Leopard tank gun stabilisation.

[David Merritt]

I see the correspondence under recoil system, and agree that inertia is a potential problem with the gun elevation mechanism, as the expected total weight of the gun with its counterbalance will be in the region of 10kg. The device providing the elevation control is of the screw jack design and when power to tbe drive is removed the gun should stop almost instantly, (Hopefully). The main problem is when the screw jack is moving one way under power and the stabilisation unit decides that the movement needs to be reversed, there will be a much higher (very short duration) current in the circuit, (as the polarity is reversed,) which has the potential to cause problems with the 2 amps limitations of the wires in the slip ring, the ECU or the motor. Time will tell if these problems exist and if the do then they can be overcome. It may be that the ECU will prevent this from happening to protect itself, which may cause a slight delay in the change in direction of elevation of the gun.
As I say time will tell.

David.

[Dennis Jones]

I think Stephen White spent a lot of time and money trying to do it on his Centurion with little success.

Dennis.

[Andre Meylan]

Well, I will try myself again and rebuild the gun using carbon materials and get it done below 500g. Servo and gyro system will easily handle that as from my experience with large turbine helicopters in the 25kg range, they can handle heavy weights. Tail rotors grasp easily 40% of the total system power and this is just a matter of counterweight in the turret I believe ...

Best regards, Andre

[Stephen White]

I applaud anyone who tries to solve this longstanding challenge but it’s got a number of issues. It would seem a logical solution to use RC aircraft gyros. You really need to measure absolute position, rate of change and have a feedback loop. The latter is what usually proves the biggest challenge. If the response times of the whole system (stabiliser and actuators) diverges from instantaneous to any degree, the barrel gets out of phase with the hull and becomes divergent. Bang - and not in a good way. So, there is a solution out there but no-one has really achieved a wholly satisfactory solution yet.

[Andre Meylan]

I applaud anyone who tries to solve this longstanding challenge but it’s got a number of issues. It would seem a logical solution to use RC aircraft gyros. You really need to measure absolute position, rate of change and have a feedback loop. The latter is what usually proves the biggest challenge. If the response times of the whole system (stabiliser and actuators) diverges from instantaneous to any degree, the barrel gets out of phase with the hull and becomes divergent. Bang - and not in a good way. So, there is a solution out there but no-one has really achieved a wholly satisfactory solution yet.

Dear Stephen - with the fullest respect on all attempts and experience made, please accept my foolish words above ... I might well see the challenges coming while getting along and trying to solve this issue

[David Merritt]

The use of a gyro stabilisation system for a aircraft to provide stabilisation to the gun of a tank is not a straight forward as it may appear as the stabilisation requirements are different.

The aircraft stabilisation system, when initiated, attempts to maintain the attitude of the aircraft when the gyros were activated, which are, in theory, the positions that the servos were at when the gyros were activated. The gyros uses the movement of the servos to maintain this initial set position. Therefore the gyros attempts to maintain their orientation in space by the movement of the servos and the set point for the servos, when the aircraft is in its initial set position, when the gyros were activated, is their neutral set positions. Therefore when the aircraft is in the position as set by the activation of the gyros the will be no output from the gyros to move the servos from this initial set point.

When the aircraft gyro stabilisation system is used to stabilise a gun in a tank there is no neutral set point for the servos as they are not using air to maintain the position but are moving the gun relative the turret which is moving relative to the ground the tank is passing over. Therefore there is a requirement for a continual varying output from the gyros to reposition the guns relative to the turret and the main tank body. To do this successfully there would be a requirement for gyros to be on the gun, within the turret and within the hull of the tank. (Which is how the gun stabilisation system works in reality).

A simple way to overcome this problem in the model and provide a reasonable representation of gun stabilisation is remove the servos from the setup and to use simple motors that can be varied in speed and rotate forward and reverse. As these motors will only move when power is applied and there position is not dependent upon the signal they are receiving they can quite happily elevate the gun and rotate the turret as requested by the gyros and will stay in the position requested by the gyros when the signal from the gyros returns to its neutral point when no stabilisation input is required.

The rotational motor for the turret and the screw jack for gun elevation as provided by Armortek should perform these tasks adequately. There may be a requirement to modify the speed of these devices or to restrict the speed of the tank to provide a better simulation of stabilisation, but only time will tell.

Sorry about the wordiness of this post and I hope my explanation has provide some level of clarification to this matter.

David

[Andre Meylan]

Thank you David - I still cannot see a problem having programmed more than 100 helicopters over the laste 15 years, using Futaba GY401, MicroBeastX, Bavarian Demon, VStabi and PowerBox Systems ... you have normal and AVCS mode. Simply define the middle and end points of the servo travel such as the gun shall never touch the body or anything else, rotating the turret 360°. Define the sensitivity of the gyro (low here around 20%). Mount the servo/gyro unit on the counterweight inside the turret and then start thinking of how to deal with the 3 wires from the gyro to the receiver ... so put one receiver in the turret and the other one for the motion and other functions into the body ??

You could even leave the gyro in the body but as soon as you change direction, I can see confusion ...

I might just give it a try and see what troubles I will get into ...

Best regards, Andre

[David Merritt]

Hi Andre,
What I am trying to say is,
If the gyro is initiated when the tank body is on horizontal ground and the gun barrel in horizontal this will be the position that the gyro will remember as the finish position for the servo when the gun is horizontal.
The tank then goes onto a slope of 10 degrees down,
The gyro will then move the servo to put the gun barrel back to the horizontal position by moving the servo to raise the gun barrel by 10 degrees.
When the gun barrel is again in the horizontal position the gyro will want to return the servo to the start position as the gyro is back in the start position.
There is now a conflict, the servo is in the correct position to hold the gun barrel in the horizontal position but is 10 degrees out of position as far as the gyro is concerned for its horizontal position.
The gyro tries to move the servo to where it was in the start position which it where the gyro thinks it is now that it is back in the horizontal position it started from, but each time it tries to move the servo the gyro goes back out of the horizontal position and finds it impossible to resolve the situation.

Regards
David.

[Adrian Harris]

Somewhere there's a video of the gun stab I did for the M3 using an Arduino, which was very basic and very slow, but proved the concept. That was just in the vertical axis though, as that's all they had in the M3 and M4. The faster you make the response, the better your PID has to be to prevent oscillation.

As far as I can recall, the 3DF board doesn't care where the body of the tank is, just that the barrel stays at the preset attitude.

In a tank, the gun might be facing sideways, with the tank pitching up and down longitudinally. Or turning side to side, so the turret has to turn to keep the gun aligned with the target. And the gun isn't just pointing along a heading, it's pointing at a fixed point in space, so as the tank moves sideways across the target, the turret has to correct to keep the barrel on the target.

It also needs to know the relative attitudes of both the gun and the hull, so that it can adjust the gun depression to prevent it impacting the hull.

Adrian.

[David Merritt]

Some progress has been made on sorting the power required to elevate the gun. With CV0704 and CV0704M in the position required by the gun recoil system the load require to elevate the gun is between 10 and 15kg, depending upon the angle of pull.

Based on this information I do not intend to use a counter balance for the gun barrel as it is impossible to remove all of the free movement from the gun elevating system so this imbalance of weight will help to limit the bounce in the gun caused by this free movement.
The current free movement in the elevation system will enable the end of the gun barrel to have an uncontrolled movement of around 15mm, which has to be acceptable due to the design of the system.

Only a hydraulic system stands a chance of eliminating this free movement and I don't intend to design one of those.

The rotational position of the gun turret with respect to the hull of the tank to enable the gun can be lifted from a negative angle of depression zero angle of depression with respect to the hull can be achieved by the use of a reed switch and magnetic tape, with the magnetic tape attached to the rotating part of the turret ring within the hull and the reed switch position to be activated by it as the turret rotates. The only requirement is that the turret position on the ring is identified so that it can be repositioned in the same position if it is removed.

There is a requirement to limit the elevation movement of the gun as the supplied screw jack has a extent of movement of 50mm but only around 25mm of movement is required to fully elevated the gun. The screw jack can exert around 100kg of force, so, if its extent of movement is not restricted it can easily cause some damage to the elevation system and burn out the motor. This is not a problem under manual control as the operator will restrict the movement, but under automatic control the stabilisation system will keep driving the system until it fails.

To this end, I have repositioned the fixed end of the screw jack so that it, when fully retracted, will define to fully elevated position. Thus I only need to define the zero elevation position and the -10deg position, again using magnets and reed switches. To stop the system depressing any further I will use the control system to switch in a diode at the required point so as to stop the depression of the gun any further but still allow it to be elevated.

I think it is important to set up this movement limiting control system before moving on to the gun stabilisation system as a lot of damage could bu caused without it in operation.

David.

[David Merritt]

Well progress is being made, all be it slowly.
I have sorted the setup for sensing the angular position of the gun and the rotational position of the Turret using reed switches and magnets.
The position of the reed switches and the magnets are proving a sensitivity of less than 1deg, which is more than acceptable.
I have yet to assemble and wire in the 6 relays required to control the extent of the movement, but their operation is sorted.
The normal elevation of the gun will be 0 to +20 degrees to prevent the possible control failure znd the gun barrel causing havoc with the rear of the tank.
It shall only be allowed to go to -10 degrees when the reed switch mounted within the hull detects that it is safe to do so, if for any reason this signal is lost the system will default to the 0 to +20 degrees operation.
There are three reed switches mounted within the turret, one for the -10 degrees position and two for the 0 degree position.
The reed switch in the -10 degree position shall switch in and out a diode that will prevent the gun depressing beyond the -10 degree position by preventing the current from flowing to the screw jack to depress the gun but enabling it to flow to elevate the gun.
One of the reed switches at the 0 degree position operates in the same way as the one at the -10 degree position by switching in and out a diode to control the extent extent of the gun depression.
The second reed switch will, if the gun is below the 0 degree position and the signal from the reed switch in the hull is lost or removed by the reed switch, it shall operate two relays that will power the gun up to the 0 degree position by bypassing the gun elevation control system and powering the screw jack directly until the 0 degree position is obtained at which time it will revert control of the elevation of the gun back to the normal control system.
The attached photos indicate what I have done.
The adjustment for the reed switches is made possible by using adjustable shock absorber mounts for 1/10 radio controlled cars.