What is the technical definition of 'rapid'?

[Cole2534]

I fully understand what it does, but what is the actual definition of 'rapid'? What does 25%/50%/100% rapid really represent?

My observations suggest that it means the control will apply x% of max current to the motors to make them move with no (or very little) attention to accuracy.

Am I close?

 

[EmGo]

I fully understand what it does, but what is the actual definition of 'rapid'?

Faster than slow ?

What does 25%/50%/100% rapid really represent?

Using "rapid" avoids having to put the feedrate into every block where you want to just move the thing, don't care so much about accuracy etc etc. Mostly it just saves typing.

Whatever the "rapid" on a machine is set to, the % overide will be a fraction of that feedrate.

My observations suggest that it means the control will apply x% of max current to the motors to make them move with no (or very little) attention to accuracy.

Two different common ways to do it - one is just max out the axes, that will create a path that is not what you'd think. The other is to interpolate the move so the path is a straight line where the longest move will go at the rapid rate, the others slower.

Version two does what you expect, version one can be a surprise.

The G00 G01 on a K&T were actually a little different even at the same feedrate - G00 would ensure that the slides were in position before going to the next block, for doing things like hole location. G01 was just a linear move which could head off on the next block before fully arriving due to following error, look-ahead, all that stuff. So rtfm is a good idea. Controls aren't all the same.

 

[sinha]

the control will apply x% of max current to the motors to make them move with no (or very little) attention to accuracy.

Some hardware expert may please explain what the control does to achieve, say, 50% rapid rate. Since too much accuracy is not needed, the method is likely to be different from the G01 feedrates.

 

[???]

My observations suggest that it means the control will apply x% of max current to the motors to make them move with no (or very little) attention to accuracy.

Incorrect, one of the points touted by Kira regarding the rapid moves on the Tap Mill centers was the positional accuracy at speed. I once came across a Canadian company that extolled the virtues of their Kira with regards to its positional accuracy.

The Kira has no rapid override button on the rapids either.

 

[Mtndew]

Vroom Vroom mode.

 

[D Nelson]

I fully understand what it does, but what is the actual definition of 'rapid'? What does 25%/50%/100% rapid really represent?

My observations suggest that it means the control will apply x% of max current to the motors to make them move with no (or very little) attention to accuracy.

Am I close?

I think it means the code has turned over the movement of the work piece to the machine control then it waits for the control to say I have arrived at the position here is the control of the code do what you want while it waits for the next rapid signal
Don

 

[Vancbiker]

I fully understand what it does, but what is the actual definition of 'rapid'? What does 25%/50%/100% rapid really represent?

My observations suggest that it means the control will apply x% of max current to the motors to make them move with no (or very little) attention to accuracy.

Am I close?

Depending on control make and model and parameter and servo configuration, how rapid is accomplished and behaves will differ.

I have a pretty fair understanding of how many of the older Fanuc models from 6 series to 16/18 series operate and could do up an explanation for a control in that range.

 

[706jim]

Each servo turns as fast as it can for the required distance without regard for what the other servos on the machine are doing.

More or less.....

 

[???]

Each servo turns as fast as it can for the required distance without regard for what the other servos on the machine are doing.

More or less.....

Depends on how you set up the rapids. There are generally two modes, full feed on all and then a single axis move or all arrive at the destination at the same time. First one is sometimes referred to as dog leg move. Interesting results when the CAM is set up for a different one to the machine and the cutter is zooming around features. Generally proves that two solids cannot occupy the same space.

 

[Houdini]

Dog leg is bad!

 

[Houdini]

Looks scary when you been watching standard clearance rapids, then dude programs shortest distance arc rapids WTF

 

[706jim]

"Dog leg is bad".

True, but for whatever reason is standard on many machines.

I ran a Makino VMC with a Fanuc 7m (ever hear of one of those?) that had linear rapid, yet later machines dog leg only.

 

[CarbideBob]

Talking about max motor current as rapid will be bit misleading a bit here.
In fact at rapid speed and moving the motor current will not be near max. Normally much less power than when cutting and much lower speeds.
Rapid speed has to do with the drive,motor,pulley/gearbox, and things like screw critical speed.
For any arrangement of the above there is a max speed possible. Inside this a tad one establishes a "rapid speed".
At this speed the servo errors are larger that what one would want on feed moves but allowed.
Whatever this rapid number in IPM in rapid override you get 10,50.75% of this just like a feedrates.
Difference in a G00 and G01 show up most when one axis faster than the other.

 

[Cole2534]

Started this thread and forgot about it. Thanks for all the replies.

It seems that maybe 1 way to sum it up is that the servos are moving as fast as possible without any regard for following error.

And yes, dog-leg vs linear can be spooky.

 

[EmGo]

It seems that maybe 1 way to sum it up is that the servos are moving as fast as possible without any regard for following error.

No no no, they all ... hang on, all the ones I've owned, who knows what the four-eyed japs do - have a max following error. Exceed that and you will get a drive error. At least on the stuff I've owned, if the rapid was, for example, 400 ipm, then the drives were adjusted for that and the control output commands to go that fast, not just whatever the heck it could manage.

 

[Vancbiker]

All motion has following error. How it is expressed varies between control builders and models.

 

[Donkey Hotey]

It seems that maybe 1 way to sum it up is that the servos are moving as fast as possible without any regard for following error.

And yes, dog-leg vs linear can be spooky.

Depends on brand. That farthest back that I can confirm is my 2004 Haas VF-2. Rapid is a parameter and can be changed. Mine is 650 IPM but the cabinet says 1000 IPM. I changed it once to the higher value and it did indeed speed it up. I was warned to put it back because that voided the warranty. Not sure why the cabinet says 1000 IPM then but, whatever.

I don't know why people think the machine can have a following error. It has to at very least know where the Z-channel pulses are firing or it would lose position and have to home itself again.

 

[Northman Ind.]

Rapid, is a max travel speed for given drive motors, the percentage cut, is just that an approximate percentage of full speed, as for accuracy, location accuracy doesn't seem to be effected on modern machines, I wouldn't get out a stop watch and start tracking lap times though, as each motor is going to have slight variables with given voltages and fluctuations, older machines would slow down just before reaching destination to maintain accuracy, and to reduce shock load to ball screws... (I wanna say there is an M Code that can turn this on and off too... just haven't seen it used in decades)

I have ran machines that were more accurate when ran at slower rapid and for that matter feed rates, and it has more to do with age and condition than type or brand. Fadals have a 'positive' approach M code that is pretty skookum for holding stupid tight tolerance hole locations (m79? I think? again its been 15 years since I've needed to run one) Did something similar with Haas, and Bridgeport VMC's (<junk) but that is closer to 25 years ago or more lol, and it might of been editing code to "create" a positive approach.

 

[EmGo]

I don't know why people think the machine can have a following error. It has to at very least know where the Z-channel pulses are firing or it would lose position and have to home itself again.

Servos have to have following error. That's how they work. No following error, they would just sit still.

Abbreviated explanation : Command a position. Difference between commanded position and actual position (following error) creates command to move. When the position the slide is at equals position that's commanded, then movement command stops and slide movement stops.

Grab a ball screw and turn it, you can feel the distance between where it's supposed to be (commanded position) and the distance you twisted it to (actual position) create a reaction from the drive because of the following error. That's what makes a servo a servo.

Never tried to see on a fauc but on bendix there's a display screen, that's how you adjust the drives. Set display to following error, command a move at a particular speed, adjust the drive to give the required error. Faster feedrates, larger error. Rapids are just the max feedrate, they aren't an uncontrolled "here ya go buddy, have a 20 volt signal !" speed.

 

[Northman Ind.]

Servos have to have following error. That's how they work. No following error, they would just sit still.

Abbreviated explanation : Command a position. Difference between commanded position and actual position (following error) creates command to move. When the position the slide is at equals position that's commanded, then movement command stops and slide movement stops.

Grab a ball screw and turn it, you can feel the distance between where it's supposed to be (commanded position) and the distance you twisted it to (actual position) create a reaction from the drive because of the following error. That's what makes a servo a servo.

Never tried to see on a fauc but on bendix there's a display screen, that's how you adjust the drives. Set display to following error, command a move at a particular speed, adjust the drive to give the required error. Faster feedrates, larger error. Rapids are just the max feedrate, they aren't an uncontrolled "here ya go buddy, have a 20 volt signal !" speed.

I had an argument with someone recently about balls screws and back lash... its really kind of surprising how much back lash they can get in them...

that aside, anything with balls screws that I've played with has had a way to adjust them, usually hiding in Parameters/settings somewheres, bendix/baldor/fanuc/mitsubishi the CNC mechanics get all cranky when you figure out how to adjust them yourself.