EMC and Galil and I/O

I've seen some capacitor ESR (equivalent series resistance?) meters that are supposed to be able to locate dying capacitors in circuit. They put out a high frequency at low voltage, maybe 100Khz at 100mV IIRC, and measure the equivalent resistance to the AC signal. I've been tempted to get one, or use a signal from my signal generator and measuring with an oscilloscope, to find bad capacitors on old motherboards, and replace with new long life capacitors. I've heard of others having success repairing old motherboards this way.

The autobalancers at work have a program in them but we don't have source code or anything to rebuild it. The company that makes them probably has a nice expensive upgrade for them. So far we have been able to replace their PC's with industrial PC's with an ISA backplane. These PC's are around $2500 IIRC, Some of the PC's for our truck tire x-ray machines are running around $12,500 because they use obsolete video boards, some cirrus logic chipset with 2MB and a high resolution monochrome mode that displays the x-ray tire image. Once I looked on eBay and saw the video boards for $15 but we pay $4000 for them from the OEM.

RogerN

IIRC the industrial PC they buy is around $2500. It has a 19" rack mount case, a backplane with ISA (and PCI?) slots. The CPU board plugs into the backplane just like any other cards. The pricing I remember is probably from 4 years ago, I'm now working at the other end of the plant in mixing so I haven't kept current with the other end.

RogerN

Thanks Roger..I guess the ones Im getting are not quite as Industrial as yours.

Gunner

One could not be a successful Leftwinger without realizing that, in contrast to the popular conception supported by newspapers and mothers of Leftwingers, a goodly number of Leftwingers are not only narrow-minded and dull, but also just stupid. Gunner Asch

I can still buy Intel boards brand new with full ISA slots. I am not sure where the factory is getting them from..but they have no issue getting them.

Gunner

One could not be a successful Leftwinger without realizing that, in contrast to the popular conception supported by newspapers and mothers of Leftwingers, a goodly number of Leftwingers are not only narrow-minded and dull, but also just stupid. Gunner Asch

The original EMC (EMC1) was indeed well written, but it was a fairly narrow implementation. 3 cartesian programmable axes, although the machine could be entirely non-cartesian, such as a SCARA robot or Stewart platform (hexapod). It had no ability to do any motion synchronized to anything outside the computer, so lathe threading was out, and even rotary axes (ABC) were not well supported. Also, the G-code interpreter, the motion planner and the servo control were tightly integrated, which made modifying anything quite difficult.

There were some who thought John Kasunich's idea of a hardware abstracton layer would foul up EMC terribly, and we even had one complete defection from the project over it (Paul Corner, the developer of the BDI, a live CD distribution kit). However, I think John's vision has been completely vindicated. There IS a measurable performance penalty to this level of abstraction, but PCs have improved in speed quite a bit since this started, so I really don't see it as a problem. It has made it vastly easier to create drivers for different motion or I/O hardware, and the set of HAL components to do various tasks interfacing this to that is a real blessing. It also made it possible to connect to a special real-time version of Classic Ladder, for machines with more complex I/O requirements, like tool changers.

I can't say that HAL made it possible to include both lathe threading and rigid tapping in EMC2, but I think it was instrumental in breaking out and isolating the major components so that this was also doable.

I also have to give credit to John Kasunich in writing the framework of a VERY flexible driver for my boards, so that one driver handles all 3 boards, and they can be combined to work together in one system. I never would have taken the time to craft such an advanced driver.

Plenty of other people deserve credit too, Chris Radek and Jeff Epler developed the Axis GUI, which not only is a great screen presentation but also fixes some rough edges in EMC, like making some problems with G92 just "disappear". They now have a touch-screen GUI, too.

But, it never would have started without the great work at NIST, mostly by Fred Proctor and Will Shackleford. I was the second outside user of EMC after Matt Shaver, and Fred helped me a lot get EMC running by sending me CDs and answering my emails. That was back in mid 1998.

Jon

Its actually "Enhanced" not extensible, but that label may be more fitting, now.

Jon

EMC2 uses a SPECIAL version of Classic Ladder, which runs in real time. Some features of Classic Ladder are not available, and the Modbus is just available now, I think.

Jon

Jon, thanks for shring, very interesting. I used TkEMC for now , but will give axis a try.

As for rigid tapping, it pretty much requires a servo driven spindle, right?

i

I happen to have a ISA pcb - NIC network interface card - and DOS driver.

If someone is trying to rebuild or put on a net an old machine.

I also have for HP Printers a number of NIC cards - Jet Direct.

I'm trying to find a SATA PCI card myself. Shop computer motherboard is starting to fail.

Tough working on a pc in 103 degree (F) in the shop. I think that is partly what killed it. But I don't use the computer when very hot.

It started when I had to leave it one night installing updates upon shutdown. It is a hyper-threading dual core cpu running XP - and one update shutdown the computer while the other was installing. I think after my comments to MS about it - they might do some checking. Been tough.

Mart> >

Can't speak for EMC, but I just have an encoder on a VFD driven spindle in both mills and lathes to do threading or tapping. Spindle encoder is master, axis is the slave. lathe is slightly more complicated as you hit the same spot several times at high speed for multi pass threading.

You won't hand tap again once your CNC does it.

Karl

Dan

One of my favorite hardware sources. And of course Ebay has them cheaper. I have a Microcenter near my clients in SoCal..so can take a DOA back in and exchange it.

It was 112 Saturday (yesterday) and a bit higher in the shop here at the homestead. I didnt run anything.

One could not be a successful Leftwinger without realizing that, in contrast to the popular conception supported by newspapers and mothers of Leftwingers, a goodly number of Leftwingers are not only narrow-minded and dull, but also just stupid. Gunner Asch

Thanks. It ould be difficult to add to my BP Series II Interact 2, I think, since the rear of the spindle is not exposed.

i

I had to get tricky on mine. On top of the machine, there's the hole for the drawbar. I made a little expanding collar. Compressed it, and slid it down on a little 1/4" lip. Removed the compressing bolt and tension holds it in place. Then just a little timing pulley over to another timing pulley on the encoder. Not much force here.

Karl

Karl, I do not even have a hole for the drawbar. I think that rigid tapping will be one of those things that I will miss. I am very reluctant to mess with the actual hardware of this mill.

Now, as an aside, having a servo driven spindle would be very cool.

i

For reference, Mach3 is also based on the original EMC.

There IS a way to put an encoder on, somehow. You got to figure it out. Now, this is certainly a future improvement that you don't have to get this summer.

My big mill has a brushed DC motor. An AMC drive for this will make a real nice combo.

Karl

I think, definitely not for this summer.

I am close to a big breakthrough on the mill work.

i

[ ... ]

Hmm ... doesn't the top end of the spindle go down with the quill feed? How did you accommodate that? (Something to consider for my Series-I BOSS-3.)

Enjoy, DoN.