Re: Electronic Lead Screw & Power Cross Feed | Poor Man's CNC ?

Bob:

You've done an excellent review of the entire issue. Full CNC is
clearly far more complex than electronic lead screw - and requires far
more learning time. CNC leads to a different lathe machine altogether
- where "machining" is replaced by "programming" - even for simple
one-off pieces. I understand - once converted to CNC - a "CNC lathe"
cannot be a good "hands-on" manual lathe - except maybe with expensive
electronic handwheels.

My thinking so far is to do a CNC conversion of my Super X3 Mill -
because, in a mill, full CNC has overwhelming advantages. In a simple
bench lathe, however, electronic lead screw is all I really need - for
control of threading ratios and traverse speeds - with perhaps an
extension to electronic cross feed.

Full CNC does have the large advantage - as you point out - that much
of the R&D work has already been done. Would a few people be
interested in setting up a Yahoo Group devoted exclusively to
developing a generic and adaptable Electronic Lead Screw design - with
perhaps SIEG's C6B as a test bed ? Once developed as an "Open Source"
design, electronic lead screw implementations should be very cheap
compared to full CNC.


---"BobWarfield" <bob@t...> wrote:

> I think the CNC setup is actually considerably more complex than
> what is contemplated in the strictest sense of an electronic
> leadscrew and I'll tell you why. It all sounds like a case of "Gee,
> if I have to hook up a stepping motor anyway, why not just go CNC?"
> However, let's look at a couple of different reasons why that is not
> really the case.
>
> First, we have to make sure we are doing an apples to apples
> comparison. It is certainly true that there are no electronic
> leadscrews to be bought off the shelf today unless you want to count
> the short-lived Frog CNC. I will therefore readily admit that the
> development of an e-leadscrew from scratch will certainly be more
> complex than to piece together a CNC system from off-the-shelf
> components. So let's assume the e-leadscrew exists. Not so far
> fetched, BTW. As was pointed out, there is the Frog CNC and also
> this conversation keeps coming up on all sorts of boards. Clearly
> there is demand or interest, and someone is going to get around to
> building one. Heck I heard Cletus Berkeley say he was going to look
> into it over on another board, and there is a fellow on the Home
> Shop Machinist board that has built an electronic gearbox for a gear
> hobbing machine, so someone WILL do it!
>
> So what would this new animal be like? Well, it's going to be very
> simple because it doesn't have to do much. Again, if you have ever
> seen a Frog CNC, there is nothing to it. It's a tiny little circuit
> card with a membrane keyboard and LED readout that's built around a
> PIC controller. It's got a stepper driver on board and takes a
> spindle encoder signal in.
>
> Let's imagine something very similar to the Frog CNC will work. In
> fact, I think it will be very similar to the Shumatech DRO in terms
> of the complexities involved (hence a thread about this on that
> board too). Lots of folks manage to get one of those DRO's up and
> running for around $150 plus the cost of the scales. Picture
> something similar here. You order the blank PC card from a
> Shumatech-like entity, buy your parts to populate it, solder it
> together, and $150 plus a little time later, you are ready to plug
> it into your spindle encoder and leadscrew stepper motor you got on
> eBay.
>
> Which brings me to my second point. If that solution existed today,
> it's a lot simpler than full CNC. If I can really build the thing
> for $150, buy a surplus stepper from eBay, and hook together a few
> parts for a spindle encoder, it is hard for me to regard that as
> even approaching the complexity of CNC.
>
> Let's just break it down. I have to have a PC for CNC. What if I
> don't have a spare sitting around? Even a cheap obsolete PC costs
> as much as the DRO. I have to configure it properly for my
> software. TurboCNC doesn't want to run on Windows XP, so I gotta go
> deal with that. Mach will run on XP, but the software costs as much
> as my PIC controller for the leadscrew and its just one piece of the
> solution. I need a Gecko drive, which is another $100. It's very
> possible my PIC-based project can deal with driving one stepper
> right on the card for its $150 cost as we already saw with the Frog
> CNC device. What about backlash? The CNC crowd certainly spends a
> lot of time talking about it and going on about ballscrews to avoid
> it or various other solutions. Ballscrews are big bucks. The e-
> leadscrew doesn't care about backlash because the operator deals
> with it as a matter of course. How about a breakout card so my PC
> can talk to my Gecko drive? Campbell designs makes a great one, and
> oh by the way it too costs as much as a Shumatech DRO. Maybe I'd
> like a nice pendant, because I really don't care to type on the
> keyboard and click the mouse, I want to work on my lathe for crying
> out loud! www.lowcostcncretrofits.com has one that plugs into my
> USB port (hope I didn't buy an old 486 machine to save money!).
> Guess what, it too costs about what that darned Shumatech DRO would!
>
> Now just about every component in the CNC system has cost as much as
> the e-leadscrew electronics, each and every one of them has a
> learning curve before I can get it to work at all, and we haven't
> even gotten on to the real learning curve associated with the fact
> that your machinist may know full well how to make parts with his
> lathe but not know the first thing about CNC, or care to learn.
> He's just tired of digging around in the swarf to find the right set
> of gears to cut a thread he needs or change the feed speed to get
> rid of some chatter.
>
> It's all fine and well to argue CNC is better, but its really better
> at solving a completely different problem and it is far from being
> simpler or "you might as well go for CNC if you are going to go to
> all this trouble anyway."
>
> BTW, the clever thing about the Frog CNC is you can use it as an
> electronic leadscrew AND as a breakout box + stepper driver. This
> way you get both CNC if you hook up all the extra components and a
> vastly improved manual lathe if you want to focus on one-off parts.
> That too would be a nice attribute for the e-leadscrew when it is
> available.
> BW
>
> ---"Paul Kelly" <tech@c...> wrote:
>> This thread crops up with some regularity.
>> Yep, you could make a setup, as described, work. Or you could
>> add one more motor. And use simpler, off the shelf electronics
>> (or simpler, 'you build em', electronics) and have a CNC setup.
>> Typically the proponents of the electronic gearing systems feel
>> that, because you can do so much more with it, a CNC setup must be
>> more complex. In fact, this is not the case. Vaso alludes to the
>> reality of the situation. The issue is mostly about stepping out
>> of your comfort zone and learning a new thing...
>> PK
>>
>> -----Original Message-----
>> From: On Behalf Of Vaso
>> Sent: Sunday, 4 December 2005 5:54 PM
>> Subject: Re: Electronic Lead Screw & Power Cross Feed |
>> Poor Man's CNC ?
>>
>> I've done machining for several years, but I'm new to CNC. I'm an
>> electronics engineer, so an electronic approach to gearing -
>> rather than mechanical gears - appeals. This PLL (Phase Lock
>> Loop)method is well-known in electronics.
>>
>> Maybe electronic gearing could be called the "Poor Man's CNC". I'm
>> less interested in full programmable CNC right now than I am in
>> correcting some deficiencies in the SIEG C6B lathe I just bought.
>> The C6B has variable spindle speed, but fixed saddle traverse - so
>> feed rate is either 0.1mm/rev or 0.2mm/rev and switching between
>> the two is awkward. There is no Quick Change Gear Box, so
>> threading requires manual gear substitutions. There is no Power
>> Cross Feed, and there is no taper attachment. Seems to me PLL
>> control of the lead screw, as described in this article, with the
>> enhancement of microprocessor control of speed ratios, could give
>> practically unlimited threading and traverse speeds. Power cross
>> feed could also be arranged by this method. If Power Cross feed
>> linked to lead screw speed, then this is effectively an electronic
>> taper attachment. Several tricks short of full CNC could be
>> microprocessor-controlled.
>>
>> Am I missing something ?