The jaw clutch is getting disengaged at TNH of 39% speed

We have GE Frame -III machine which is equipped with gas turbostarter. The jaw clutch is getting disengaged at TNH of 39% speed.
We found that the angle of the teeth of jaw clutch was changed by friction and became wider.
-Is this problem the cause of disengagement?
-If the L14HA constant is increased to around 60%, what is the consequence, and will the problem be solved?
jaw clutch01.jpg

Hmmm .... Well, those jaw clutch teeth are pretty damaged. (They get damaged by closing them before the unit reaches zero speed when trying to re-start. (That's the most common method of damage, anyway.))

Jaw clutches rely on two (2) things to stay closed: 1) the angle of the teeth (and those teeth are extremely worn down and have virtually no angle left on them); and, 2) the transmission of torque from the starting means (the expander turbine in your case) to the accessory gear shaft. The transmission of torque occurs when the starting means (expander turbine in your case) is providing more torque to the accessory gear shaft--which implies the starting means half of the jaw clutch is spinning (or trying to spin) faster than the accessory gear shaft half of the jaw clutch.

The angle of the teeth, when existing and proper, will help to keep the jaw clutch halves engaged even if the torque from the starting means starts to drop off a little bit (the starting means half of the jaw clutch would be starting to spin a little slower than the accessory gear half of the jaw clutch).

There might be one of two--or possible a combination of the two--problems. First, the torque being produced by the starting means (expander turbine in your case) is insufficient. Something is blocking the flow of gas into the expander turbine, or the expander turbine internal clearances or seals are worn (probably a safe bet judging from the condition of the jaw clutch halves).

Second, the amount of fuel being burned by the gas turbine is causing the turbine shaft (and accessory gear shaft) speed to either suddenly increase or to increase at a faster rate causing the accessory gear half of the jaw clutch speed to exceed that of the starting means half.

One--or both--of these issues is the likely cause of the problem you are experiencing. And the reason I say that is because the starting means half of the jaw clutch is slowing down sooner than it should. OR, the accessory gear half of the jaw clutch is speeding up faster than it should.

Again, sometimes (often) good jaw clutch teeth can alleviate this problem so some extent--as long as the starting means can "keep up" with the HP turbine speed, OR the turbine speed isn't increasing excessively because of increased (intentional or otherwise) fuel flow.

A LOT of time, people just start jacking up fuel flow-rates to try to get the machine to speed faster. This usually also goes along with trying to re-start before the unit reaches zero speed--which causes the damage to the jaw clutch teeth (which is clearly evident in the photograph).

My best guess--based on the information provided (and it would have been VERY GOOD to know at what speed the jaw clutch previously disengaged...!!!)--is that the starting means (expander turbine) is tired and worn. And the jaw clutch teeth are also extremely worn, AND the fuel to the turbine is being increased (intentionally or unintentionally) very quickly causing the turbine to accelerate faster than normal. All of these things, together, are most likely causing the problem. And continually trying to re-start the unit before it reaches zero speed has already caused a lot of damage to the jaw clutch teeth, and possibly the expander turbine and its internal clearances and/or seals.

Hope this helps! Please write back to tell us:

1) When did the jaw clutch halves disengage before this problem started (actually, when the jaw clutch teeth were in good condition)?


2) How you resolve the problem(s).

Because it's likely more than one problem. But that jaw clutch is BEGGING to be replaced. And properly aligned when it is replaced. AND for the flying re-start attempts to stop.

Thank you for your answer, it was really appreciated.

Based on what you are mentioned above, some actions have been performed as follows:

I- Check gas flow path
- Clean the filter
- Adjust flow control valve
- Check turbo expander's wheels by the borescope

II- Engagement system (clutch and hydraulic cylinder)
- Clutch was worn and to be replaced later, actually we don’t have a new one in our store
-Replace hydraulic cylinders, because in some start attempts, clutch was getting disengaged at crank phase, it was the result of internal leaks
Bringing a new jaw clutch may take several days and for production reasons, we decided to force the constant L14HA1 at 55%.

Notice that in normal operation of start means devices :

When (L14HA=1) is picked up the electro valve L20CS will be disenergized (L20CS1X=0), which implied hydraulic cylinder to return to its normal position, and starting means have self-sustaining torque (without hydraulic cylinder force), as HP speed is less than 60%.

Moreover, we fed the turbo expander by 10.5 bars gas pressure, before reaching 55% of HP speed the jaw clutch tried to disengage.

Fortunately, the two halves of clutch keep attached, till HP speed equal 60%

A bog down of 2% occurs, after that the turbine speeded up until the final sequence.

Best regards

Thanks VERY MUCH for the feedback!

If I'm not mistaken, on MOST GE-design heavy duty gas turbines 20CS-1 is de-energized when the shaft breaks away from zero speed (during a START). It's the shape of the jaw clutch teeth that keeps the clutch halves engaged during starting, and when the turbine speed is higher than the starting means side of the jaw clutch the jaw clutch halves automatically separate and the springs on the clutch make sure the halves separate and stay separated. The limit switch then tells the Mark* the starting means is no longer needed.

Further, I believe the speed level you are referring to is "designed" to only open the jaw clutch halves (by stopping the starting means (expander turbine in your case)) if the jaw clutch hasn't automatically opened during starting.... Because at the design speed level the turbine is supposed to be self-sustaining (not require a torque assist from the starting means to maintain speed).

So, it's recommend when you do install a new jaw clutch you need to return all parameters to normal and see how things work and make adjustments from there. You didn't say in what condition you found the expander turbine internals.... and seals (though if the seals are leaking that could be detected with gas sensors/monitors).

When the "bog down" (decrease in turbine speed) occurs after the jaw clutch halves separate it's usually caused by insufficient fuel flow to the turbine--OR insufficient torque from the starting means (expander turbine in your case).... It's also noted that you didn't answer the question of when the jaw clutch "normally" disengaged.... (There are LOTS of similar machines in your Company from which to get data.)

BUT, if you're "goosing" the expander turbine (applying extra pressure to the expander turbine) then that could be contributing to a faster gas turbine acceleration rate than normal and the fuel/speed aren't well-matched. (Of course, it's also possible that incorrect calibration of Gas Control Valve LVDT(s) might be responsible for a slight change in acceleration rate.) The Mark VIe usually uses an acceleration rate reference (TNHAR) to control turbine speed during acceleration (comparing TNHA, actual turbine acceleration rate, to TNHAR, acceleration rate reference. You can monitor both signals during starting to see if possibly the expander turbine is accelerating the gas turbine faster than the Mark VIe's programmed acceleration rate.

But, thanks again for the feedback! It would be GREATLY appreciated if you would continue to let us know how this progresses when you compare TNHAR and TNHA--both now and after the new clutch is installed. (Personally, I suspect the rounded (flattened) jaw clutch teeth are the biggest culprit in this problem, in addition to the other possible causes outlined in this thread.)
To add a little more about "goosing" the expander turbine causing the actual acceleration rate to exceed the acceleration rate reference, when the starting means is disconnected (when the jaw clutch opens) if the actual acceleration is above the required the Mark* will reduce the fuel to try to slow the actual acceleration rate--which could be causing the drop in speed. Again this is speculation (based on experience), but it's something that could be monitored, before and after the jaw clutch is replaced.