GE FUNAC LM 90-30 PID Problem

[turbohamad]

Hello Friends;

I am working on a GE Funac LM 90-30 software. I face the problem to set the CV value when PID in auto mode. The parameters are set as under
1.Enable,Override,Manual,UP,Dn are OFF
2.Sample Period 0.01 sec.
3. Dead Band + and - are 00003
4. Error Term=sp-pv
5.Derivative Action = ERROR (Error/PV)
6. Biase=00000
7.Min Slew Time=00000
8. Upper Clamp=32000
9. Lower Clamp=00000
10. Output Polarity= POS

11. KP=11.18%/%
12. KD=0.00 sec
13. KI= 0.055 rep/sec of 32000

CV value is 22000 even when sp and PV are almost same.
I want that CV will become between 10000 to 14000

Please help me how i fix it?

The other point is My PID loop starts from %R01099 and i use the
BIT SET WORD AND CLR as follows

%R01111=IN of BIT SET WORD use when engine speed cross the 60%
LEN 000001
Const=00005
Same parameters are used for BIT CLR WORD when any shutdown HAPPEN TO THE MACHINE

According to my knowledge it is Anti Wind RESET of the config word of PID.

Please help me

Thanks

[Steve Bailey]

CV value is 22000 even when sp and PV are almost same.
I want that CV will become between 10000 to 14000

Please help me how i fix it?

The action of the PID function is to modulate the CV in such a way as to minimize the difference between the SP and the PV. From your description it sounds like the PID is doing its job.

[NetNathan]

Your "Sample Period" may be a little short...
This is the "slice of time" of information the PID block uses to calculate reaction.
SO....right know you have it re-calculating every 0.01 sec.
Does your process (do not know what you are controlling) react that fast?

Also....you need some "Bias". This is were the CV "thinks" it should be. Usually... this is somewhere near 50% (16000), but I usually end up with it around 8000-10000, because my furnaces have a lot of power.

[NetNathan]

Also...
CV is changed by PID to bring PV to SP.

[turbohamad]

Respected Friends;

Thanks for the reply. In PID block there is also a parameter name
MANUAL COMMAND it is 22000 always. I enter the new value like 14000 but it goes back to 20000.

Is it use only when PID is in Manual mode or is it CV value?

how can i adjust the cv value when sp and pv are almost same is there any way?

Regards

[Steve Bailey]

You could use bias. That term is independent of the error (difference between SP and PV).

What are you trying to accomplish? The SP value is presumably what you want to achieve. The PV value is a measurement of it. The CV is the command to the process that is necessary to produce the desired result.

If you put in a bias value of -10000 to get the CV to a value of 12000, you can expect the PV to decrease, which will cause the PID function to increase the CV

[NetNathan]

I assume you are using the PID_ISA block and not PID_IND
Maybe this will help...
It is for RX3i, not LM90, but it is very close...

[turbohamad]

Respected Friends;
Thanks for the guide lines. The programmer use the Anti-Reset windup action as follows but i doubt on it.

PID loop starts from %R01099 and programmer use the
BIT SET WORD AND CLR WORD as follows

%R01111 (Anti-Reset windup action)in config Word=IN of BIT SET WORD use when engine speed cross the 60%
LEN 000001
Const=00005
Same parameters are used for BIT CLR WORD when any shutdown HAPPEN TO THE MACHINE

Is it ok? i think in config word the BIT 4 is for Anti-Reset windup action.

Please guide me?

Regards

[NetNathan]

More info that might help....maybe

[Steve Bailey]

It would be helpful if you would describe the system controlled by this PID function and what you are trying to achieve or fix. By the nature of your questions I assume that you are not the person who originally programmed the system.
Has it ever worked properly?
If so, are you trying to correct something that has recently started causing problems and what has changed since the last time it worked properly?
If it has never worked properly, why is the person originally responsible no longer involved?

Or, are you simply trying to understand the intent of the original programmer?

With regard to your question about the anti-reset windup parameter, I believe it only comes into play when the CV reaches the upper or lower clamp value. Its intent is to prevent accumulating a large integral component.

[NetNathan]

^ What Steve said....would make it way easier to understand what you are doing and why.

[turbohamad]

Hello Friends;

The programmer left the company and now i face the problem that the engine overshoot and not stay on idle speed. The PID output control the 4-20 mA electro hydraulic actuator which control the amount of fuel to the engine.

The idle speed is 62% of 11520 and at that speed engine overshoot and trip at over speed.

Please guide me that is anti reset windup bit address for SET and CLR bit is correctly used?

As the programmer pass the parameter as constant 00005 as i describe in my previous post?

Thanks

[sjohnson]

You may try Uploading the Program or Just do a copy of the rungs and Paste them in the text box here. It may Help clear up what you are trying to describe as there will be a visual and may explain a lot more in detail and get you a solution much quicker.You can also Take a snapshot of the pid parameters and rungs associated.


GFK-0467 Chapter 12



On Your KP You have alot Compared to Your KI. Maybe reason for the overshoot. But not knowing what you are controlling and how it reacts.. I use this theory when setting up my Pid loops. KP how fast you want your Control getting to the SP. KI determines How fast you want it to react so it doesn't over shoot and KD would be you How it reacts to holding it to the SP but too much KD can have erratic effects also. I usually leave this at 0 till i get the control really close sometimes it isn't needed.


Your control Loop is 1099R and you state %1111 is Reference +12
Config Word. It modified the default pid settings.



So back to Steve's question did this work before and it never work properly and you now have the task of setting this up. There is probably easier ways to accomplish what you are trying to accomplish.



First I would limit the output CV Upper Clamp to a value that doesn't allow the over speed of the machine.


I see what you are Trying to accomplish with bit 4 of the config word Address +12 but in my opinion the pid loop itself is way off track for this to function properly as re configuring the pid loop to get closer to a good control should be done first.



Try setting your KI 35% of your KP with little to none KD. Set the upper clamp that it doesn't over shoot until you can dial in the control correctly. The idle speed 62% of 11520 is 7142 are you writing 7142 into the SP for idle or maybe use this value as your lower clamp. Is something else writing to the CV to keep it at 22000 as it should be dropping unless something else is writing to it or you may have a manual control enabled or make sure a register isn't writing from somewhere else. Pid loops can be a bit overwhelming at first but start with the bare minimum and try not to make it more complicated than it needs to be as it will work much better reliable in the end..

[Steve Bailey]

Answering your only question today:
Bits zero through four of the PID function control word (%R01111 in your case) are the only bits of that word used in the PID function. Using the BIT_SET and BIT_CLR function with an "IN" value of 5 controls bit 4 of the word. So yes, the logic you described is setting and clearing the bit that controls the anti-reset windup action. Personally, I don't think the action controlled by that bit is the source of your problem because it only becomes a factor when the CV is clamped at either the minimum or maximum value.

What do SP values of zero and 32000 represent?
What is being measured by the PV signal and what do PV values of zero and 32000 represent?
Am I correct in assuming that a CV value of zero represents zero fuel flow and a CV value of 32000 represents maximum fuel flow?