from a "first do no harm" strategy to avoid being blamed if something goes wrong with the equipment, is to try to not add new sensors or valves and etc to the equipment, but try to take advantage of the increases in computing power to simply read it and recognize the situation exactly as a human would.

Check out this project: [] It describes software that uses a webcam pointed at an anologue needle gauge to recognize the position of the needle. Why not, as much as possible, set up passive sensors that don't touch or interact with the equipment in any way, feed them into a cheap multi-gigahertz computer, and process everything that way ? If the furance has a big accident, it would be hard to blame your apparatus.

Re:Reading without integrating sensors (1)

SixFactor (1052912) | more than 7 years ago | (#18993761)

This is brilliant. The philosophy you describe for "first do no harm" by not adding extraneous and intrusive components is actually in use at nuclear plants today. We use closed-circuit TV to watch dials in high dose areas, reducing operator dose, and freeing the operator up to do other things. We don't quite go as far as using an OCR-like process described by the link, since we'd have to qualify that software (which is a buttload of paperwork, honestly, and not worth it), and human eyes are more than good enough for data gathering at the periodicty of interest.

"Homebrewing" for your application is OK - BUT! (4, Insightful)

SixFactor (1052912) | more than 7 years ago | (#18993367)

...let's be clear about this: all you're doing is monitoring and gathering data - there is to be no feedback signal from the homebrew rig to control the valves. There's a whole field devoted to control theory, one that is best not trifled with, especially with industrial processes that can potentially cause fatalities.

If you really want meaningful data from those process streams, you're much better off installing calibratable (calibrable?) flowmeters on those lines that cover the performance range of the process fluids you're working with. If you've got the flow, you don't need the valve position, unless it's for a secondary indication to validate the valve's performance (e.g., position vs. Cv vs. measured flowrate). The flowmeters can be hooked up to provide data for remote collection, or more simply, display data for periodic local reading. Here's a mess [] to start with. Whomever you buy from, you'll need to develop specifications defining the operating range, operating conditions (pressure, temperature, humidity), power requirements, tolerances, calibration frequency, etc.

Translation (0)

Anonymous Coward | more than 7 years ago | (#18993373)

I'm a high school student with a part time weekend job at the local bottling plant. I'm supposed to walk around with a clipboard and check the furnaces every half hour, but I'd rather just sit in my chair all night and read bad science fiction books. Does anyone know how I can use linux to solve my problem?

Re:Translation (1)

Linagee (16463) | more than 7 years ago | (#18995131)

Just make sure you don't engineer yourself out of a job. :)

Need sleep (2, Funny)

glwtta (532858) | more than 7 years ago | (#18993381)

I read that as "Creating a Hebrew Industrial Process Monitor?"

Really wouldn't know where to begin with that one.

Re:Need sleep (1)

Steve-o-192.168 (1096403) | more than 7 years ago | (#18993425)

Sounds kinda neo-nazi to me.....

Homebrew industrial tool control, eh? (2, Informative)

stonecypher (118140) | more than 7 years ago | (#18993531)

Yeah, because if there are two words that go together, it's industrial and homebrew. :D

There are a whole lot of things that go into handling an industrial system. If you're really going to try to do this on your own, you've got a lot of reading ahead of you; the cost of faults in an industrial system is typically prohibitively high. You're going to need a deep familiarity with modern methods. You're going to need to be familiar with direct hardware control, realtime coding (which is harder than most people think,) constant test polling, and all sorts of stuff most programmers never, ever have to deal with.

This is not something you can take lightly. If you're going to do this, you have to get it right, the first time, and that means your test cases and regression tests have to be diamond-hard, your specification has to be absolute, and you have to know your timing will not fail. These are difficult issues, but with the appropriate know-how, this can be done.

Here are some places to start:

Embedded Control:




Specifications and Modelling:

It may be instructive to get Software Cost Estimation with Cocomo II [] and hack together a clearer view of the cost of doing this yourself. You'll find that those obscenely expensive control systems suddenly seem much less obscenely expensive.

Re:Homebrew industrial tool control, eh? (0)

Anonymous Coward | more than 7 years ago | (#18995363)

Disclaimer: I'm not a lawyer, an industrial technician, master brewer, or EE guy. I'm nobody.

There's another side to going homebrew that doesn't quite go with an industrial application:

  • Insurance
  • Liability
  • Exposure to Lawsuit

If you're doing this on the side, then you're exposing yourself to a boatload of legal risk. It's one thing if you're doing it in a lab on company time, as you would only loose your job if something went wrong. Even if you were a contractor, you could set yourself up as a corporation prior to accepting the contract and you'd have some ruidmentary legal shielding. Here, you're prone to being sued into oblivion by offering to attach something you made in your apartment, to a piece of (possibly vital) industrial equipment - it doesn't matter if you're in charge of the thing while you're on the job or not.

Another way to look at it: how much R&D went into that old, closed system that's in the way now? The people who sold it to your employer - did they have some kind of statement absolving them of certain liabilities?

I think it's commendable that you're looking to go above and beyond to improve things at work - every company needs more people like you out there. But when you're talking about large companies, huge amounts of materials and gobs of cash, you've gotta look out for #1. Do the research on how to handle this in a safe, standard and widely-adopted way, and find something as close to pre-fab as possible (of course, that's why you're posting here). Then, get your employer to agree to let you put things together *on the clock*.

Selling your employers on the new equipment should be as easy as figuring out how much more data precision this is going to offer those bean counters, and how that difference equates to savings and a realized increase in profit . If those savings work out to more than the cost of time and materials, over an acceptable time-span (X months, or Y years - whatever they want), it'll be easy. If not, then demonstrate the amount of risk demonstrated by the existing, inadequate, obsolte, impossible to support equipment should it break - "a stitch in time" and all that.

Good luck!

A few suggestions from a CNC guy... (4, Insightful)

spazmonkey (920425) | more than 7 years ago | (#18993565)

I build CNC and automation equipment, so I can pretty safely say that from what you describe this is a brain-squishingly trivial project. Probably one that can be done over lunch - After you spend five years climbing the learning curve, which is not at all trivial.

    I would just ask someone who does do this for a living out for lunch, it'll take them ten minutes. I do this when I need coding done. The price of a few beers to get the occasional patch or script written is a lot more efficient than many years learning coding to do it myself the one time a year the need comes up.

    The learning curve on automation hardware is at least as steep as learning Linux, and with crappier documentation. Coding guys usually seem to underestimate the complexity of the physical engineering and design side, and think they are always bright enough to just pick it up and do our jobs. There is more to hardware engineering than the butt-crack guy with a monkeywrench, just like there is more to coding than script kiddies.

  In short - unless you want to go into this as a hobby or career change, just treat a hungry engineer to lunch and call it good. Even if you paid him it'd be less than the books you'd need.

How did this get passed moderation? (0)

Anonymous Coward | more than 7 years ago | (#18993643)

The question is incredibly stupid. Rather like:

"I've been put in charge of network security, and I've noticed that intrusion detection is a problem, so I've decided to build a secure operating system to prevent this. Would you recommend a PC or a MAC architecture."

There is a WHOLE industry that provides solutions to this problem. How do you think facilities are automated, some random choice of micro-controller? You should BUY a solution, it will do more an cost less and do less harm.

The hacker spirit revived (1)

athloi (1075845) | more than 7 years ago | (#18993653)

I'm sure this post will start with bad karma, and be modded down to -273.15, but here's my take on it:

It's very cool you're doing this. As you stated it, this is a prototype so you're not about to rely on it, just play with it and test it. That's hacking at its finest: getting equipment to do unexpected but useful things, efficiently.

The people who say "contact a robotics firm/authority" are probably technically correct, but as Robert Heinlein was fond of saying, specialization is for insects. You might want to hit those resources for influences but keep going on the project.

My advice would be to decide whether you want a net of smart controllers, or a central controller processor connected to dumb sensors. Other than that, I know zero about this technology but think your project sounds like a lot of fun.

Never accept doubt as a substitute for enthusiasm.

Another way to monitor the valves (1)

Kerrygeek (950934) | more than 7 years ago | (#18993763)

If there's a way to read a voltage level, you're halfway there. If not, you'll have to build or buy some kind of sensor. There's an open source software package called Argus [] that I use to monitor network devices but I've heard of people using it to monitor various manufacturing functions. It's overkill if you're just wanting to monitor 2 or 3 valves but if you have a lot of them it's a perfect way to do it. It has a web interface and can e-mail you when something changes state. It's pretty easy to install but you'll need somebody who knows a bit about Apache and CGI to help you set up the web interface.

I just tried the newest version yesterday and it started sending me alerts via my e-mail while I was still configuring it.


Pitfall #1 (0)

Anonymous Coward | more than 7 years ago | (#18994045)

Tell us why your Engineering department isn't involved. Right now it sounds like politics is going to be your biggest pitfall.

If you don't have an Engineering department, then Maintenance. Somewhere you must have people with titles like Millwright.

Even if all of that is contracted out (either third party, or a separate entity within the "major beer company" family) then you need to find those contractors and have a chat.

If you're trying to joe-together a proof of concept so that funding & authorization will follow, tell them that. Be clear you want it to become a proper project they will install. If you sound like to want to just fudge something together that they're going to have to repair later, they won't give you the time of day.

Also I'd look to the manufacturers of the existing equipment. There's good chance they've developed 'bolt-on' upgrades to do what you want.

Also I'd think about bouncing it off the younger Millwrights first. They're likely less bitter and more recently educated. You could have better luck getting them interested in talking about your ideas over their coffee break.

Anyway, let's empahsize that the Millwright/Maintence crews tend to be grumpy because they have to fix dumbass ideas all the time. Use that. Tell them straight up you've got an idea and you want their input so it doesn't become a hassle. That show of respect will go a long way.

Simple (1)

ratboy666 (104074) | more than 7 years ago | (#18994139)

Obtain a 68HC11 microcontroller (they are available for around $30, quantity 1). You will need an RS232 adaptor (figure another 10 bucks).

Attach a small disk to the valve. Put aluminum foil on one side, and earth the foil. Cut a series of holes around the disk (before attaching the foil, and the disc should be a non-conductive material). Put a sensing whisker so that the whisker is in contact with the foil, or disc. Now when the valve turns, the microcontroller will get pulses.

Of course, this doesn't tell you which way the valve is turning. Use a second circle of holes, offset, and another whisker, not offset (or the other way around). The phase between whisker activation will tell you which way the disk is turning.





Where H is a hole, and W is a whisker pickup. If the lower pickup activates first (from a state of no pickups active), the disc is being rotated counter-clockwise. If the upper pickup activates first, the disc is being rotated clockwise.

The amount of rotation is determined by counting the activations on a single whisker. The speed of rotation is determined by activations/time.

So, the pickup shouldn't cost more than an evening to build, and two bucks in parts.

You can dispose of the microcontroller and directly (I would use a 7404 for buffering and some isolation) drive a PC parallel port.

Re:Simple (0)

Anonymous Coward | more than 7 years ago | (#18995545)

heh.. what a great solution! is this 1981 though?

Re:Simple (0)

Anonymous Coward | more than 7 years ago | (#18996415)

Interesting.. but I think he should use a comparator as an input stage to logic rather than a 7404 which only has TTL inputs and might behave oddly with a non-TTL signal.

I'd personally use an optical pickup of some sort because this would not be subject to wear and tear.

Direct measurement. (1)

Sqweegee (968985) | more than 7 years ago | (#18994453)

I looks like you're trying to measure the flow of natural gas and oxygen by measuring valve position. While it is possible to do, you would need to be able to calibrate the position to flow rate, this would also require very consistent supply pressures and back pressure. On top of that the flow is not likely to be linear with valve position. I've worked with valve positionners that can do this but you'll need to do some redesigning of your control system to integrate them, for that you'll need an expert.

If you have any budget at all to work with I would suggest direct measurement of flow using appropriate instrumentation such something from: productlevel1.htm []
They cover all sorts of volumes of flow, can track total volumes consumes, and can be interfaced with other systems either through analogue or digital interfaces.

I would also suggest checking out various industrial control/instrumentation/automation forums out there and maybe hunting down someone locally to consult with.

Don't... (1)

simp (25997) | more than 7 years ago | (#18994563)

Monitoring gas and oxygen supplies and all the hardware that controls it can be dangerous. What if your homebrew micro-controller with its connection wires makes an earth-loop and noise gets on the signal control lines?

What is your fail-safe strategy? Do the valves fail open or closed if something short circuits in your setup? Does your home-grade micro controller survive the heat & vibration of the environment?

Buy something of the shelve from a "standard" industrial automation firm. You can get anything from a $49,- PID controller to a $5 million DCS just my googling. Used industrial control parts even turn up on Ebay these days.

Disclaimer: I actually seen a fully functional glass furnace in action. And I have also seen a fully functioning glass furnace during a control computer blackout. For the people that don't know glass furnaces: once you fire them up most designs need to run continuously until they reach their end-of-life. Once they powerdown and the glass solidifies the only way to get it running is to break it down and build a new one. Imagine that: A 12 year lifespan and you need a 100% uptime for the control computers (redundancy is your friend)..

I have thought of a solution for you... (1)

Kymermosst (33885) | more than 7 years ago | (#18994717)

I have a solution for you that will work. I thought I might give it to you for free (as in beer) but decided I would rather exchange it for beer (as in free). But, I only drink microbrews (being an Oregon beer snob)... so I guess you're out of luck.

Re:I have thought of a solution for you... (1)

Doctor Memory (6336) | more than 7 years ago | (#18995671)

*snort* There are about 10,000 different microbrews made in Oregon, and 9,982 of them are IPAs. Good luck if you're looking for a decent porter or pils...

Re:I have thought of a solution for you... (1)

Kymermosst (33885) | more than 7 years ago | (#18999691)

Naturally there are breweries that only sell their beer on-tap that have good porters, but you have to come to Oregon to find them!

For those that are bottled, I would have to say that you need to give the following a try: Monkey Face Porter from Cascade Lakes Brewing and Mocha Porter from Rogue. Second place are Black Butte Porter from Deschutes Brewery (which has suffered in quality as it has gotten bigger, IMHO) and offerings from McMenamins.

I don't know why you'd want to drink a pils from any brewery (or any other lager for that matter).

Get pro devices (1)

ericferris (1087061) | more than 7 years ago | (#18994725)

Are you trying to extract flow information from non-instrumented valves? If so, you need to either replace the valve with something modern that outputs its data cleanly, or just insert flowmeters in series with the values to be monitored. Either way, get something clean and professional, don't do a DIY job that's hard to maintain. Also, once you have an instrumented device, I'd not recommend putting a homebrew micro-controller system on a factory flow, especially a glass factory, which abounds in abrasive dust. Any consumer cooling fan, for example, would quickly see its bearings ground and clogged by silicate dust. Get a real industrial device with sealed bearing, etc. Devices that can widthstand factory floor abuse, high temperature and high dust levels are expensive, but there is a good reason.

Standard industrial solutions (1)

Animats (122034) | more than 7 years ago | (#18995185)

There's standard hardware for this sort of thing. You can get little industrial I/O units with Ethernet interfaces. [] They're little boxes with an Ethernet connector on one side, and digital, analog, or thermocouple inputs on the other side. They're built to hold up in a factory floor environment, and easily replaced if damaged. You can even get wireless ones. []

If you're going to send signals around a plant, 100baseT is actually quite noise-resistant. More so than TTL signals or even RS-232, because it's a twisted-pair differential signal.

To talk to these things, LabView is often a good option. You can create panels of dials, gauges, and graphs with little effort, log and analyze data, and give the plant operators the tools to see what's going on.

Re:Standard industrial solutions (1)

pionzypher (886253) | more than 7 years ago | (#19042079)

Those two are shaping up to be great solutions. Thank you.

Allen Bradley (1)

Wonko the Sane (25252) | more than 7 years ago | (#18995561)

You can not go wrong with Allen-Bradley PLC systems. Find a large Allen-Bradley vendor (Rexel-Nelson is a good one) and use them as a resource.

Have them send a representative out to you and help tell you what product line will fit you best. That's what vendors are for.

If there is a possibility of becoming one of your suppliers, any good vendor will bend over backwards when you need help like this.

Silicon Labs 8051 family. (0)

Anonymous Coward | more than 7 years ago | (#18995725)

check out silicon labs' (formerly Cygnal's) line of 8051 micros (up to 100mips or so I think). They can be ordered from digikey, cost $5-$10 and provide dozens of ADC channels, DACs, comparators, PWM, timers, onboard flash, ram, temp sensor, voltage references, internal clock, etc. etc.
Their development kits run under $100 and include target prototyping board, debugger/programmer, compilers, etc. I like the c8051f31x family, as they come in 32-LQFP packages, which are not overly fine pitched, so easy enough to work into homemade PCBs. there are also DIP packaged versions of a couple families. Handy for cranking stuff out quick. Think they even have nifty USB dongles ready-to-program.

I've used these in diy projects ranging from wine cellar temp controllers, weather station data loggers, cinema timecode generators / slates, robot control, arcade machine controls, not to mention all the boards these chips get designed into at work.

Just the info you need (1)

itwerx (165526) | more than 7 years ago | (#18995729)

If you want to reply offline using the email address above we can refer you to an industrial monitoring systems specialist who does work for, among other companies, Corning Glass.

Gas Codes/Hazardous Areas (0)

Anonymous Coward | more than 7 years ago | (#18996211)

It's a furnace and there are rules that apply (eg AGA) which are in place to stop people doing things in an unsafe manner. There will be issues with explosive atmospheres (hazardous areas) and special instruments/equipment required.

I do this sort of stuff for a living and it is bad enough when an experienced engineer does not interperate the gas code correctly, you don't even seem to be aware of some of these issues.

Pay a professional who do this for a living and you won't have to be explaining why your employer needs to buy a new furnace, if not deal with death and injury.

National Instruments (1)

alienw (585907) | more than 7 years ago | (#18996305)

Have you looked at all the data acquisition stuff National Instruments sells? They ought to have something that will work, although it will pretty much have to be PC-based. Definitely beats the hell out of trying to homebrew this, though. I'd say that if you have to ask, it'll be too much work -- microcontroller projects have a way of taking way more time than they should.

Re:National Instruments (0)

Anonymous Coward | more than 7 years ago | (#18997679)

Actually, National Instruments does make hardware for the very purpose of industrial control and monitoring. There are PC based versions, but some are ruggedized and can run in real-time autonomously and headless (with ethernet connectivity). They work well, can make direct sensor measurements, and are super easy to program. Check out: []

Just buy the parts and do it yourself (1)

Robotics integrator (1097935) | more than 7 years ago | (#18999217)

Why use home brew you could do it with a PLC of almost any brand(most likely AB or Omron) over an Ethernet network that way you could automate the whole plant and remote monitor from home. If you go with AB you can use the DF1 protocol to read the information directly off the PLC it's not difficult at all. I just graduated from college and my senior project was similar to this with some other fancy stuff such as QC and a nice HMI and SCADA system. I say for 25k you could automate your plant and control product flow. By the sounds of it that might put you out of a job. well best of luck and make sure you don't use anything home brew because when something goes wrong with the hardware you made you will be the one in court trying to explain why the valve read open but wasn't and the pressure vessel exploded. PS. you could do it for cheaper but then you sacrifice quality and that means more repair time and down time.

Homebrew? (1)

ClamIAm (926466) | more than 7 years ago | (#18999843)

So you're trying to use a "homebrew" solution at a major beer company? Isn't that like dividing by zero or something?

(note to morons: this is a joke)

Amateur attempts (0)

Anonymous Coward | more than 7 years ago | (#19000541)

It would help to know if you know whether the valves are calibrated in some way, as well as whether you can see the position of the valve in some way (not all valves you really can easily, like a gate valve, as opposed to a ball valve). Is valve position to gas flow known here?

If this is like a ball valve, you could find position with a camera aimed at the valve. As the valve opens or closes, the lever moves, which you can capture. As long as you have a standard behind it (like a ruler, circular ruler/dial), you could just simply use a camera and record it to a file. Then run an analysis program that takes stills from the vid file and notes position along with the frame (frame correlates to time) and get your info from there. All this is a high tech dial capture they use to use in regular furnaces that had a pen plot onto paper (the pen would move outward over time and the paper was gridded accordingly)--you'd see these on metallurgical furnaces.

If more like a gate valve, you could do the above by simply putting a mark on the handle (paint) and monitor that position.

Better would be to use a rotary encoder. There can go ridiculously high, but even nowadays, 2048 or 4096 positions per turn are typical (you can go far higher depending on your needs). Many, many hacks exist for this in the CNC world, but look up "rotary encoder" on google. US Digital makes one that you could probably figure out pretty quick, and you would have to hook it up to a counter (see boondog automation), for which there are chips to interpret the quadrature single and then you can choose your capture method (ISA card, conversion to serial, regular DAQ, etc.).

Note that with rotary encoding, there are sort of 2 types. One is a counter--it counts ticks of light/dark light as the encoder moves, and since there are 2 disks, it can tell position by the spectrum of light. The other is absolute position--this will tell you the position of valve in a circle. Both have their places and purposes--fast turning counts, I'd go with the counter type. Slow turn valve or especially valves like a ball valve that can't go over a revolution, I'd go absolute position.

btw, this is from an amateur.
Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>