Bendix FCU/GOV fundamental question
Thread Starter
Join Date: Jan 2019
Location: Mill Valley
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
To put things in order -
Flyweight governors have inherent droop - this gives stability to stop it hunting incidentally.
Droop occurs because the spring tension is non-linear and when it is back "on speed" to equal the spring tension it will be in a different place to equal the spring tension = droop. Google "mechanical governor droop"
If you think about it if the tension was linear it wouldn't work.
The linkage from the collective is the "droop compensation" to adjust the new set point with the change in power back to where it should be.
To adjust for more rapid changes the droop compensation is normally "over compensated".
Set 100% N2 in a Bell 206 flat pitch - pull it up into the hover and it should be higher.
The beep trim is there to compensate for atmospherics - the P in DP N1/N2 is a pressure which varies from day to day.
Set the correct N2 with the beep trim and forget about it - if you are constantly playing with the beep trim something is broken or out of rig.
The Bendix system has been around since Pontius was a Pilate and works pretty well considering. The operation and maintenance manual is around online for free.
Flyweight governors have inherent droop - this gives stability to stop it hunting incidentally.
Droop occurs because the spring tension is non-linear and when it is back "on speed" to equal the spring tension it will be in a different place to equal the spring tension = droop. Google "mechanical governor droop"
If you think about it if the tension was linear it wouldn't work.
The linkage from the collective is the "droop compensation" to adjust the new set point with the change in power back to where it should be.
To adjust for more rapid changes the droop compensation is normally "over compensated".
Set 100% N2 in a Bell 206 flat pitch - pull it up into the hover and it should be higher.
The beep trim is there to compensate for atmospherics - the P in DP N1/N2 is a pressure which varies from day to day.
Set the correct N2 with the beep trim and forget about it - if you are constantly playing with the beep trim something is broken or out of rig.
The Bendix system has been around since Pontius was a Pilate and works pretty well considering. The operation and maintenance manual is around online for free.
Thread Starter
Join Date: Jan 2019
Location: Mill Valley
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
thanks!
FYI: In a B206, the term "droop" relates more to the positive/negative shift in M/R RPM when power is pulled into a hover from 100% Nr at flat pitch. If the Nr does not recover to 100% in a hover then that "droop" is compensated by airframe rigging adjustments. This adjustment physically changes the throw and range of the GOV arm arc as it moves in relation to the collective. It does not reset anything internally in the GOV.
Thread Starter
Join Date: Jan 2019
Location: Mill Valley
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
Curious. What are you defining "static droop" as? In the big picture, "droop" compensation is a function of the airframe N2 control system. Internally (engine/FCU/GOV) the GOV's sole job is to "reschedule" the FCU to maintain N2 turbine speed. There are several "precursors" which must happen before the GOV detects and reacts to maintain that N2 speed. Since the system is mechanical there is a certain "dwell" period between these "precursors" which adds to the transient time period between the GOV's detection and reaction to the power changes. Plus add a second or two for the GOV reaction signal to make its way to the power output of the engine.
FYI: In a B206, the term "droop" relates more to the positive/negative shift in M/R RPM when power is pulled into a hover from 100% Nr at flat pitch. If the Nr does not recover to 100% in a hover then that "droop" is compensated by airframe rigging adjustments. This adjustment physically changes the throw and range of the GOV arm arc as it moves in relation to the collective. It does not reset anything internally in the GOV.
FYI: In a B206, the term "droop" relates more to the positive/negative shift in M/R RPM when power is pulled into a hover from 100% Nr at flat pitch. If the Nr does not recover to 100% in a hover then that "droop" is compensated by airframe rigging adjustments. This adjustment physically changes the throw and range of the GOV arm arc as it moves in relation to the collective. It does not reset anything internally in the GOV.
I think I understand how the compensation works, as it just pre-loads the speeder spring enough to offset the inherent static droop (figure from manual). In addition to helping the steady state be closer to 100%, the compensation also lessens the transient droop by introducing fuel immediately on collective movement as opposed to waiting for the RPM drop to manifest in N2.
RVDT , in is first answer, asserted "Droop occurs because the spring tension is non-linear and when it is back "on speed" to equal the spring tension it will be in a different place to equal the spring tension = droop. Google "mechanical governor droop"
Maybe this is the whole story and the inherent droop is just determined by the speeder spring strength. (and non-linearity - fuzzy on this detail)
I may have become confused by looking at "Governing Fundamentals" (Woodward power) They show how static droop can be engineered with a "droop lever" which "weakens" the speeder spring tension when fuel is being added. I think I assumed that this extra mechanism was necessary to create droop and we don't see anything like it in the 206 governor. However, if the governor can exhibit static droop just by virtue of the spring characteristic, then the question is answered.
To be clear, my question is just one of curiosity not an operational question, I'm figuring out how to fly it ;-)
John
By the way, I'm sure my reply other reply to wrench1 will show up in several hours after "approval". I just have to say, this forum is extremely unfriendly to noobies. The long lags in post approval combined with the inability to include images. It's like 1980 all over again. Maybe that's how you want it ?!
Now in some other older droop designs they also added an "anticipator" or "bias" device that "anticipated" the droop and reduced its effect (value). While the "non-linear" side of your statement is at or beyond my skill set, I seem to recall this type of governing system had to be developed in some form of non-linear fashion in order to function throughout its control range due to the variables of the signal inputs/outputs. Its my understanding that a digital controlled governing system is more linear due to the fact the input/output signals can be processed almost instantaneously via the EECU or similar unit and acted upon.
Having been riding shotgun in a 206 when it had a high side governor failure in a difficult location you will appreciate what that GOV does. Lucky for me, the pilot had a light touch and we made it down without me having to change the engine that night. If I can get it to post, I'll put up a booklet put out by Bendix/Allied Signal on your GOV/FCU system for your reading pleasure. Good luck.
autojohn
Pr = Pressure regulated
Pg = Pressure governing
The PTG tells the FCU by biasing the Pg pressure relative to the fixed Pr pressure. It is simply a lever with a face over the end of a hole in a jet.
To change the bleed out of the jet to a different value means that the lever will be in a slightly different position to maintain that value = droop.
The difference in the value Pr/Pg = fuel flow required to maintain onspeed at the higher load. The onspeed condition where the lever arrives at equilibrium will be different.
Droop stops hunting or instability - imagine if there was no speeder spring - wouldnt work at all.
Droop compensation from the link to the collective adjusts the speeder spring load to remove the effect of droop but the characteristic is still there just masked or compensated for.
Most mechanical systems that I have seen in helicopters are overcompensated. Nr or N2 will rise with the application of collective and vice versa (if correctly rigged) This deals with the rate of change in power setting.
Lookup PID controllers - in a mechanical governor you basically only get P = proportional.
FADEC you get the lot - Proportional Integral and Derivative
Where the Bendix has an issue is the quality of the air it uses - smoke particles corrosion and other detritus accumulate on the lever and on the jet face and wear it out as well as changing the jet size.
Even FADEC's are not immune as they still sense P. On the PW200 series, for example, they put a tee in the P line that feeds a rather cunning setup that uses air flow to the output shaft seal. Where the P is sensed at the tee there is no flow
just pressure. Guess where all the dirt likes to sit!!
I hope this makes sense.
Pr = Pressure regulated
Pg = Pressure governing
The PTG tells the FCU by biasing the Pg pressure relative to the fixed Pr pressure. It is simply a lever with a face over the end of a hole in a jet.
To change the bleed out of the jet to a different value means that the lever will be in a slightly different position to maintain that value = droop.
The difference in the value Pr/Pg = fuel flow required to maintain onspeed at the higher load. The onspeed condition where the lever arrives at equilibrium will be different.
Droop stops hunting or instability - imagine if there was no speeder spring - wouldnt work at all.
Droop compensation from the link to the collective adjusts the speeder spring load to remove the effect of droop but the characteristic is still there just masked or compensated for.
Most mechanical systems that I have seen in helicopters are overcompensated. Nr or N2 will rise with the application of collective and vice versa (if correctly rigged) This deals with the rate of change in power setting.
Lookup PID controllers - in a mechanical governor you basically only get P = proportional.
FADEC you get the lot - Proportional Integral and Derivative
Where the Bendix has an issue is the quality of the air it uses - smoke particles corrosion and other detritus accumulate on the lever and on the jet face and wear it out as well as changing the jet size.
Even FADEC's are not immune as they still sense P. On the PW200 series, for example, they put a tee in the P line that feeds a rather cunning setup that uses air flow to the output shaft seal. Where the P is sensed at the tee there is no flow
just pressure. Guess where all the dirt likes to sit!!
I hope this makes sense.
Last edited by RVDT; 24th Jan 2019 at 23:53.
Thread Starter
Join Date: Jan 2019
Location: Mill Valley
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
autojohn
Pr = Pressure regulated
Pg = Pressure governing
The PTG tells the FCU by biasing the Pg pressure relative to the fixed Pr pressure. It is simply a lever with a face over the end of a hole in a jet.
To change the bleed out of the jet to a different value means that the lever will be in a slightly different position to maintain that value = droop.
The difference in the value Pr/Pg = fuel flow required to maintain onspeed at the higher load. The onspeed condition where the lever arrives at equilibrium will be different.
Droop stops hunting or instability - imagine if there was no speeder spring - wouldnt work at all.
Droop compensation from the link to the collective adjusts the speeder spring load to remove the effect of droop but the characteristic is still there just masked or compensated for.
Most mechanical systems that I have seen in helicopters are overcompensated. Nr or N2 will rise with the application of collective and vice versa (if correctly rigged) This deals with the rate of change in power setting.
Lookup PID controllers - in a mechanical governor you basically only get P = proportional.
FADEC you get the lot - Proportional Integral and Derivative
Where the Bendix has an issue is the quality of the air it uses - smoke particles corrosion and other detritus accumulate on the lever and on the jet face and wear it out as well as changing the jet size.
Even FADEC's are not immune as they still sense P. On the PW200 series, for example, they put a tee in the P line that feeds a rather cunning setup that uses air flow to the output shaft seal. Where the P is sensed at the tee there is no flow
just pressure. Guess where all the dirt likes to sit!!
I hope this makes sense.
Pr = Pressure regulated
Pg = Pressure governing
The PTG tells the FCU by biasing the Pg pressure relative to the fixed Pr pressure. It is simply a lever with a face over the end of a hole in a jet.
To change the bleed out of the jet to a different value means that the lever will be in a slightly different position to maintain that value = droop.
The difference in the value Pr/Pg = fuel flow required to maintain onspeed at the higher load. The onspeed condition where the lever arrives at equilibrium will be different.
Droop stops hunting or instability - imagine if there was no speeder spring - wouldnt work at all.
Droop compensation from the link to the collective adjusts the speeder spring load to remove the effect of droop but the characteristic is still there just masked or compensated for.
Most mechanical systems that I have seen in helicopters are overcompensated. Nr or N2 will rise with the application of collective and vice versa (if correctly rigged) This deals with the rate of change in power setting.
Lookup PID controllers - in a mechanical governor you basically only get P = proportional.
FADEC you get the lot - Proportional Integral and Derivative
Where the Bendix has an issue is the quality of the air it uses - smoke particles corrosion and other detritus accumulate on the lever and on the jet face and wear it out as well as changing the jet size.
Even FADEC's are not immune as they still sense P. On the PW200 series, for example, they put a tee in the P line that feeds a rather cunning setup that uses air flow to the output shaft seal. Where the P is sensed at the tee there is no flow
just pressure. Guess where all the dirt likes to sit!!
I hope this makes sense.
I think you're saying that the Pg/Pr differential biases the FCU speeder spring continuously across the range from flight idle fuel flow up to max power fuel flow. At flight idle, you need the biggest differential to suppress N1. This corresponds to the lowest Pg => most open port => highest flyweight position. The beeper lets you tune this to 100% for variation in Pa. At max power, you need the least bias = highest Pg = closed port. The Pg jet is the most closed so the arm is displaced the least => maximum static droop! (please let this be right ;-)
The compensation cam just adds a bit of tension to the spring so that you really need a bit more RPM to get to this almost closed arm position.
I hope I'm getting closer and I appreciate your explanation RVDT!
Thread Starter
Join Date: Jan 2019
Location: Mill Valley
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
We're talking about the same topic just opposite ends of the spectrum. A couple more tidbits. Keep in mind this droop is nothing more than a "feature" of an analog-controlled governing system. In digitally-controlled governing systems droop is basically non-existent even on the same core engines. But from my understanding there are a few other items that determine the inherent droop other than spring strength in an analog system. There is also the possibility of positive "droop" which I've seen called "droop cancelling"(?).
Now in some other older droop designs they also added an "anticipator" or "bias" device that "anticipated" the droop and reduced its effect (value). While the "non-linear" side of your statement is at or beyond my skill set, I seem to recall this type of governing system had to be developed in some form of non-linear fashion in order to function throughout its control range due to the variables of the signal inputs/outputs. Its my understanding that a digital controlled governing system is more linear due to the fact the input/output signals can be processed almost instantaneously via the EECU or similar unit and acted upon.
Having been riding shotgun in a 206 when it had a high side governor failure in a difficult location you will appreciate what that GOV does. Lucky for me, the pilot had a light touch and we made it down without me having to change the engine that night. If I can get it to post, I'll put up a booklet put out by Bendix/Allied Signal on your GOV/FCU system for your reading pleasure. Good luck.
Now in some other older droop designs they also added an "anticipator" or "bias" device that "anticipated" the droop and reduced its effect (value). While the "non-linear" side of your statement is at or beyond my skill set, I seem to recall this type of governing system had to be developed in some form of non-linear fashion in order to function throughout its control range due to the variables of the signal inputs/outputs. Its my understanding that a digital controlled governing system is more linear due to the fact the input/output signals can be processed almost instantaneously via the EECU or similar unit and acted upon.
Having been riding shotgun in a 206 when it had a high side governor failure in a difficult location you will appreciate what that GOV does. Lucky for me, the pilot had a light touch and we made it down without me having to change the engine that night. If I can get it to post, I'll put up a booklet put out by Bendix/Allied Signal on your GOV/FCU system for your reading pleasure. Good luck.
And indeed the manual shows the anticipator (aka droop compensator) which allows collective position to turn a biasing cam in the governor. Such a cool document!
Thinking of the governor taking a nap gives me sweaty palms, glad you made it out!
John
Now if you really want to get confused you should read about the CECO fuel control system that was offered on C20/C20Bs back in the day. It used metered fuel pressure as the signal medium between the FCU and GOV instead of air pressure as the Bendix does. The Bendix system was a godsend--at least from a maintenance point of view. I still get a tick when I mention CECO... CECO... CECO...
And just to clarify. A true "anticipator" or "bias" airframe N2 system works a bit different than the Bell airframe compensation system and is more responsive in correcting droop.
FYI: I wouldn't sweat the probation period on new posts. I've noted most websites that allow open, anonymous viewing of posts usually have a similar policy when it comes to attachments. Only the sites that require a login to view posts seem to have a lesser restriction.
Thread Starter
Join Date: Jan 2019
Location: Mill Valley
Posts: 10
Likes: 0
Received 0 Likes
on
0 Posts
Ha. You don't want me explaining that pilot stuff. As my username indicates, I turn wrenches on them. My stick wiggling is limited to only about 300hrs in 38 years or so. The real stick wigglers on this forum are the ones you need to inquire with. Some of the best around. However, suffice to say, it wasn't an "overspeeding governor" but rather an overspeeding engine without the benefit of governor control as the FCU went full open. Cracked line was the culprit. All I did was watch the gauges and tap him on the side of the head with my pilot stick when he got close to a limit. He flew it like the old days with the "back up" governor in his left hand. Excellent pilot, whom I miss.
