General Operability Discussion

[finiteparts]

Some simple suggestions to all experimenters out there:

1.  When making introductory engine test, it is always better to make small or slow engine rpm changes. Large transient shifts can eat up a lot of the clearances and the potential to experience compressor or turbine rubs increase.

1a. If we take the turbine as an example, to increase the engine rpms, we increase the fuel flow, which increases the combustor discharge temperature. Since the rotor has a finite amount of rotational inertia, it does not respond instantly to the increase in available energy. So what happens is there is a momentary temperature overshoot due to a rotor torque imbalance, the turbine has increased its torque, but the compressor is still demanding the earlier lower torque demand. But, before the rotor catches up, the local increase in temperature at the NGV, pushes the compressor operating point up the constant speed line, increasing PR and reducing mass flow. This off-condition persists till the rotor is finally accelerated to the new steady state condition.

1b. The other factor of concern in addition to the rotor inlet temperature overshoot, is the thermal time constants of the various parts. The thermal time constant is the amount of time it takes for a mass to reach some level of the steady state temperature, often taken as about 77% (RMS) of the final temperature. The rotor is immersed in the hot gas and responds very quickly to the increase in gas temperature. The shroud does not respond as fast and thus has a longer time constant. The problem comes up when the time constants are sufficiently different, because, the rotor will thermally grow outward, while the slower shroud will not grow out of the way. The clearance is reduced due to the mismatch in these thermal growths and if sufficiently different, the clearance can be reduced to zero or even negative clearance values (clashing), thus a rub.

If the thermal changes are done slowly enough, the thermal time constant mismatch is less significant as both the shroud and turbine can thermally grow at a more similar rate.

2.   Rapid changes in rotor speed can drive instabilities in the rotordynamics.  Since a high speed rotor is subjected to gyroscopic effects, a sudden change in rotor speed can produce a reaction that can lead to a precession like effect that may force the rotor to tilt.   This reaction can trigger an instability especially if the rotorynamic behavior of the system is teetering on the edge of resonance or bending mode.

3. Finally, the response of the thermocouples and pressure measurements also need some amount of settling time to give accurate values and rapid changes can give very unexpected results.

[finiteparts]

Hi Ralph,

I thought it might be better to move this discussion over here as opposed to hijacking John's thread.

jetandturbineowners.proboards.com/post/34888

Things are going well for me in general and I hope they are also going well for you also.

Ok, so if you replace the compressor wheel in an existing system, all dimensions are the same except now you have a shorter tip height, which I am reading as B-width, what happens?

Generally, the flow is limited in the system by the compressor inducer's, the nozzle guide vane's and the exhaust nozzle effective throat areas.   If all of these areas are staying the same, what else might change the flow rate?  The flow rate through a given orifice area is controlled by the pressure ratio across it and the local loss features that impact the discharge coefficient.  So would the change in the impeller's b-width impact the compressor system overall pressure ratio?

I think so.   The primary role of the b-width is to set the impeller exit conditions and reducing the b-width increases the relative and absolute discharge velocities. If only the b-width is reduced, the absolute discharge velocity increases and thus reduces the absolute discharge angle. The lower tangential velocity falls out from a reduction in the impeller work coefficient and by extension, a reduction in the exit total pressure, i.e. less kinetic energy has been imparted by the impeller to the flow. Remember, the amount of power absorbed by the rotor correlates directly to the difference in the amount of whirl the airflow exhibits between the inlet to the exit. 

As has been discussed elsewhere, the impeller can be thought of as a rotating diffuser stage that is stabilized by the centrifugal force imposed on the flow, thus the impeller is more capable of diffusing the flow with lower losses than typically done in the stationary hardware.  So by reducing the b-width, the diffusion capability of the stage is effectively reduced (to some extent) since you are now placing more demand on the stationary diffuser components to recover the flows kinetic energy in an efficient manner.   This is usually reflected in the map as a shift of the peak efficiency island to a higher rotor speed, but with a vaned diffuser, this would also imply that the diffuser was re-staggered to meet the new impeller discharge angle.  If the vaned diffuser was left as was (except maybe shorted to meet the new b-width), it would likely be mismatched at the design point.

Finally, the tip clearances are now a larger portion of the b-width, so the efficiency of the stage will usually be decreased.

So how does this impact the whole of the system? 

Disclaimer: This is just how I would think of it, I haven't done any calculations to actually predict the overall system behavior, so I could be wrong. 

The reduced work coefficient would suggest that the torque balance between the turbine and compressor would likely happen at a higher rpm, so that at a matched compressor overall PR (for original vs reduced b-width), the rotor would be going faster.  The reduced compressor efficiency may or may not be an issue since the peak efficiency island of the compressor map would have likely shifted upward and may be at a similar or better matched condition...this can only be know from calcs or testing.
The map width of the modified compressor will likely be narrower.  The engine overall flow rate should be similar, since the controlling flow areas haven't changed and the system can re-balance to achieve similar pressure ratios.

So to answer your questions:

1. the maximum pressure remains the same but the RPM is lower?

Not necessarily.  The engines overall PR may be similar or larger, depending on how the comp and turbine balance out.  But, the compressor system PR will be lower at the same RPM, or rebalanced to a similar PR at higher RPMs


2. the flow rate and also the thrust is lower?


Again, the engine flow rate shouldn't be much different since none of the controlling areas changed.  The thrust should also be similar if the rotor torque can re-balance at a higher rpm, since the mass flows shouldn't change significantly and the PR can be similar.


3. what about the temperatures of NGV/turbine?

If the comparison point is a similar PR, then the expectation is that the temps will be up slightly, but not largely different if the rotor can torque balance at a higher rotor speed and there are no really off conditions on the vaned diffuser.

I hope that makes sense...not saying it is totally accurate, but just my thoughts on what it may do.

- Chris

[finiteparts]

jetandturbineowners.proboards.com/post/34899

I was giving this some thought and the problem I am having is, the exit temperature doesn't give enough information to say that it is an good or not.

If the longer inlet causes a slight inlet pressure loss, then the compressor will do slightly less work, thus lowering the exit temperature all other things being equal.

Or maybe the inlet pressure loss makes it require more work due to a lower efficiency of the compressor stage and this is reflected by a larger temperature drop through the turbine stage...

Does it reduce the EGT while maintaining exactly the same thrust level and the same fuel flow?

There are a lot of variables that come into play and I don't think that just a statement of lower EGTs gives enough information to figure out the physics of it.

Finally, I was going down the same line of thinking as John stated, show it aids the flow direction at the impeller inlet face, but I am diverging from that line of thinking.  There is nothing in the tube to provide any sufficient flow control.  I am more on the line of thinking that the longer inlet would impose a slightly larger pressure loss prior to the compressor inlet and this causes the overall system to re-balance somehow and the lower EGT is a side effect of that.

- Chris

[ripp]

Hi Chris,

Thanks for your detailed answer!

Ralph