MARK-VIe TMR Servo Control Issue(Valve Fluctuating)

We are using mark-vie for combined steam turbine control. After commission, we have some problem in TMR Servo Valve control. Three lvdt are installed in one valve and Three Servo Regulators control the servo valve. There are cv#1, #2 which have same parameters in our control system. Only cv#1 have been fluctuated with 3% range (ex> demand: 40% , feedback: 37%~43%) does not only just feed back also actual valve fluctuating. so when unstable condtion Electrical Hydrauric Control Line are vibrated criticaly.

FIRST: we already changed servo valve to new one, but problem still exist.
SERVO VALVE: REXROTH HYDRAULICS: 4WS2EM, nominal current per coil 30ma, resistance per coil 85ohm)

second : Althought we don't think there are some problem in our lvdts. we changed a one lvdt to new one because of both low probability in lvdt malfuction and our system with tmr control, but problem still exist

third: that is very interesting.
after we disconnected the R core from TMR servo control, Valve status back to normal
so we tested other core S,T one by one. Results are same above.


fourth: in normal operation our cv#1#2 are full open status. At that time, Valve is not moving any more (demand over 100%), but in other positions (20%, 30%, 50%, 70%..). During start up , we suffer the problem.

Could you let me know how to solve this problem.
Now i consider the pvso regulator gain adjust, but i want more information about this.

I wish I could say with a high degree of certainty (as I'm sure you would like to hear, also!): Do this and your problem will go away.

But, I don't have any (as in zero) experience with Rexroth servos. I have no idea if they are unipolar or bipolar or anything about them. You haven't told how the output is configured (in ToolboxST or via the jumpers on the TSVO card). I have seen a couple of combined cycle steam turbine applications recently that DID NOT use LVDTs for control purposes--ONLY for position indication and nothing more.

I have a couple of suggestions, but neither is likely to be the problem. First, have you performed a servo current polarity check? This is probably NOT the problem because, if I understand your "chart/table" properly you operated (opened/closed; "stroked") the valve under the control of <R> servo output only. And, if the servos (and the servo output) are unipolar (meaning the current output is ONLY positive (or ONLY negative), then this likely is NOT the problem. (Bipolar servo current is current that goes positive and negative during the course of normal operation, and when the valve is in a steady-state position and stable the current is very close to 0 mA (not exactly zero, but very close).

The second is to check the reference being developed by <R>. It sounds like there is something amiss with this, or the position feedback (if the loop is even using position feedback). If there are unusual Diagnostic Alarms (voting mismatches, probably, and possibly associated hardware Diagnostic Alarms (such as with the PSVO I/O Pack for <R>, or <S> and <T>).

If this is an upgrade from an older turbine control system to a newer turbine control system (again, something you didn't tell) and it worked before the upgrade, and there wasn't any changes to the control valve during the outage (valve; actuator; etc.) then it could be a problem with the <R> I/O PSVO I/O Pack. But, again, I would suspect some Diagnostic Alarms (possibly lots of them!) for that pack, or for that pack as well as the others on that TSVO card (I'm presuming this is a TSVO card with three PSVO I/O Packs--we don't know for sure...!).

That's about all I can think of. I am actually thinking this is a control system problem (based on the information provided--and there wasn't a lot of that) and not a mechanical problem or an input/output (LVDT or servo) problem. If the unit control system has some kind of inlet pressure control that is in service during start-up/shutdown and there are multiple inlet pressure transmitters that could possibly be a problem--depending on how the transmitters are connected to the Mark VIe (difficult to know without seeing the ToolboxST configuration or drawings). Or, if power control is being used and there are multiple, redundant load transducers being used that could also be a problem. But, again, with EITHER of these possibilities (input/feedback problems) I would expect there to be associated Diagnostic Alarms.

Please write back to let us know what you find. The more information you can provide the better, too, AS ALWAYS. Hope thuis helps!

I want to also add that it could be a mechanical issue with the actuator of the control valve with a problem--leaky seals; worn cylinder; etc. Particularly if the Mark VIe is an upgrade to an older control system on an older unit.

I have even seen low quality actuators used on new units have problems, as well as poor oil flushing (or NO oil flushing) cause problems with good equipment during a new unit installation.

Again, there's a lot we don't know about the situation at the site.

Please do write back with information about what you find!