Jump to content

Recommended Posts

HI,

I'm having a problem working with PID on V570. I have 2 units, all the same configuration. The first one works, and the second one couldn't pass through the AUTO TUNE, Status stuck at 2.

I am attaching the program, I would appreciate if you can help me. 

Again the same program works perfectly on the first unit, but get stuck on PID Status=2 on second unit.

Please advice.

 

REgards,

kvj

 

 

 

V570_PID.jpg

Link to post
Share on other sites

kvj,

Please provide more details, then community will help you better.

1. Post project file.

2. Publish sensor type, output /actuator type, settings, what kind of device you manage - heater/cooler, etc.

3. Other details.

B.R.

 

Link to post
Share on other sites

Hi, these are just my quick comments, but one key thing with autotune and heaters is to choose the right setpoint.  The autotune can get "stuck" or just take a really long time if the wrong setpoint is used.  Autotune drives the process to the setpoint, then applied alternative step changes to the setpoint, about 5-10% above and below the setpoint.  It measures the step response of the process to these changes in setpoint.

For example, if your ambient temp is 23° C and you choose a setpoint of 30°C the process will ramp up nicely to 30° (or 33°).  However for the negative ramp (back to 27°) it will take a very long time, especially if the process vessel is large and/or insulated, as the difference between the process temp and ambient temp is very small.  On the other hand, if the process temp is 90°C and the vessel is relatively small and uninsulated, the autotune will cycle much faster, as the vessel is able to lose heat quickly once the heater is tuned off.

Hope this makes sense, it seemed like a complete mystery to me until I worked through it.  My desktop test system is a 5W ceramic resistor inside a plastic cup 🙂

Link to post
Share on other sites

One tip for autotune....

Please check carefully is it heater for this zone  conected to the correct PID output (via hardware and via software). 

I found one installation where part of heater from one zone is connected to another one and PID AUTOTUNE  stuck

Link to post
Share on other sites

Hi,

 I’d like to ask, I’m just trying to do simulation using one of my spare unit and using memory to feed in SP and PV. I can mimic the issue I have and when I try to use the PID SERVER tool to START AUTO TUNE, it forces my program to get pass to STATUS-4. It works fine, but I’m concern when the system powers off and on, does it need to auto tune again? I tried on my unit to power cycle, it still retains the value and its works perfectly now.

 

My main question is, is doing the auto tune using PID SERVER tool will work for good ? What will be of any possible issues later on, if any?

 

Thanks.

 

Regards,

kvj

Link to post
Share on other sites

Hi kvj,
Your question include two different parts.

My main question is, (is doing the auto tune using PID SERVER tool will work for good ?)[1] (What will be of any possible issues later on, if any?)[2]

Part [1] can be answered:
PID Server is running on PC.
It can consume much more resources (CPU, memory), and it may make PID Autotune more precise. But it is dependent of other conditions - PID startup parameters, Sensors/Actuators models and phisical location, inertion of sensors/actuators, etc.
And good side of PID Server is possibility to save PID parameters graph.
You can use this graph for feature reference, send it to someone for rteview, etc.
And yes, you can change PID parameters and make optimization by compare results of an original PID Autotune calculated PID parameters and changed ones.

Part [2] is acadenic, so everybody can add answers in accordance to experience.

B.R.
 

Link to post
Share on other sites

Hi, 

Thanks guys so much for the info.

I tried to RESET and INITIALIZE and then do the auto tune again using the PID SERVER, and NOT WORKING.  What I am doing wrong in here?

Pls advise.

REgards,

kvj

Link to post
Share on other sites

Hi,

I am not onsite, and using my simulator unit.

I'm checking the consistency of the PID server tool if I can use this one next week onsite.  I can't see anything that I have done differently except that I RESET and INITIALIZE and start over. Now the PID server auto tune isn't working too.

 

Link to post
Share on other sites

If it not work with simulator, then control loop does not similar to real one, or PID startup parameters are wrong.

First, you have to look for description of status=-2.

image.png

1. Clean MI 304, as it calculated by PID Autotune.

2. MI 222 should be 1000, and you can use it to indicate percentage.

3. In linearization swap 4096 and 1000.  *4096 is wrong. Should be 4095 (range is 0 to 4095).

image.png

MI 2000 will be in range 0 to 4095 and you should link it to Analog Output.

4. In PID Autotune select 2 cycles and run it again.

Does it work now?

B.R.

 

 

Link to post
Share on other sites

Hi Alex,

Thanks for the input. I am yet to see this tomorrow.

image.png.092ceed01646353dbfa9bee809053316.png 

So far, I have not seen this status during the test. 

I'll try to follow yours and see how it goes. Much appreciated.

 

Link to post
Share on other sites
On 10/10/2020 at 9:21 AM, AlexUT said:

000 will be in range 0 to 4095 and you should link it to Analog Output.

Hi,

Unfortunately the issue is still there.

1. Adjusted all the ranges.

2. Tried Stage 2

3. The first unit that works only do auto tune in 5 to 10 mins.

I would like to try manual PID, how is this done?

 

Thanks

Link to post
Share on other sites

Any help? Basically, what I did is to get the parameters from the working auto tune unit, and encoded to the second unit for manual tuning.  It works for 24 hours and for some reason client made adjustment on the mechanical side and it trips again. And my question is, is this the right way shown below  for V570 forcing it manually for PID ?

 

image.png.1e9f96ef418e157d1130fc71aa2e1fa7.png

Link to post
Share on other sites

I would not continually force a value to the PID status register.  The PID block should write that value all by itself to report what it's doing.

Here's how I do PID-

image.png.92da99ae051349608e69d953a0277ace.png

 

I also always make something on another screen to allow manual tuning-

image.png.f3e8394643dd75959bc81f1d94e48185.png

 

As far as manual tuning goes, I posted this a while back-

Manual loop tuning 101:

Start with Kp = 100 (10.0%) and Ki = 0.  Then start lowering Kp until the process starts to just ring around the setpoint.  If it overshoots wildly you need a bigger Kp.  

Measure the time between peaks.  This is the natural cycle of your process.

Then take your Kp value and double it.

The units on Ki are seconds.  Enter the time between peaks value.

 

If you want a better understanding of the best way to manually tune a PID loop Google "PID Loop tuning methods".  My method is loosely based on Ziegler-Nichols.

Joe T.

 

 

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...