I stroked the actuator to 50% to insert the metal block that relieves spring force, so that I could remove the actuator to replace the valve. New valve set at fully closed position, reinstalled actuator, removed block. Started valve calibration application, starts up with -2.5% setpoint, LVDT shows 0% actual position with 0.7 volts. When I enable manual function, desired setpoint goes to zero, but SRV moves to 1.7% actual position with LVDT voltage at 0.75 volts.

The SRV valve symbol on the HMI DLN screen also changes to red (shows open).

 

You Didn't mention machine configuration anway. you do auto calibration of SRV from .m6b and see the the online trend of SRV auto calibration. After completed the auto calibration then give manual command from HMI AND See the performance according to command.

 

I don't know what .m6b is. This is a GE 7FA machine with Mark V.

A question I have is that because the I/O Config is set to look for 0.700 V as a zero value for SRV, that when the default -2.5% bias is removed and set to 0% by the Mark V valve calibration function, the servo outputs from R,S, T would move the valve to 0.700 V feedback.

 

Jeff Ladwig,

I don't recognize the type of SRV you are describing.

I also don't understand what you mean when you say the Mark V is configured to look for 0.700 volts at zero stroke. The Mark V doesn't care what the zero-stroke voltage is as long as the feedback is linear. The LVDTs GE uses are supposed to be adjusted such that the zero-stroke voltage is 0.700 VAC RMS, +/-0.02 VAC RMS. Setting the zero-stroke voltage to this range ensures the output (feedback) from the LVDT will be linear over the range of travel of the device to which the LVDT is connected.

When AutoCalibrate is used, it will measure the exact zero-stroke voltage (and 100%-stroke voltage,also) so that it can then be entered into the I/O Configurator.

There is an exception to this for some SRVs with rotary variable differential transformers (RVDTs). They usually have a linear output range of 0.05 - 1.20 VAC RMS, which results in a very poor feedback but more importantly if set to 0.700 VAC RMS at zero stroke the feedback as the valve was opened to 100% would become non-linear.

Lastly, with the above exception, the linear range of most LVDTs used by GE was 0.700-3.500 VAC RMS. That <b>DOES NOT</b> mean that in every application the feedback at 100% will be 3.500 VAC RMS. It only means the feedback is linear <b>BETWEEN</b> 0.700 and 3.500 VAC RMS. If the stroke (travel) of the device the LVDT is connected to doesn't make the output of the LVDT reach or exceed 3.500 VAC RMS then that's perfectly okay--because the LVDT output (feedback) will still be linear at values less than 3.500 VAC RMS.

But if the LVDTs were replaced or disturbed during your work, it will likely be necessary to readjust the zero-stoke voltage, but it absolutely be necessary to re-calibrate the LVDT feedback to get the new values for the Mark V to use to scale the feedback such that the displayed position very closely matches the actual, measured position.

Hope this helps!

Even if the LVDTs weren't disturbed during the work it seems the feedback may need to be re-calibrated.

 

dear,

so, example you mean if the actual feedback just 2.4 Vrms for Open 0%, even the linear 2.7 Vrms for open 0, doesn't it matter?

thank you

 

yoga,

What I'm saying is the Mark V doesn't care what the zero-stroke voltage is or what the 100% stroke voltage is--<b>AS LONG AS THE VOLTAGE FEEDBACK BETWEEN THEM IS LINEAR.</b>

A lot of people believe the Mark V <b>MUST</b> see 0.700 VAC RMS at zero stroke and that's <b>SIMPLY NOT TRUE.</b> The Mark V needs to see a zero-stroke voltage that is in the linear output range of the LVDT. Most of the LVDTs GE and the packagers of their turbines supply are chosen such that when the zero-stroke voltage of the LVDT is set to 0.700 VAC RMS the output is in the linear range of the LVDT.

And when the device the LVDT is connected to is at its 100%-stroke position the output of the LVDT will still be in its linear range and will not be more than 3.500 VAC RMS.

LVDTs have a non-linear range at the low end of their output <b>AND</b> at the high end of their output--and a linear output between those two non-linear ranges. That's why it's important to set the zero-stroke voltage to an output value that is above the lower non-linear range. GE or its packagers are supposed to have chosen the LVDT so that when it's set to a linear output value at the zero-stroke position the range of travel will not exceed the linear output range of the LVDT. Further, they typically supply (purchase) LVDTs that will have a linear output between 0.700 VAC RMS and 3.500 VAC RMS on the device they are attached to.

But that does not mean the LVDT output will always be 0.700 VAC RMS at zero stroke <b>OR</b> that the Mark V is configured to (must have as its input) 0.700 VAC RMS at zero stroke. The only thing the Mark V must have is a linear voltage feedback between the zero-stroke position and the 100%-stroke position. The Mark V will take care of the rest as long as the input is linear.

Does that help?

If the 100% stroke voltage is 2.4 VAC RMS or 2.7 VAC RMS, it doesn't matter as long as the voltage output of the LVDT is linear between whatever the zero-stroke voltage is and the 2.4 or 2.7 VAC RMS value. The Mark V doesn't care what the zero- and 100%-stroke voltages are--as long as the voltage between them is linear over the range of travel of the device the LVDT is attached to.

You're welcome.

 

Dear CSA,

it is very help me. I agree with you, 2.4 VRms or 2.7 Vrms doesnt matter with the different voltage as long as the actuator will move linear

mmhh, i have a problem when i must calibrate PID. have u manual tunning of Woodward 505? or it just uses trial and error method?

I think use control with servo valve is too risk, because it is very sensitive, because why?

if there are contaminants, such as water contents, impurity and etc.
when servo valve has affected by contaminant, it will make servo valve not work well. it will make trouble in its mechanical valve movements

thank you

 

yoga,

You are going to have to refer to the documentation provided with the equipment at your site for information about calibration procedures. You may find some useful documents at the Woodward Governor website.

Hydraulic actuators are very commonly used--and for the most trouble-free operation it is very important (critical) to take the necessary steps to ensure the oil is clean, free of impurities and serviceable. There is an entire industry segment devoted to oil treatment and purification including testing.

 

Yes, explanation is helpful. I am still wondering why when the SRV command setpoint changes from -2.5% to 0.0%, the valve actually moves to about 1.3% open. Seems to me that the valve should not move off of zero actual position.

 

Jeff Ladwig,

Either the LVDT calibration is not correct, OR the servo null bias is not correct, OR the servo wiring (polarity) is not correct, OR some combination of all of the above.

 

Does your control does calibration procedure include a null bias check? If so, have you adjusted the null bias value in the I/O configuration per the procedure?