variation in speed and amperes of Siemens perfect harmony drive 2

[fiaz1404]

We have ID fan equipped with 2000kw, 207 amp motor which is controlled by Siemens perfect harmony drive. Normal operational current is 110 amps and speed is 810rpm. We are facing a problem that drive amperes and speed suddenly start to vary from 75 amps to 160amps while speed varies between 800 and 830 rpm. I checked the drive set point reference which is quite stable in those moments.
To normalize the system we have to reduce the set point reference to minimum 200 rpm and increase again up to required operational level. Everything then becomes normal, No variation in speed or amperes. If we will not exercise this practice, drive will continue this variation for hours. I believe to exclude mechanical or process problem as system becomes absolutely normal after this practice. This phenomenon is happening once or twice weekly.
Thanks for any help
Fiaz Ahmed

 

[itsmoked]

What's an "ID fan"?
What is the fan actually doing, its 'system load'?

Is the drive running a PID? If it is it sounds like it is out of tune and causing the system to oscillate.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[davidbeach]

Induced Draft fan.

 

[itsmoked]

Thanks David. A 'sucker' fan.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[fiaz1404]

Dear itsmoked,

1. It is Induced draft fan which is responsible for flow of necessary gases from kiln to exhaust fan in cement plant.
2. I will check if drive is running a PID
3. if drive is running on PID,What are the possible causes of PID loop out of tune?

 

[itsmoked]

Hi fiaz.

Any kind of change in the process air flow path could shift the PID possibly to an out of tune condition. It probably would've been tuned when, perhaps, the kiln was new and now over time;
The fan's blades have gotten fouled
The stack is reduced in diameter due to residue
Dampers have changed
Prevailing winds have changed with climate
A paired forced draft fan has changed its operating point
A change in firing rate
Fan bearings are failing changing the torque requirements
Someone changed fan blades


Other possibilities unrelated to a PID issue could be:
A bad connection causing brief single phasing or feedback errors.

I am unfamiliar with that particular drive but if semiconductors are involved I can see thermal problems causing behavior like that. Slowing it way down and slowly speeding it back up will have the effect of dropping temperatures everywhere in the system, the semiconductors themselves and all wiring connections. Once cool they work fine again for a while.

Has anything changed that could cause the control boards and circuitry to run hotter than 'before'? Clogged filters, bad fans, dirt and dust? Clogged room A/C filter? I've seen this kind of behavior many times from overheating controls. It causes the classic open panels with floor-stand fans blowing on the controls.


Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

This part of your question says you have already checked the analog command signal and it's not oscillating which means the PID loop controlling fan speed should be fine.

I checked the drive set point reference which is quite stable in those moments.

The VFD also has some tuned loops inside. I'd suspect those possibly because the drive was poorly tuned to the motor during commissioning.

 

[fiaz1404]

Thanks all for your valuable inputs
This drive is receiving one fixed analogue input as speed reference from central control station which is only changeable manually by operator. There is no any PID control involve in operating station.
There are two types of internal PID control loops available inside the drive. One is the control regulator for flux and torque producing components. I think, parameters of these control loops are updated when auto tuning is performed. Other type of loop which may be used as external PID & analogue input may use for this PID reference is not active.
Does performing auto tuning will help?
itsmoked,
We didn’t change anything in drive, cleaning of drive air filters is weekly basis, room temperature is maintained to 20 deg Celsius. As we are able to restore normal operation within one minute so I think it is not a thermal problem as you indicated.

 

[jraef]

But poor tuning would not “fix” itself and run fine after a while.

The Perfect Harmony drive is a Medium Voltage low harmonic drive using what’s called a “Cascaded H Bridge” topology that accepts input from a multi-phase (24, 36, 54 pulse depending on voltage) isolation transformer into a set of low voltage rectifiers and inverters with their outputs put in series to recreate the MV output. It’s likely that one (or more) of your “power cells” is intermittently failing in some (so far) non-catastrophic manner. In that drive design, they have a feature called “automatic cell bypass” in which a dead (or failing) LV power cell is detected, then bypassed with an internal contactor.

But in order to maintain balance to the motor, corresponding cells in the other phases are also bypassed and the effective output voltage lowered. This results in rendering the drive incapable of providing full output, so the drive speed modulation is throttled back irrespective of the commanded speed. The basic concept is “giving you something is better than shutting down completely”.

In that process, there will be messages and warnings displayed on the HMI of the drive, also made available to any control system communications (if any). SOMEONE in your facility should be getting these messages and/or understand what they mean. If not, you need to contact Siemens for some training!


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[fiaz1404]

Automatic cell bypass feature is optional in perfect harmony drive; this feature is not available in our installation. When this problem rises, there is no any alarm or fault message appears on the drive HMI

 

[LionelHutz]

One is the control regulator for flux and torque producing components

This is the loop I'm referring to. The drive can do odd things when this loop is poorly tuned.

 

[fiaz1404]

Thanks to LioneIHutz and all participants in this discussion.
I did auto-tuning and didn't face problem since more than a month.

 

[waross]

If possible I would try to add a few percent to the proportional band of the PID loop.
The auto-tuning may be leaving the PID loop to close to instability.
I have seen a similar issue with conventional PID controllers.
A small increase in the proportional band solved the issue.

Bill
--------------------
"Why not the best?"
Jimmy Carter