Have you tried with the feedrate reduced by about 50%? If so, is it any better?

First, I'd reduce the stepover to 5% and see if it helps.
But to me, it looks like you're getting flex somewhere.
I know you said you have a solid machine, but to me, it doesn't look all that rigid.

Are the grooves always along the same axis? Are you doing a roughing pass first, or a single finishing pass?

Is this an actual 5 axis tool path ? ... or just 3 ?
Could it be related to the resolution of the rotary axis's ..... are they fine enough ?
Has this worked fine before ? or has this always been this way ?

Aspire doesn't do 5 axis, so I'm assuming it's a 3 axis toolpath.

because it is a 5 axis machine I did also check that the tool was running vertical etc.

 

Quote from: Alan B on July 08, 2013, 05:27:03 PM

because it is a 5 axis machine I did also check that the tool was running vertical etc.

 

Because he's using a ballnose bit, being vertical makes no difference as far as the grooves are concerned. The tool is the same shape regardless of the orientation. The only thing being out of vertical would cause would be a slight shift in the location of the carving.

Gerry, I was merely pointing out that was not using the 5th axis, that's all.........

Alan, If this is the same tool path as in the picture earlier, it seems much better after taking a very light second cut. If that is true I wonder if a third pass (very light) would get rid of all the tool marks.  If it does, I would have to agree with Gerry thinking it may be flex. With the bottom of the cut looking so good and only getting the tool marks as you get near vertical makes me wonder if the tool is chipped as RC asked. Dave has a very good suggestion as well about asking Vectrics. A video of the process would be worth a lot.

Brett      

Thanks
for all the response much appreciated, Gerry why do you say that the machine does not look rigid?? the frame of this machine weighs around 1/2 ton and is as solid as it could be and the sides and the gantry etc. are also very solid! just a bit curious about that, the tool path is a 3 axis was I can see now a bit confusing saying that it was a 5 axis machine, I do a roughing tool path and then a finish path and the finish path cuts at a depth of 0.5 mm (0.01968498 inch) the tool is new so there is no damage to that as for feed rates etc. I have been down to 75 mm (3 inch) per minute makes no difference as for step over have not been quite as low as 5% been down to 8% but will try going lower and see what happens, you make a comment about flex where might this flex be? I can run this machine at some 8 meters (26 feet) a minute and there is no flex/movement anywhere that I can see at that sort of speed not that I ever have or would try and machine anything anywhere near those sort of speeds what I am saying is that if I am not getting any flex/movement at those speeds then at the speeds that I use which for large areas of roughing I might at times go up to 1500 mm (59 inches) per minute and finish speed of 700 mm (27 inches) per minute then I certainly should not be getting any flex etc. anywhere, the gantry and all axis move as smoothly as can be no juddering or jumping around it never loses steps, so can you give me some idea as to where I should look for flex? the spindle motor again there is no vibration or run out it does not lack power (4 Kw - 5 HP) in fact it is way over the top for my needs really, anyway if all else fails I will do as you say and put it over on the Vectric site.
Thanks to everyone.
Alan

Try this, put a dial indicator on the table, adjust the dial to zero with contact on the cutter and then push on the side of the uprights.
I am betting you need gussets for rigidity. Looks a little weak to me from the photo. That's why you need to prove it to yourself.
If you were to try different materials with the same gcode program you would get different results of wavy lines. That would prove visually a rigidity or spindle issue. If they were all the same it would be a programming issue.

Edit: Also check with a prybar at the spindle mounts to see if you see movement on the dial, might be a backlash issue on the rotary. I can assure you you have something not quite up to snuff. But cool project thanks for showing it, have fun playing with it.

Craig,

Glad you brought this up. I have seen many say the same thing as you, but really give no tolerance for what is acceptable to a particular machine. On your(Craig) machine, if you put a indicator on any of the axis, are you saying it does not move at all? What resolution indicator do you use (tenths, thou, hundred or a metric equivalent ?) and what would you find unacceptable for your particular machine, and most important of all, what's it made of and how's it put together?

 I can put an indicator all over the place on my router, and move every axis if I really wanted too. Especially if I use a crowbar 

I'm being totally serious, and I'm not looking for an argument, it's just that I see this all time, but nobody ever gives acceptable examples for a wood cutting CNC router table. I personally think there are just too many variables being inserted into each home built table to come up with a standard.

For my own machine, I was a little concerned until I started cutting my first pieces, and then relieved to find that it's cutting to my satisfaction. I possibly should have started a new thread about this, but thought the topic was appropriate here as well.

Dave