Messages

701

General Mach Discussion / Re: Set Steps Per Unit function not accurate

« on: July 18, 2014, 05:46:19 AM »

Sounds to me as though you have some serious backlash somewhere. That will always produce those results. And no, making sense out of it before you deal with the backlash is probably not possible.

Cheers

702

General Mach Discussion / Re: Spindle speed crossover eliminating

« on: July 18, 2014, 05:43:05 AM »

It sounds as though you don't have a sync signal coming back from the Lathe to mach, or maybe you just haven't enabled the sensing of it. Does threading work for you?

Cheers

703

General Mach Discussion / Re: How to correctly ground very noisy spinddle

« on: July 09, 2014, 12:09:56 AM »

The bottom line in all of this is that power supplies are not the simple beasts they are so often made out to be. Without being able to look at Adrian's system with a CRO, it is very hard.

I do remember listening to a conference presentation on a 5 V 600 A PS once - it was for a rather large computer (and no, there are no typos there). The complexity needed in the design was ... iinteresting!

Cheers
Roger

704

General Mach Discussion / Re: How to correctly ground very noisy spinddle

« on: July 08, 2014, 05:18:03 PM »

Since the supply(s) are already rectified within each series supply, you're not dealing with the spikes that occur from the bridge mains rectifiers -  so a 200VDC electrolytic filter capacitor will be just fine at the 90 DC volts

All of that would be correct if the laptop supplies are simple linear jobs. But since there seems to be a of of RF around, I question that assumption. My experience has been that in order to get an input rating of 100 - 250 VAC, which most of them have these days, they use a switch-mode supply (SMPS). I have a tiny charger for a camera battery from eBay which cost me just $5.05, and it too has an SMPS in it. Dunno how they do it for the price.

Cheers
Roger

705

General Mach Discussion / Re: How to correctly ground very noisy spinddle

« on: July 08, 2014, 08:44:04 AM »

What would be the correct value for choke? I already tried a 200V electrolytic capacitor without choke, and it got hot within 1 minute

Ah, well that is a good question! But not a simple one.

I think you need a choke, an inductance, rated for a bit more current than your motor draws, in one of the wires to the motor, plus a mains-rated film capacitor across the motor. The cap should be rated to something like 630 VAC, and will only be maybe 0.1 uF. Not expensive stuff. An alternative would be to buy an IEC filtered mains socket, but then you would need an IEC plug as well. Not a silly idea, mind. You would (have to) earth the earth pin on the socket.

The 200 V cap was simply the wrong thing to try. It sounds right, but it could not handle the RF spikes.

A caution: do not try using an 'ordinary' capacitor. It would probably fail fairly soon. By the sounds of things, this is a harsh environment!

Now, this is quite different from putting in a large filter capacitor. But try it and see how it goes. If you need to add a large electrolytic filter cap as well, you will probably need a rating of about 400 VDC at least.

I doubt you need a bleed resistor across the 0.1 uF capacitor: it's very small, but it won't hurt.

Cheers

706

General Mach Discussion / Re: How to correctly ground very noisy spinddle

« on: July 05, 2014, 08:12:58 AM »

Um - OK.

I suspect, from that info, that the PS you had been sold with the spindle, was a VERY crude switch-mode unit emitting huge amounts of RF. The RF would have been heating the motor. The laptop units will be made to a much highr standard simply because they have to be, to get FCC aproval. Oh well, hooray for Standards!

You could use the original PS IF you filtered the output well: a series choke and a good shunt capacitor (or two) should work a lot of good.

Regardless: earth one end of the overall PS!

Cheers
Roger

707

General Mach Discussion / Re: How to correctly ground very noisy spinddle

« on: July 05, 2014, 01:43:01 AM »

Frightening!
Yes, you should ground one of the wires going to the spindle - I think! I am assuming the laptop power supplies are all floating. If they aren't, you'll soon find out when the smoke comes out.

Better, MUCH better, would be to buy a proper power supply !!!!  Please!

Cheers
Roger

708

General Mach Discussion / Re: Losing Input Pins

« on: July 05, 2014, 01:35:28 AM »

See reply at Warp9 web site

Cheers
Roger

709

CS-Lab / Re: Servo drive as main spindle

« on: July 01, 2014, 06:16:30 PM »

Thanks Hood.

I do good single point threading on the lathe already, so that's under control.
I do a lot of thread milling on the mill, so that is under control as well.
But ...
I have to replace the spindle drive on my mill (the old one has died) and i was wondering about putting a DG4S16032 drive in and having full servo control of the spindle, and got to wondering - and searching. The number of options exceeds my budget ...

Cheers
Roger

710

CS-Lab / Re: SwapAxis

« on: June 28, 2014, 09:40:45 PM »

Hi Hood and Terry and all

I must be blind or stupid, or something. No worries.
I know I have read about 'swapaxis' - somewhere, but for the life of me I cannot find it any more in any of my manuals. Is there a good primary reference to this, or is it a miasma of allusions plus a small PCB?

I have to replace my spindle as the one I have is dying (has died), and I am wondering about putting a full servo drive in. Yes, the spindle motor does have a 512 line  encoder on it, even though all that has been used in the past was the Index pulse.

Mind you, I am happily single-point thread milling down to M2 and up to M24 with and ordinary spindle drive, but ... you know ...

Oh - yes. Mach3 and ESS.

Cheers
Roger