Shark SD120

How are other systems alike/different from the Shark?

Moderators: al wolford, sbk, Bob, Kayvon, al wolford, sbk, Bob, Kayvon

Post Reply
Mad Max
Posts: 1
Joined: Sat Oct 16, 2021 2:34 pm

Shark SD120

Post by Mad Max »

I was a Tool and Die maker for 30 years , NO CNC machines
How accurate is this CNC?
For a 1.000 block, what is the plus or minus making multiple passes

Rando
Posts: 724
Joined: Tue Jan 06, 2015 3:24 pm
Location: Boise, ID
Contact:

Re: Shark SD120

Post by Rando »

Mad Max:

Depends on the material and F&S used.
Most people only do wood on these machines, and wood parts often only need to be accurate to 1/32" (0.03125").
That's even more true of things like wooden signs, where most people might not even notice 1/8"
Inlays, however, often want things a LOT tighter, but even then we're talking ±0.005", often more is tolerable.
And, of course, wood changes shape and size due to even humidity.

I'd say overall the machines are typically reliable to within ±0.010"
If done with a 0.005" finishing pass, potentially ±0.002-4"

However, even though the "resolution" of the machine is listed as 0.001"...
Unfortunately, the placement of those one-thou on-location points actually goes more like 0.0008", 0.0008", 0.0014".
Use a dial-test indicator attached to the Z-axis carriages and jog...you'll see what I mean.

I've done aluminum on an older HD-class NWA machine and managed to get it reasonably within 0.002" for round features, and square-to-axis edges sometimes within 0.001".

However, because the axes are not very rigid, the "number" of structural members involved in absorbing the cutting forces measurably affects the deflection.
So, the X (along the gantry width) will be off by the bit and spindle body deflection (z-axis carriage), any backlash in the X axis, and the side-to-side rigidity of the gantry arms. That's one amount.
The Y (along the bed length) will be off by the spindle body deflection, any backlash in the Y axis, and the length-wise rigidity of the gantry arms. That's a DIFFERENT amount.
And finally, the Z-axis is susceptible to flex of the Z-axis carriage and spindle mount, the backplate rigidity, and the x-axis rails (under the bed) rigidity. That's yet a third amount.

Which means that each axis has a different deflection-under-load "total Young's modulus" in that direction...which means that round holes will essentially **never** come out perfectly round unless done with
VERY low cutting forces. You'll never get boring-bar accuracy on these machines, sad to say. :cry:

Because the machines are made of a combination of steel, aluminum and HDPE, it's not like a fully cast hardened steel machine. Deflection is definitely an issue with getting accurate parts, and has to be carefully managed to get there.

If you don't already, get yourself a GOOD F&S calculator like the one from cnccookbook dot com. One that tells you the spindle HP, and deflection forces resulting from a cut in a particular material.

Well, that is if you actually **want** machinist-precise parts ;-). If you're just making cribbage boards, it's entirely possible this amount of accuracy will be more than adequate.

Regards,

Rando
=====================================================
ThomR.com Creative tools and photographic art
A proud member of the Pacific Northwest CNC Club (now on Facebook)

Post Reply