Comments by "ke6gwf - Ben Blackburn" (@ke6gwf) on "" video.
-
As a hydraulics guy, I just want point out the bits that I recognize lol
On upper side of the upper blue actuator cylinder, you can see a silver rectangle, with a smaller silver rectangle one above it that has white wires coming out.
The larger one will be the 3 position spool valve (extend/hold/retract) with the pressure and return lines coming into it, and the smaller one is the wiring junction box.
It will then either have hard lines running from the valve to each end of the cylinder, or it may have bored channels inside the walls of the cylinder.
Then at the rear end of the cylinders, you can see another box with wires coming out, and that is going to be the linear positioning sensor that tells the computers exactly how far out the cylinder is extended, thus what angle the grid fin is at.
Judging by its position, I would guess that the center of the piston and rod is bored out, and a skinny tube runs down the center of the bore from the rear end of the cylinder. Inside the tube would be magnetic sensors, and there would be a magnet imbedded in the piston, allowing the exact position to be detected.
511
-
4
-
@engineer9673 I went back and looked at the pictures in high resolution, and I see no sign that any of the cylinders were extended beyond their limits, in fact the cylinder on the left is extended even further than the one on the bottom, and you can see that the left grid fin is within normal range of motion, so there is no reason to suspect that the cylinders were damaged.
It is also rare for a piston to break off a rod, since the entire system is designed to be stronger than the maximum forces applied, and the grid fins don't face very high forces.
I also suspect that the cylinders are also the mechanical limits of the grid fin travel, since the greatest force they will see is from the hydraulic cylinder action, and so there is no need for any other stops.
Generally hydraulic systems use the limits of the cylinder for the stops, unless there is some reason for another stop, but for a rotary device like this, you just make your cylinder the correct length for the desired throw, otherwise you are putting full hydraulic pressure against the linkages, and have to make them stronger to withstand it.
By using the cylinder as your limit, you ensure that the greatest force applied to the system is the aerodynamic forces created by the grid fin.
3
-
@sunside79334 , lightweight linear position sensor hydraulic cylinders are a very common off-the-shelf product, as are the linear sensor components for inclusion in a custom cylinder design.
It is one of the most preferred means to measure position, because it is completely protected from external interference or damage, unlike rotary shaft encoders or external scales.
Since this is being controlled by a computer, all you have to do is tell the computer what angle of grid fin equals what cylinder position.
If you did a rotary encoder on the shaft, you have to find space for it, add complexity to the shaft, either a disk mounted to the shaft, gears to operate a parallel axis rotary encoder, or try to fit 4 of them in the center between the end bearings and the shafts and the stage seperation cylinder, protect it from damage (it would need to be a high precision encoder, which means very tiny slits which are susceptible to debris clogging them leading to an incorrect reading), and adding additional components to the system.
Then you would have to put that data into the computer, which would have to count pulses and then calculate rotary position.
The linear approach prevents calibration errors or position error, because instead of counting clicks, it can tell what part of the sensor the piston is at.
No, having it built in to an existing component that very possibly is an off the shelf part, where it is protected from space and heat and debris is a very smart thing.
Also, if they wanted pressure sensors, they would need to have 2, one for each end of the cylinder since it is a double acting cylinder.
These would probably be mounted flanking the valve if they needed them, but they aren't dealing with high forces here, so they don't really need pressure sensing.
As for the plumbing, it is really simple.
On the lower left side, 2 metal lines come from the hydraulic pump. One is Pressure and one is Return. (the smaller one is Pressure)
They connect to 2 square manifold loops, a larger and a smaller, and then there are 4 sets of red hydraulic hoses coming off the 4 sides of the manifolds to the valves, which are mounted on the cylinders.
The reason for the square loop manifolds took me a second to figure out, but it is to smooth hydraulic flow and prevent pressure drops.
Each valve and cylinder is separately controlled, and as they cycle, the hydraulic flow needs vary widely, but since they tend to move in opposed pairs, with one extending while the other retracts, and it takes a lot more fluid to extend a cylinder then to retract it, by allowing fluid to go either direction around the loop, it allows fluid to bypass lower flow cylinders to get to the higher flow cylinders.
If it only went 3/4 of the way around, then the cylinder on the far end would be starved if 2 cylinders before it were in high flow demands.
So by adding a few inches of pipe and closing the loop, they were able to reduce the size of the manifold pipe, and avoid pressure drop issues. I call that genius!
There is no need to try to hydraulically synchronize cylinder pairs, because that is done through positive position feedback through the computer, and since it is flying the rocket with the fins, it does not always want them synchronized.
If they were hydraulically synchronized, you would see lines going directly between the cylinders.
So, I am also quite impressed with this design, because it is simple, and uses standard parts and systems that are well understood and reliable, except maybe for the hydraulic pump, which is probably some special helium driven gas over hydraulic unit, which always have problems lol
2
-
@firesculpturevideo a servo hydraulic valve is a specific part. It is designed so that the valve can be opened part way, rather than the normal valves that are either all the way open or all the way shut.
It is normally only used on larger valves, but it is used when the continuous speed of operation of a cylinder or motor is needed, rather than just position.
So to position grid fins, you don't need smooth gentle variable speed motion, you can pulse the valve to get the fin to the needed position, and so the valve just needs to be on or off.
If you are dealing with a larger device with a lot of inertia, then you might need a servo valve, so you can command it to 20 percent to start the load moving, then to 50 percent, then to 80 percent, and finally fully open, etc.
The servo in the name refers to the valve operating mechanism being something other than on or off.
I have worked with these types of valves using either direct magnetic core designs, or rotary motors operating the valve.
Now maybe in some fields some people refer to position controlled hydraulics as "servos" because it reminds them of electric servos from airplanes or radio controllers, but that isn't the normal name for having motion controlled hydraulics, because other than occasionally using a servo valve, it isn't really the same thing as a servo.
Oh, and the reason servo hydraulic valves are not very common is because they generate lots of heat, and heat kills hydraulics.
Normally if you need to control the speed of a cylinder, you use some form of pressure control or variable output pump, etc.
Anything that uses friction to control flow is like using the brakes to control the speed on your car, instead of taking your foot off the gas. Not a good idea, unless it is only occasionally needed lol
2
-
1
-
@anuragshinde9208 , I think you are talking about the small extended cylinder that you can see at the top mount of the landing legs.
That looks like a pneumatic cylinder, operated by helium, that kicks the legs open by pushing them away from the rocket body. Once they are beyond its reach, I suspect that gravity and inertia does the rest, since they extend during the landing burn with the rocket rapidly decelerating, so I would not be surprised if they actually have to be slowed down in their extension, rather than powered.
But there is no mechanism built in to retract the legs, and none needed at this point, since it always lands on the ground to be serviced by ground crews.
Now, the future designs will need self retract capabilities, but only on missions where they will be landing on the moon or Mars, and then take off again.
For ideas like the BFR being used to from city to city, I suspect that they will have a launch gantry, so once it lands, it will be captured, set on the launch lugs, and the legs folded up by ground support equipment, since there is no need to add the weight and complexity of a retract mechanism.
1
-
1
-
@engineer9673 as I said, the cylinder on the left, which you can't see in this video, is extended further than the one on the bottom, and its fin is within normal operating range, and no broken mounts.
Considering how few cycles these cylinders operate, just a few minutes of operation really, you are not going to see wear and scoring on the cylinder showing limits of operation, so you may be seeing soot deposits from the landing burn or just shadows, but it isn't scoring and wear after only a few minutes of operation. Also, most of the operation time on these cylinders will be in testing, and they would be very clean, and stopping at many different positions, and rarely going to full extension, because it is just one out of many positions to test.
And remember, this system dumps its fluid as it uses it, so they aren't going to be testing it for long periods, and then having to refill every few minutes. So landing is probably the most movement it gets, and it normally would never go to full extension, because the rocket should not be that far off course.
The high resolution pictures, including from straight on where you can see the left cylinder are on John Krauss's Twitter feed if I remember correctly, or maybe it was Instagram.
Look him up and see them all. He's worth looking up anyway lol
1