reading pressure gauges

Thanks for your questions,
I am Susan from Dynisco China, for your questions, I would like to make some comments.

Why do clients use pressure gauge on extruder or plastic machines?
To test pressure(sometimes test temptersature both), then can gurantee the quality of medium, and also protect machine working in good status, and protect operator well.

What function on presure gauge or called indicator or controller ?

Clients can set relay alarm, set high limit or low limit, when value is exceed set scope, then alarm works, and operate can adjust promptly, to avoid any bad influence for machine ,staff or products.

So, that is why we called control system works. To get first-hand data, then collect, then analysis, then help u make a precise decision.

Hope my answer can help you well
If you are searching for these kind of products, welcome to contact me.

Susan From Dynisco China
liming.yang@dynisco.com

I’ll have to agree with Bill here. First you have to ascertain the purpose of this data collection before you determine the periodic collections.

Bill & Brent,
Both of your replies were based on injection molding, though this is the Extrusion forum. That is probably because the topics in the upper left of the main page are not further separated, and most of the visitors to this page in the past have been people who work with injection.
However, much of the comment still applies to extrusion. I'll reply here:

Brent points out the advantage of using recorded data for SPC. This is a great idea, provided you have people who read the records, who know what they mean, and who can do something about what they show.
He also notes the advantages of calibration and alarms.

Bill asks the purpose of data collection. Here are some: to become "fluent" in the process, to guide any changes in process conditions, and to investigate past performance whether good or bad. This "fluency" is getting to know the process in a way that can be communicated to others. "If you know what "good" is, you'll know what "fishy" is.

He also notes that melt pressure readouts are typically imprecise, but better ones are more expensive and seldom seen. I'd like to have more accuracy, too, but what I find on the machines is still useful (to be covered in more detail in my Modern Plastics/Plastics Today webinar, "Get the Numbers!" on January 20). I'm grateful if I find a working, calibrated gauge at all.

I agree with him that data is not 'Holy Writ' but there are good lessons in the Bible even if you don't believe every word. Much data is useful even if it's not all perfect. Like the lessons, it depends on how you apply them.

I worry about the people who don't want data because they may contradict the ideas they already have. To them I cite the following proverb: "If anyone says "trust me," don't." Or, "anyone who resents being double-checked, should be."

Remember the old computer acronym GIGO (garbage in, garbage out) and its even more dangerous version (garbage in, gospel out). Data is a form of vision: we see the best we can with what we've got, but we don't close our eyes just because it isn't 20-20.

Bill is more openly critical of new engineers than I might be, but I secretly agree with him, as I got my (chem) engineering training when that meant pilot plant labs at school and summer jobs in real plants (mine was in an oil refinery).

In my first job (Union Carbide) I had to specify a reaction temperature for a pilot plant run of PS: Production just asked me: 135, 145 or 155 C? and I had to choose fast, as the run was scheduled for the next day. And that's what I mean by the "good and fishy" proverb above. Getting (and understanding) the numbers will better prepare you to say 135, 145 or 155 when the time comes.

Awareness of profitability: I couldn't agree more. Many people still obsess about energy costs in extrusion, even though material costs are far more important, and they worry about production rate without knowing if they can sell the increased production. My engineering school (Cornell) gave us a semester of "Chemical Engineering Economics," which led to the business-and-reality-based approach we need to have today. We used casino games to study risk and reward (no field trips).

As for alarms, I'm less skeptical and more worried than Bill. We don't need to prove limits by ourselves, as equipment makers and experienced production engineers should advise, and we should protect ourselves accordingly. No experience? Find someone who has it. Too much is at stake.

Yes, studying variables is full of surprises, but I'd rather prepare for a party than be surprised by it.
ALG

When collecting data you need to answer a few questions:

1. What's the purpose of the data collection in the first place?

2. Your data collection equipment must have a degree of precision and accuracy 10X better than the answer you're looking for. Temperature and pressure gauges are notorious for being "10%" tolerant unless you buy specifically matched and calibrated equipment. Just because the gauge reads 1100 PSIG could mean the absolute pressure can be + or - 10% of that figure.

If you're looking for general trends, a commercial gauge is fine. If you're looking for absolute data it's gonna be a little tougher.

3. Don't believe your data as 'Holy Writ'. When comparing a part measurement to a machine parameter. You are correlating a DIRECT measurement (measured on a calibrated machine traceable to the National Bureau of Standards) to an INDIRECT measurement from commercial grade gauge (read: not particularly calibrated).

4, Data is Data. That's all it is. Don't expect it to translate directly to all processes and all machines.

This is always fun to watch the Newbie Engineers ("I have a DEGREE in PLASTICS") make pronouncements from their experiments and watch their absolute predictions fail when we try to implement them. I let them off the hook by showing that in General they are right but it usually doesn't work out well if they get too knit-picky just because their computers can give them an answer in five decimal places.

Before embarking on a Post Doctoral Thesis type evaluation balance the effort you expend and the equipment you use to the profitability of the outcome you expect. Spending a few day's pay to find out why you lost $10 is a bad investment unless you can leverage your work to save thousands in the future.

Don't set alarms on upper and lower limits until you've proven to yourself this is "the Limit". In many cases (for example) picking a pressure limit is only valid within a very narrow band of temperature and speed.

Remember: studying variables in Injection Molding is like your Birthday - Most of the time you get the one present you hoped for. But all of the time you get a few surprises you didn't expect.

That's why this business is so much fun!
-- Bill

You can do both. You are referring to the actual gauge (manometer) not the digital readouts that most presses have? The readouts make tabulating pressure data easy over time to see how the system or process is behaving.Here the readings are over time and you need enough time or readings to give statistical significance. I mean like to do SPC

The manometer is great for troubleshooting the hydraulic system and the readout on the screen is great for watching the process and seeing how your machine and settings interact. You can see screw pressure and relate to recovery and heat/BP settings and also it is good to track pressure at transfer to judge your process. In the case of pressure at transfer, just a gauge would make it hard to capture the moment at transfer, this is where the screen readouts are nice.

Also, it is a good idea to compare readings from a known good manometer (gauge)with your digital readouts to see if the press needs calibrating.
Many new presses don't have analog (or digital for that matter)gauges. You can buy one and put a female hydraulic quick connect on it and put male ports on your presses on the mentioned test ports and move the manometer from press to press.

The some pressures on the screen like transfer and peak injection pressure can be alarmed and even programmed to shut down the press.

Keep on Molding! and don't hit your pressure limits,
brent

www.matrixtooling.com

Thanks for the questions, Allan.

I look forward to comments from the field. My bet is that the readings should be made over time.....but if so then how much time is necessary to get a good indication?

BR,
Matt