boiler level control with honeywell UDC 6300

[7465]

Dear all,
I want to know anyone using honeywell UDC 6300 for drum level control in boilers? I want to configure and I am not very certain about some settings.
pl reply urgently and thanks
7465

 

[danw2]

Do you need help with UDC configuration or boiler control strategy?

 

[7465]

Dear Danw2,

I need the UDC6300 confiuration for boler drum level control.It is a big story.The original configuration done by the commissining engineer was not correct and he did not connected external hardware properly and he had connected both steam and fee water flow in one circuit and messed up.Even the configuration was wrong.I have never configured UDC 6300 before,I tried for some time but it was not responding well.The feed watr valve was opening with increase in feed water flow.so I had to put a RTK 3 element controller for time being but I want to put back 6300. Now I have connected all the inputs properly.That is feed water flow (input 1) in loop1 and the level( input 2) in loop 2 and Steam flow(input 3) as a feed forward summer.The algoritham is on cascade.
I want to give you my configuration record so that u can check what i have done is correct.Hope u have the manual for UDC 6300.

ALGORITHM
loop rate- 19 X 23X
control alg -PID A
pid loops -cascade
cont 2 alg pid A
in alg1 -none
in alg 2-ff sum2
alg2 inA -input 3 (steam flow)
alg2 inB -loop2 output
alg2 inC -none
8SEG CHAR, 8seg cha2,poly nom,totalizr -disable

ADV MATH- NOT USED

OUT ALG - CURRENT

INPUT 1 ( feed water flow)
Xmitter1 -liner
in1 hi-28
in1 lo 0
ratio - 1
bias-0
filter- 1

INPUT 2 (level)
Xmitter1 -liner
in1 hi- +150
in1 lo (-150)
ratio - 1
bias-0
filter- 1

INPUT 3 (Steam flow)
Xmitter1 -liner
in1 hi-28
in1 lo 0
ratio - 1
bias-0
filter- 1

CONTROL ( I hope this is for loop 1,feed water flow)

pv source IN1
pid sets - 1
lsp -1
autobias- disable
sp track -none
pwr mode- A or M sp
sp hi- 28
sp lo -0
action -Reverse
outrate -enable
the rest are factory settings

CONTROL2 ( The level+feed forward algoritm)

pv2 source IN2
pid sets - 1
lsp -1
rsp source- none
autobias- disable
sp track -none
sp hi- 80
sp lo -0
action -direct
outrate -enable
the rest are factory settings

Any other settig is needed? The boiler steam emad is very widly fluctuatig load and so the feed water respose should be good and within 10 mm drum level range.
The boiler is 25 bar with a steam flow varying between 14t to 20t/h.

Pl check and let me know the confign is correct.Is it necessary to use the math functions and 8segs ?

regards
7465

 

[danw2]

You're too fast for me; you're on page 8, I'm on page 1.

Is this a viable control strategy? (the Sq. root function assumes a DP steam flow meter)

 

[7465]

Thanks for your response.
Yes.This is the control strategy.The diag u have given is for the UDC 3500 the present version.But the logics are same except that it is having passwords.I checked that.But not much of help from that site. In UDC 6300 the sq rt has been given in the loop1 and loop2 also internally,whereas in 3500 it is given only in loop2 as shown in your drwng.That is the only difference between the 3500 and 6500.
yes.It is a DP -steam and feed water flow tx.Even the level is measures with a DP tx.I have a doubt here.In the controller configuration,for the flow inputs, do i have to put a linear or Sq.root??.just because it is a DP transmitter,it is already sq rooted? That is what u mean?.
is is any possiblilty of chatting with u in yahoo,so it will be faster?

regards
7465

 

[danw2]

Yes, CONTROL = CONTROL for loop 1

No, math functons and 8 segment characterizer are not needed, they are used for pressure compensated 3 element control.

> character used for navigation through menus, not as 'greater than'.

** asterisks are critical settings **

1) ALGORITHM LOOP Rate
19 X 23 X cannot be used because that setting is limited to 2 HLAI (high level Analog inputs)
see page 82 (paper) Algorithm Data set up Group:

** For 3 high level analog inputs, the setting for Loop Rate = 1 6X 2 3X **

2) The use of the square root function in the UDC depends on whether the transmitter(s) are configured for pressure output (which uses 'square root' on the UDC) or square root within the DP transmitter (which uses 'linear' setting on the UDC).

Most modern DP transmitters can be either pressure output or flow output (where the pressure is square rooted).

If the transmitters are square root output, then you do not use the square root function in the UDC, you
use the "linear" setting in the UDCs's
INPUTx > XMITTRn parameter.

3) PID loop action: (direct or reverse)
Feedwater flow (loop 1) is reverse action: as steam flow FT PV decreases, PID calls for higher output.

Drum level flow (loop 2) is reverse action: as the level decrease the PID calls for higher output
** Your CONTROL2 > ACTION = direct. Should be = reverse **

4) Ratio & Bias settings (under INPUTx)
Your ranges of 0 to 28 for feed water flow and steam flow seem problematic.
The summer's output = the remote setpoint of loop 1, the feedwater loop.
Since the output of a PID loop ranges from 0-100, and that value has the steam flow value added to it, there's a problem feeding numbers of this range to a loop whose PV range is only 0 to 28.

Although the inputs can be scaled by a ratio and bias setting, the output of PID loop can not, it will always be 0-100%.

I suspect the work-around is to scale the feed water and steam inputs up by some factor, so that the range is greater then that of the PID output. But my notes aren't here, I'll have to check on Monday.

(The ratio and bias settings under INPUTx apply a Y = Ax +b correction to the input;
where ratio = A and bias = b)

5) Output rate (under CONTROL and CONTROL2)

Output rate should be disabled, unless you have a compelling reason to define the output slew rate.
The parameters that follow Output rate, PCT/MIN UP and PCT/MIN DOWN affect the output slew

rate.

6) PID B is sometimes used for loop 2 drum level
Algorithm > CONT 2 ALG + PID B or PID A

PID B definition from manual, page 86 paper; 97 pdf:

PID B: Unlike the PID-A equation, the controller gives
only an integral response to a setpoint change, with no
effect on the output due to the gain or rate action, and it
gives full response to PV changes. Otherwise controller
action is as described for the PID-A equation.

7) "SP Track" parameter under CONTROL
Feedwater loop 1 sometimes uses PV track, where local SP tracks PV
when in manual mode.

8) Setpoints:
Drum level Loop 2 uses a local setpoint
Feedwater flow Loop 1 uses a remote setpoint from loop 2's output.
The manual suggests that such is automatically configured on UDC 6300, page 175:
"Loop #1 Remote Setpoint is fixed as loop #2 output."
I'm not sure what CONTROL2 > LSP = 1 means in a cascade context..

Dan

 

[7465]

Dear Dan,
Thanks for your reply.It was wonderful.I am not clear with point 4 of ur note ( ratio and bias settings).Is it possible to throw more light on that?

Your ranges of 0 to 28 for feed water flow and steam flow seem problematic.
( i have defined this as the engineering units low and hi corresponds to 4-20 ma output from the tx)
The summer's output = the remote setpoint of loop 1, the feedwater loop.
( this i strue i can understand)
Since the output of a PID loop ranges from 0-100, and that value has the steam flow value added to it, there's a problem feeding numbers of this range to a loop whose PV range is only 0 to 28.
( i dont get this part )

Although the inputs can be scaled by a ratio and bias setting, the output of PID loop can not, it will always be 0-100%.
( i dont get this part also)

I suspect the work-around is to scale the feed water and steam inputs up by some factor, so that the range is greater then that of the PID output. But my notes aren't here, I'll have to check on Monday.

(The ratio and bias settings under INPUTx apply a Y = Ax +b correction to the input;
where ratio = A and bias = b)
( is it possible to give a readymade soluton for this?)

all others are ok.I got it.
control2 > lsp1 means ( local set point-the set point from the key board )it has a single set point ( in this case it is the level)

Is it any way of chatting with u at your free time? or can you give your number to call u ?

Regards
7465

 

[itsmoked]

It's best that you continue in here if you can, as others are reading this and learning.

But if you 'must' you can both contact me at my info email and I'll connect you.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[7465]

dear itsmoked,

It is fine.I am ready to share.But unfortuatly,i want a faster solution as I am really want to put it back as the system i have put is a makeshift arrangement which i have taken from my plant.I want to put back my sysytem also.So i promise you that i can put back the result in this forum sothat you all will know.
regards
7465

 

[danw2]

Telephone won't do much good. I'm a plodder. Last time I did 3 element feedwater was 1997, 10 years ago. I need time to ponder and think it through, especially for the old UDC 6300.

Note: Although I use the expression 'PID' from habit, both the feedwater flow and the drum level loops are frequently (usually) PI loops. You can turn off the derivative rate factor by putting in 0.00 under TUNING > Rate Min = zero. It has to be done on each loop separately, because each loop has its own tuning parameters.

I seem to recall having to scale the summer output to match the range of the feedwater flow rate, or maybe it was scaling the feedwater flow PV to match the summer output, but I can't find notes from that job. So, try the bias setting (below) and see what you get.

The rules I remember for 3 element feedwater feedforward control are

1) The steam flow rate and the water flow rate must have the same engineering units and same range. You've done this already.

2) The steam flow signal is ratioed and biased to create a bipolar signal that can add to or subtract from the PID output which cascades into the feedwater loop as the remote setpoint.

The higher the ratio above the value 1.00, the more "gain" the feedforward has on the feedwater loop.

Try a bias of -10 (negative 10), which is half of your nominal 20t/hr flow rate for steam flow on INPUT 3, with a ratio of 1.00 (minimal gain) and see what the response is.

Dan

 

[danw2]

FYI,
The UDC 6300 equation for the feedforward summer is
Final output = [PID output] +
[([Input X * Ratio X] + Bias X} *
(100/(Input X Hi - Input X Low)]

(all the above is supposed to be on one line)

where X can be any of the 5 inputs on the controller