PDA

View Full Version : Danfoss ICAD (1200) Valve stalling.







Grizzly
14-11-2015, 03:41 PM
Hi Guys!
A site (Un manned by Engineers) has a large surge drum on Ammonia. With it's liquid level being metered by a ICAD /ICM Valve.
Said valve has started to randomly shut closed and the surge drum gradually empties.
Causing the plant to cycle off on it's suction Pressure control. (No Alarms generated).

Resulting in long shutdowns before anyone on-site notices the chill room temps rising.

As anyone will appreciate once emptied getting a large Ammonia plant away again is, Shall we say
"Is a right Pain in the butt, Particularly if the plant room is on the ground floor and the control room is 1st floor!

Once onsite the valve opening is shown as 100% open but is actually closed.

Because the refrigerant level in the receiver is high and opening the Liquid Line bypass.
Allows the refrigerant to flood into the surge drum.
Well at least until the standing pressures have equalised then you have to create a HP /LP differential to further fill the drum.

So no power cuts, ICAD shows 100% but is actually closed?
Any ideas?
Grizzly

Brian_UK
14-11-2015, 06:15 PM
Don't know the valve myself but reading - >https://www.google.co.uk/url?sa=t&source=web&rct=j&url=http://www.danfoss.com/NR/rdonlyres/A3EA3045-9E71-4052-88E0-C3E2CFE5A542/0/ManualICMeICADUSA.pdf&ved=0CB4QFjACahUKEwjulLLKtJDJAhUCgBoKHYcqCeE&usg=AFQjCNG8X-O6ArWyHwzYMaiXpcqgTA6Wcw&sig2=A8YitA0JC9W8MCSC8yl09A

Would ask if the valve has been recalibrate?

RANGER1
14-11-2015, 10:17 PM
Grizzly,
Suggest inspect valve using manual Danfoss magnet tool see if valve turns freely & number of turns specified for that valve.
You could also put guts out of valve to see the same.
See if anything obviously wrong.
Maybe seizing & magnetic drive slipping.
Then recalibrate as Brian suggests.
Have had jammed up threads occasionally.

At work last week same situation as you on same application.
ICAD was first suspected, but ICM was found faulty.
Using the magnet tool it was turned to close, in this case 1.25 turns, remove magnet, then put on again, it could be turned to close again & again etc.
Danfoss will check that one out as weird..

What type of level probe is being used, probe, radar, other?

Grizzly
14-11-2015, 11:15 PM
Hi Guys.
I will let you know regarding the recalibration. As I had to attend site earlier and used i26 setting to recalibrate the ICAD.
Watching it remotely as we speak!

Historically I have always assumed that when you interrupt the power supply.
The settings are factory reset and the valve recalibrates.

The issues have been going on for several weeks but are getting more and more frequent. Basically we are going to suggest that the ICAD 1200 is changed, (got to start somewhere!)
Although it could be the EKC 347 controller?

All settings check out and the level probe I think is a AKS 41?
Come to think! Ranger, don't you close the valve fully and then open 1 turn when setting them up?

I fitted a ICAD 600 and ICM unit a couple of weeks ago!
Some may remember my Plea for an SMV actuator?

Incidentally the magnet tool you mention Ranger comes in 2 sizes. I thought I had the correct one
(part no ends 181) and it turned out to be the larger type.
But if you hold it centrally, when you rotate it.
It will turn the valve so it will get you out of the Poo, should the need arise.
Grizzly

Glenn Moore
14-11-2015, 11:30 PM
Hi Grizz
The ICAD 1200 has an in built encoder so that it should know how many steps it has moved and where it actually is %. So it's a bit weird why it's lost where it is. One of the problems I found was that the controller never stops . The EKC 347 has a digital input terminal 1 & 2 often these just have link wire so the controller and valve is always running. The D1 should break when the plant stops this then closes the valve and recalibrates it. If the D1 is never broken the valve never recalibrates itself.
As Ranger says take the ICAD motor off and wind it manually it should
move the spindle easily. It could be the Pek plastic drive nut is damaged but the encoder should see this and alarm if necessary. Check the magnets in the ICAD make sure they are clean and not corroded
Removing the power will NOT recalibrate the valve only breaking the D1 on the 347 or via the ICAD itself . Ps have AKS 41 cap rod in my garage

Grizzly
14-11-2015, 11:48 PM
Hi Glenn.
I would like to see your Garage some day!

I did have a quick look at the wiring diagrams earlier, when on site with the intention of checking.
whether the terminals 1 &2 were linked or connected to the Unisab / line contactor auxiliary etc.

I think this one must be linked out I did l look at the DI status but cannot remember of the top of my head what it was now.
It all seems to be behaving itself at the moment (Famous Last Words)!

I will report back if there are any changes..

From what I am seeing there is not a problem with the AKS as both the EKC and Level column site glasses concur.
If necessary I will follow all the advice Kindly given above.
Thanks Guys!
Much appreciated.
I need sleep now, so will talk again soon.

RANGER1
15-11-2015, 08:05 AM
Grizzly find turns for each valve size in troubleshooting section


http://www.ra.danfoss.com/TechnicalInfo/Literature/Manuals/01/DKRCIEIHT0B222_ICM_ICAD%20Quick%20Start%20Guide.pdf

Grizzly
15-11-2015, 03:15 PM
Grizzly find turns for each valve size in troubleshooting section


http://www.ra.danfoss.com/TechnicalInfo/Literature/Manuals/01/DKRCIEIHT0B222_ICM_ICAD%20Quick%20Start%20Guide.pdf

Thanks Ranger a useful piece of info for future use.
I cannot give you a rep point as have not issued enough to others.
But thank you all the same.
Grizzly

RANGER1
15-11-2015, 09:20 PM
Thanks Ranger a useful piece of info for future use.
I cannot give you a rep point as have not issued enough to others.
But thank you all the same.
Grizzly

Thanks hope it helps .
I'm not here for rep points, but of course they are a form of appreciation.

Grizzly
20-11-2015, 05:42 PM
Just a quick update Guys!
I changed the ICAD 1200 For it's new ICAD1200A replacement on Wednesday last.
So Far it has stayed rock solid around it's setpoint ever since, despite two site power cuts / blips!
I am not convinced as to the old one being totally at fault so I have kept it to have a play with at some point.
Thanks Grizzly

Magoo
21-11-2015, 01:53 AM
Hi Grizzly.
a good result. Thanks for update
magoo

Grizzly
21-11-2015, 05:43 PM
Hi Grizzly.
a good result. Thanks for update
magoo
Thanks for the acknowledgment!
If more bothered to update their post then more interest would be generated on the forum.
Grizzly

RANGER1
21-11-2015, 10:23 PM
We have recently installed a ICM 50, ICAD on a liquid makeup, on a large older plant.
It is in remote location, so have backup makeup bypassing it(original one with hand expansion valve after it), if new valve cannot keep up level, or fails, old EVRA32 takes over.
We also installed solenoid upstream of ICM as well for safety & can override ICM if out of control.
Also can swing from level probe to float switches if probe plays up.
One falure or hiccup would pay for all control backup if it plays up as long as client drains oil on regular basis.

Recently saw opposition not install solenoid upstream, so will be interesting if power failure.
Don't it has UPS backup.

Grizzly
22-11-2015, 09:05 AM
Hi Ranger.
An interesting setup.
We look after various sites using the ICAD /ICM system and some have ups back-up to drive the valve shut, when a power cut occurs.
Some non critical applications don't and trip on software set within the EKC 347.
Our particular issue this time was not over filling but the plant cycling out on Suction Control. Due to the surge drum being starved of Ammonia.
Controls can be a real pain, even when setup correctly!
This week whilst on a site, the site Fire Alarm was accidentally set off. Shutting down the 3.3Mwatt Ammonia Plant. (As it should!)
Trouble was when I was allowed to return from Muster over an hour later. I had to nurse the plant back on from an elevated surge drum level.

Which being on a bordello tower was a nightmare due to the minimal head above the plates and the large size of the comps.
Long Story short-
By pulling of the saturated vapour too aggressively, a high level alarm (AKS41 Level probe) was being activated.
By generating heat exchange within the plates (Chilled water from factory at about 15c-20c by then) "Vesuvius" was being created within the drum and setting off the High Level Alarm/lockout.

Add a slow acting PID within the controller and you have a High level being created, with a working level you cannot control.
Once at High Level Lockout.
There are no pumps to run to drop the Liquid level in the drum and you cannot run the Comps to lower the level either. (This plant has no High Level overrides.)
I got it all going in the end. But continued to have problems for the next couple of days with the Surge Drum Fill Strategy!
It's been a busy Week as they say!
Grizzly

RANGER1
22-11-2015, 08:37 PM
Grizzly,
Sometimes very difficult when client just wants
production started.
High levels In some circumstances if suction pressure very high, Similar to discharge pressure & surge drum is a bit higher than liquid receiver you can drive excess liquid back up liquid line into liquid receiver.
Haveing continuallypurging ammonia off receiver into water.
If you had a seperate ammonia system you could lower receiver pressure with that using hoses.
Probably has nothing to do with your systems, just ideas.

Grizzly
23-11-2015, 12:07 AM
Falsify the High level setpoint and short spurts in Manual did the trick.
It wasn't a five minute job though, I recon it took a couple of hours the first time.
Shutting the Liquid line valve after the economiser helped also.
Yes, sometimes the High levels were suction related, hence the careful short runs!
I never had to resort to any of your other options,
But have had to connect a charging hose between a HP and LP drum before now.
Thanks, have a good week Grizzly.

PaulZ
23-11-2015, 05:28 AM
Hi Ranger
We do the same with a solenoid before the ICS valve and floats as well as a probe, good back up. If the probe fails for any reason we force the ICS to 50% and lock it and control on the float switches.
We have started using AKS 4100 Cable type probes with a fair bit of success. Oil doesn't seem to affect them like probe type level transmitters.
The other good thing about the AKS 4100 you don't need a lot of height above the control column to get it in and out.
Paul

Grizzly
23-11-2015, 07:00 AM
I totally agree about the AKS 4100 Paul,
We have found the same.

Grizzly

RANGER1
23-11-2015, 08:22 AM
We steer clear of the Danfoss probes after them prematurely bringing them on the market with faults.
Australia is a big place, imagine your in a remote location & your new probe doesn't work. We generally always put back up floats & operate like Paul says.
It's like their LLC switches, the old 38E body bullet proof, the rest is pretty average.
Hansen flavour of a he month at the moment, as PArker floats still causing us grief even after their mods
Scary bit is HAnsen & Parker look like float electrical switch made in same factory
Itall goes round in a circle after awhile

Tycho
10-12-2015, 10:52 PM
Only problem i have had with the ICAD1200 was a couple of years ago.
We opted out the EKC and added the regulator to our existing PLC program.
After all, all the EKC does is send a 4-20mA signal to the ICAD, and the PID regulator in the PLC does the same thing.

the ICAD/ICS was on the liquid line between the HP receiver and LP receiver, it was set up to keep the level on the HP receiver to keep enough liquid in it to keep the oil cooling going.

this was at a fish factory, and outside the season there isn't any load on the system and the ICAD would operate with an opening degree of 2-4% after a week running like this, it would loose it's zero and not close properly, and it wouldn't be noticed before the compressors shut down on high OT and of course we would have to nurse the system back to running conditions, spending 3-5 hours on site to get it going again.

We were in contact with danfoss without getting anything back from them.

after this had happened a few times I did the calibration and then rotated the ICAD on the seat a "smidgen" towards the closed position, and that held for 3 weeks before it failed again...

Danfoss sent us new inserts for the ICS without any changes, until after about a year they sent us the upgraded ICAD to test, and it has worked ever since. :)

Grizzly
11-12-2015, 06:04 PM
Only problem i have had with the ICAD1200 was a couple of years ago.
We opted out the EKC and added the regulator to our existing PLC program.
After all, all the EKC does is send a 4-20mA signal to the ICAD, and the PID regulator in the PLC does the same thing.

the ICAD/ICS was on the liquid line between the HP receiver and LP receiver, it was set up to keep the level on the HP receiver to keep enough liquid in it to keep the oil cooling going.

this was at a fish factory, and outside the season there isn't any load on the system and the ICAD would operate with an opening degree of 2-4% after a week running like this, it would loose it's zero and not close properly, and it wouldn't be noticed before the compressors shut down on high OT and of course we would have to nurse the system back to running conditions, spending 3-5 hours on site to get it going again.

We were in contact with danfoss without getting anything back from them.

after this had happened a few times I did the calibration and then rotated the ICAD on the seat a "smidgen" towards the closed position, and that held for 3 weeks before it failed again...

Danfoss sent us new inserts for the ICS without any changes, until after about a year they sent us the upgraded ICAD to test, and it has worked ever since. :)
Good info Tyco!
All's quiet with the EKC / ICAD issues at present.
So I am keeping quiet!
I don't want to tempt fate!
Thanks Grizzly

Kenneth199
12-12-2015, 03:49 PM
we had the same issue, it was one of the parameters in the icad that was set wrong. Cant remember which one though.