[glomagno]

I have a 5 axis breakout board that directly controls stepper motors with out additional drivers. I would like to use an air cylinder to move in and out at the same time the motor turns cw or ccw. The connections on the board are pretty straight foward. A+A- B+B- connections. There is constant voltage going to the stepper for the holding torque. can this device be daisy chained in to move when the stepper moves?

http://enfieldtech.com/wp-content/up...262012_300.pdf

This air cylinder is to assist a stepper motor turning a robotic arm. so when the motor turns cw the air cylinder pushes up and when the motor turns ccw the air cylinder pushes down.

[vtxstar]

That device, a servo air-drive, may be overkill for your application.

Do you require variable locations for the air driven arm or just stop-stop locations. What tooling is on the air driven arm?

The motor drive pulses will only change position relative to each other depending on direction. Their mean values will be the same regardless of direction.

Better is to use the direction signal for the motor driver. This is two states and can drive a two position circuit... if a stop-stop condition works for your application.

More information is needed.

-=Doug

[rockcliff]

I have a 5 axis breakout board that directly controls stepper motors with out additional drivers. I would like to use an air cylinder to move in and out at the same time the motor turns cw or ccw. The connections on the board are pretty straight foward. A+A- B+B- connections. There is constant voltage going to the stepper for the holding torque. can this device be daisy chained in to move when the stepper moves?

http://enfieldtech.com/wp-content/up...262012_300.pdf

This air cylinder is to assist a stepper motor turning a robotic arm. so when the motor turns cw the air cylinder pushes up and when the motor turns ccw the air cylinder pushes down.

It would be possible to use the Servo Pneumatic Proportional Control Valve, But it could not be controlled from a stepper driver, (motor outputs) these servo valves need a 0-10v signal input, the signal output would need to be tied into your mechanics of the robot arm (example: when the arm is fully extended a 10 volt signal would have to be sent to the valve, as the arm retracts the voltage would need lower in increments to 0 ) as well as feedback position control, some sort of potentiometer, linear on the air cylinder or a rotary one on the joint.


From past experience in using these Proportional Control Valves, they can be a nightmare to setup to run correctly, and really should only be used when you have limited options like cylinder position control in flammable areas.

On industrial robots, like Fanuc, ABB etc. they use counterweights to help with balance.

[Al_The_Man]

The first time I have seen a Pneumatic servo valve, and I have used a few hydraulic with satisfaction.
But I would be leery of using a pneumatic due to the compressibility issue?
If I used air at all I would be inclined to use air over oil?
Back in the 80's, Bosch built a CNC machine with all Hyd cylinders in place of servo motors.
I am not sure if they are still around, but there was a manuf. that used a Hydraulic cylinder for positioning that had a stepper motor on the end controlling the spool directly, fed with a step/dir signal. Feedback was need though.
Al.

[rockcliff]

The first time I have seen a Pneumatic servo valve, and I have used a few hydraulic with satisfaction.
But I would be leery of using a pneumatic due to the compressibility issue?
If I used air at all I would be inclined to use air over oil?
Back in the 80's, Bosch built a CNC machine with all Hyd cylinders in place of servo motors.
I am not sure if they are still around, but there was a manuf. that used a Hydraulic cylinder for positioning that had a stepper motor on the end controlling the spool directly, fed with a step/dir signal. Feedback was need though.
Al.

Yes, your absolutely correct, compressibility of air, is very problematic, if the load / friction changes, from one position to the next, we found these very difficult to setup (hair pulling, head banging against the wall) , overshoot, undershoot, and wild oscillations, are common issues, + the cost of the system

[glomagno]

sorry i have been very busy and forgot about this post but i came up with a few things for my setup.

MM1 OEM Pressure Control Valve | Pressure Control Valves | Proportion-Air

Dual Output Contactless Rotary Sensor | Penny+Giles

I want the air cylinder to assist in the arm movement. the stepper motor i will be using does not have the holding torque needed to hold the weight of all the connected arms. i wanted to create a stable control.

Here is what i am building

[Al_The_Man]

I still feel pneumatics will be problematic?
Did you look at addressing the problem by gearing the motor and/or larger motor in order for the motor to be able to achieve this on its own, rather than trying to aid the low torque by adding an extra element?
Al.

[glomagno]

im doubting by going to a 1600ozin motor will due the job and a gear box with a backlash of 7arc/min will wear out to fast and deny the repeatability i need. i figured with air cylinders and position control i wont have to worry about replacing parts.

the motor im using is a 890oz stepper for the lower arm. the weight of all the arms together is about 35 pounds with all the motors connected. A 1600-1800 with a gear reduction SHOULD do the trick but. I dont have much experience with gear reduction boxes and the longetivity of its life

[Al_The_Man]

The problem as I see it with both assist solutions is that you are introducing a PID loop in parallel with another control loop, in this case open loop stepper, but it would be the same if the present loop were PID also, IMO it will be a challenge to perfectly match the two loops exactly?
This is typically done with a closed loop servo system, where the parallel slave axis, is geared by virtue of feedback to the primary master axis to move in unison.
In your case the slave would be the pneumatic servo valve.
Al.

[glomagno]

in another forum post i made i had asked if i could use the motion controller board from hvlabs.com to control stepper motors with drivers.

here is the link of the board they have created for the motion control
Im not an electronics expert but i have seen similar wiring setup for certain mach3 specific boards. HVLabs.com

what do you think al?

[rockcliff]

I agree with Al, servo valves / cylinders are possible, but would introduce a whole new host of challenges, beyond your initial issue, but there is no harm in trying at least one axis,

Even thou most industrial robots use harmonic drives,
You should really take a look at the FANUC M-2000iA heavy lifting Robot for some design inspiration, they also use a few counter weights on the back, to help assist in lifting, see,,,,, M-2000iA Car Body Transfer Robot & M-20iA Sealer Robot - FANUC Robotics Industrial Automation - YouTube


and....... [ame="http://www.youtube.com/watch?v=8St-COkyAEs"]The biggest and strongest robot of the world: Fanuc M - 2000iA 1200kg - YouTube[/ame]

[glomagno]

Ive worked with fanuc welding robots at my last job and i found that inside the joints of the bottom arm is a slew bearing with a built in planetary gear drive. no weight balancing. only balance is on the upper arm like i have designed on mine. the motor is mounted behind the joint.

[vger]

If you are just needing to support some of the weight, you might try one of those gas shocks of the type used to lift tailgates on cars. I believe they are available in various lengths and forces.

The questions in my mind are...

If this is a rotary joint, how are you planning on converting the linear motion to rotary force?

Will the force required change with the position?

On the servo pneumatic drive.... with 0-10 v drive, suppose your driver outputs 5 v (middle position) and the stepper also goes to middle position. If the stepper is off even slightly, the feedback in the pneumatic system will see the error and apply pressure to force the arm to middle position.