Feed rate slow down

[gtdinc]

I am not a machinist or programmer, I am in the IT department in a support position, so I may not know the proper terrminology.
We have a Hitachi-Seiki mill with a Fanuc controller. It is an older machine and I don't know a lot of details. The operator puts 12 in as his feed rate, but when the machine is doing X-Y-Z moves in 3D milling the displayed feed rate flucuates and drops as low as 0.5. Is this normal operation of the machine or are there parameters or G codes to use to set the machine to use a constant feed rate?
Thanks for any help.

[1234567]

If high speed machining is turned on "g5.1 q1" it should slow down and speed up .
This is normal ..depending on what you are cutting this feed rate seems very slow for 3d.

[Reyad]

old machines have slow processing speed for g-code
so they don't read many gcode lines ahead

and to give you smooth surface finish it get's slow down

[sinha_nsit]

Whenever the path is approximated by a large number of tiny line segments, the effective feedrate would slow down due to inherent acceleration/deceleration involved in each segmental movement. Fanuc provides options to overcome this problem.

[angelw]

Originally Posted by gtdinc I am not a machinist or programmer, I am in the IT department in a support position, so I may not know the proper terrminology. We have a Hitachi-Seiki mill with a Fanuc controller. It is an older machine and I don't know a lot of details. The operator puts 12 in as his feed rate, but when the machine is doing X-Y-Z moves in 3D milling the displayed feed rate flucuates and drops as low as 0.5. Is this normal operation of the machine or are there parameters or G codes to use to set the machine to use a constant feed rate? Thanks for any help.

This is quite typical for controls that don't have high speed machining functions when machining true 3D profiles. 3D profiles are generally made up of many small linear moves to develop the shape, and its the smallness of the move that causes the the problem you're having. At the end of each motion block the sides come to a complete spot, albeit for a very brief time, then try to accelerate up to the programmed slide velocity. If there is insufficient length in the motion block to reach the target velocity, before having to start the deceleration ramp, then there will be a dramatic decrease in the actual feed rate achieved. The slides will accelerate to whatever velocity they can in the length before having to start the deceleration and that's all they will achieve.

Some controls have functions where the Acceleration/Deceleration is optimized, by the control looking at many blocks ahead. A few months ago I did a test on a machine where a total linear travel of 1000mm, was made up of many small G01 moves. I would have to have a look back over my notes, but I think each block had a linear move of 0.2mm. With the high speed function engaged the programmed feed rate of 2000mm/min was achieved. Without it, I think it struggled to get to 180mm or so.

Regards,

Bill

[angelw]

Sorry Sinha,
I didn't mean to cut across, I was typing when you posted your reply.

You may recall that the test I refer to is the one I carried out when we were discussing this same subject some time ago.

Regards,

Bill

[sinha_nsit]

Bill,
You always give detailed and better explanation. So, your post is always desirable.