Bob La Londe wrote:

"Larry Jaques" wrote in message
...
On Wed, 1 Dec 2010 15:25:58 -0700, "Bob La Londe"
wrote:

"Ignoramus24652" wrote in message
om...
On 2010-12-01, Bob La Londe wrote:
"Ignoramus24652" wrote in message
...
My opinion only. Buying that Bridgeport Interact mill was NOT due to
my foresight, deep knowledge of CNC, or careful analysis of what
exactly I need. It was more of an element of dumb luck, combined with
an insight that a machine of this size is manageable, yet capable.

Most importantly, I was VERY attracted by the fact that the machine
did not seemingly need any physical retrofit (such as swapping lead
screws for ball screws and other such things). I am now of a very firm
opinion that doing physical retrofits is always suboptimal.

Yeah, and most folks don't get to buy a machine for $500 either.

It is not like I was privileged in any way. It was not an inside
deal. I saw the same ad on craigslist that everyone else saw.

So I had the same opportunity that was available to anybody.

I was not even very quick to buy it, it took me about 3 hours to
finalize all conversations with the seller and arrive at a virtual
handshake. If you recall, I was asking questions here as to whether to
go with it.

Then I paid same evening and took possession of manuals and tool
holders.

Heck anything other than full training by the factory service rep
who just installed your brand new $100K machine with tool changer
might be considered suboptimal by some folks. If I had gone that
way or anyway other than at the lowest end I would still be out in
the cold as far as making my own parts.

I think that if I was making money by machining, this is what I would
do, as well.

Knowing what I know now, I personally would stay away from any hobby
machines.

Fair enough, but how long did it take from the time that you bought
your
mill until you were CNC cutting parts?

I bought it in March around March 15. At the end of March, I picked it
up and delivered to my friend's warehouse. I got it to my house around
June 1.

On Jul 31 I made my first metal cuts (sprocket). So, it was 2 months.

I spent about 300-350 hours. If I billed someone for that at $50 per
hour, the cost of my time would be $17k. However, if I paid $17k to
get a working CNC mill, I would not have learned a lot of useful
things.

It's a good thing educational time isn't billable.


On balance, unlike making a microcontroller control for a welder, it
was worth it. The welder was the opposite, I spent a lot of time, and
a lot of money, I ended up with a machine that worked as designed,
reliably, but was inconvenient to use.

Suckage.


Let me first say I don't disagree with your reasoning one bit. Just
wanted
to point out that from another perspective for a different person another
machine might be a better starting choice depending on what they want to
do.
I think it was a better starting point for me. I have 100% confidence now
that I can take a non working machine and turn it into a working machine.

Congrats, Bob.


If I had started out with the Hurco two years ago though I suspect it
would
still be sitting under a tarp in the shop and I wouldn't have cut any
parts
in all that time. Instead I used the Taig to make parts, and even to make
parts to take another small non-working machine and turn it into a working
machine. I will probably use the two mini machines to make any parts I
need
to get the Hurco going too.

There ya go.


(I've even been playing with the idea of making mini machines for sale,
but
incorporating some of the things I have learned to make them better for
the
price.)

both ears perked up and faced forward for that

Yes, go on...

Well, some ideas you just have to go make some chips to learn what you
really need to do. Some ideas you need to wrap your mind around for a while
and then just go crank it out. Some ideas you need to do the 2nd first and
then the 1st 2nd before going back to the 2nd before getting back to the 1st
to make a real part. (Clear as mud right?) I'm somewhere in the middle.
Fortunately all the other projects and ideas help with the hard won
knowledge and experience of the chip making process. There are a number of
approaches and cost is a major factor.

For example its pretty hard to beat a preloaded ball screw nut with some
spring pressure on it. Ok... so just make it with top quality ball screws
right with antibacklash preloaded drive nuts? Not if you have a price
target. Top quality ball screws alone without the cost of the rest of the
parts of the machine cost as much as some machines. At the opposite end of
the spectrum are 60 deg V thread lead screws with simple brass (or bronze
for more money) pinch nuts. Cheap, effective, and somewhat accurate ones
can be had, but at the cost of long term life and accuracy and uneven wear
making it more complicated as your machine ages.

How about using a combination of things for a compromise. Delrin wears very
nicely, but it has a load coefficient that will destroy it pretty quickly
once you go over it... once. Backlash reduction... elimination... Backlash
nuts work, but many machines can apply more force than the springs can take
and when pushing hard you can lose your compensation, although repeatability
on the same cut will be decent. Beef up the spring? Then you are working
your motors harder too. Get bigger motors you say? Then your price goes
up.

I said was "playing" with the idea. LOL.



Quality rolled ballscrews with minimal backlash, and software backlash
compensation. Just keep the backlash to compensatable levels, and the
drive system up to performing the compensation. Don't forget the pitch
error screw mapping as well.

My little X2 mini-mill performs decently with it's stock ACME screws and
bronze nuts, driven by 600oz steppers at 30 IPM which is a decent feed
rate for a fractional HP spindle with a 2,500 RPM top speed. Even the
much crummier Z performs surprisingly well with a stepper driving it's
stock "fine feed". Certainly I could improve things, but I had about a
week to get the machine running for a project and it's done quite well.