What is the “regulator self-feedback verification method”?

What you’re seeing are oracle bones, right: lol :lol:lol　

Haven’t heard of it; asking just to broaden my knowledge

I’ve never heard of it; I even searched on Baidu and couldn’t find any information. Could you please share all the relevant details you’ve seen, so we can figure out what kind of application it relates to?

This post was last edited by zxb1990 on 2015-7-4 21:43. Here’s the situation: our company still uses old regulators in remote areas, and sometimes we also help small factories run by local farmers with basic control tasks. There is an old instruments storage area in the factory, filled with old instruments that were removed during DCS upgrades; when needed, one goes there to select a suitable one, calibrate it, and replace it. So I always have to compete with those cousins who are older than me. This time, I saw half a copy of an old magazine under a shelf that contained this verification method; it claimed to allow for a simple and quick assessment of whether a regulator was good or not, but the content of that magazine is no longer available. So I’m asking here. @feng*aosa on the 2nd floor seems to know about this matter. Could you explain it to me? I’ll choose the best one.

　　This is not a verification method, but rather a means for adjusting during the assembly of pneumatic gauges; however, it can be used to roughly determine whether a single regulator functions properly. Taking the pneumatic regulator as an example, for electric regulators, refer to the following: 1. Connect the regulator’s output directly to the input (measurement); set P=100, I=∞, D=off, with feedback and internal setting ； 　2. When the internal set value is changed from 0 to 100%, the measurement pointer should always stay in sync with the set value, with the deviation not exceeding the specified accuracy requirement. This indicates that the regulator balance is good ； 　3. Setting: P=50, the given value is 50%, the rest is the same as in 1. Placing the given values at 25 and 75 respectively, the readings should be 0 and 100 ； Setting: P=100, given value 50%, the rest is the same as 1. Placing the given values at 0 and 100 respectively, the measurements should be 25 and 75 ； This indicates that the proportional band is available ； 　4. Open the points; the rest is the same as in 2. The measurement slowly approaches the given value until they coincide after the given adjustment, and the approach speed varies with the integration time. This indicates that the integral is applicable. 　5. Enable differentiation; the rest is the same as in 1. With the given adjustments, the measurement focuses on the phenomenon of overshooting during tracking before returning to alignment; the amount of overshoot varies with the differential time. This indicates that differentiation is available. 　

Great, I’m thinking about it – what else is involved in regulator balancing?

What is mentioned on the 8th floor refers to the closed-loop tracking experiment of the regulator.

　　Yes. The difference isn’t very big. I didn’t answer at first because I couldn’t remember the name of this experiment. That “verification method” was developed in the *era, and it has characteristics of that era.

　　It’s hard to define exactly. In other words, when the regulator generates a deviation signal, the performance of the setpoint section and the measurement section (range, linearity) must be consistent in order to enable accurate comparison across the entire range. In a pneumatic regulator, the consistency in performance between the given part and the measuring part is referred to as balance. This is particularly significant in pneumatic regulators, as the actuating element (bellows) must be adjusted and matched for each gauge.