[michaelthomas]

I can probably take it down to the local print shop and get what you need copied......I haven't spent much time with it yet, so I don't know what is in there about the spindle and knob.

[erikjgreen]

Right now, that's the only part I can't figure out. Long term, I'd like to know Lagun's recommended process for tramming the mill, lubrication type and interval for the spindle motor, and I'd love a wiring diagram.

Actually, if you could get me a description of what the knob does plus a wiring diagram, that would be awesome.

Thanks,
Erik

[erikjgreen]

Ok, I found out that the knob on top of the machine is for a spindle brake (I climbed up a ladder and cleaned it off up there, simple).

From my reading online it seems like the spindle brake is adjustable, electrically operated, and that knob is something to do with adjustments. It's stuck in the "release" position at the moment... which should be good?

Other than that, I don't know what to try. I did get the 3 axes tested a bit more tonight, figured out why one of them isn't setting "ref" correctly but couldn't get the others to do so either. They seem to get errors at random locations during moves for some reason, although all three axes move without much noise or complaint, at varying speeds. They only move when the "spindle speed override" knob isn't in the off position... doesn't matter where it is if I select a speed from the menu, but it can't be off completely.

I still can't get the spindle to turn... I tried the obvious, the on/off switch on the dynapath. No luck there. Tried the "spindle speed override" knob in the off position and on.

Suggestions welcome. If I can't figure out the brake, I may take it apart or off, and trace the wires back to the controls to find out what runs it. Long term I'll use a VFD with a resistor brake anyway, but it'd be nice to make the spindle turn.

Erik

[michaelthomas]

Give these a look. Let me know if they are what you need.

[michaelthomas]

Here are some more.

[erikjgreen]

Well, I took a look. The 320 mill seems to be a nicer model than the one I have

I think some of the wiring labels are similar, so this will help me work out some of the limit switch and reference switch wiring. It seems to have a spindle drive (vfd) unit, so the contactor/brake setup I have isn't present.

There are some similarities in the cabinet layout, although I only have one cabinet and this has two.

It's definitely a step in the right direction.

Thanks,
Erik

[erikjgreen]

Okay, I had some success last night. I went through the process in the manual for referencing the axes, and got two of them to reference pretty easily. I think the reference switch for the Z axis is broken, since it seems to miss it most of the time.

Also, if it hits a hardware limit switch, there's no way to move off of it except manually it seems? I would think there'd be a way to clear the error and move elsewhere, but I seem to end up moving ballscrews by hand.

I did follow the directions for the dynapath and got a short program entered to run up the spindle. The dynapath has entries for spindle speed even though it can't control it except for off and on, and direction.

The gear lever on the head of the mill goes into the low settings easily but has more trouble with the high side, it won't lock into its detent. The gears seem to work in high more or less, but I think it slips occasionally, so I didn't try that too much. Long term I'll use a vfd for that anyway, not sure if I'll keep the gear switch. I'll pull the head apart to have a look at the gear select mechanism sometime before I start cutting things.

I made sure the lube system is working... the spindle is great, it sprayed extra oil all around when it started, and I can manually pump oil into it and get it to drip out. It runs well, most of the noise comes from the motor and gears, nothing from the bearings.

The ways are more complicated, they have an electrically operated timed pump on them. I cleaned it and filled it with oil, and I can pull the manual lever to send oil to the ways, but it seems to go back down verrrry slowly. Not sure that's normal, I really need to check and see if it's shoving enough oil out the ends, and maybe find a way to see if it's running its timed cycles.

I could possibly connect an air tank, put a tool in, and try the mill out using jogging or a short program at this point, but frankly I'd rather move on... I know what works and what doesn't, and the Dynapath controller is very hard to deal with (and the screen is very dim about half the time). My Mesa electronics order should arrive tonight, so I can start work on the conversion this weekend. Should be fun. I'll take some pics so I can have something more than words to post.

Erik

[erikjgreen]

Here are some more.

Yes, thanks for sharing these. There's enough in common between models that these have some useful labels.

These plus a circuit tracer should make my rewiring job much easier.

Erik

[erikjgreen]

Not a lot of mill stuff done over the weekend. The weather was really nice, so I got some car and boat stuff done.

I did spend some time mapping wires. Most of the limit switch, servo, etc. wiring is routed into the cabinet along the right side, where it connects to a barrier strip. The other side of the strip is mostly single lug connections, in other words all it's used for is a tie point between wires inside the cabinet and outside.

I'm going through this and noting down position of wire, number label if any, tracing it to where it goes inside the cabinet, and if necessary using an RF wire tracer to follow it to its source on the machine.

When I'm done I should have a nice map of where the connections are for the switches, servos, encoders, etc so I can install the new electronics.

I'm trying to decide what to do with the pendant... that's why Lagunmatic calls the monitor/control box where the dynapath screen is. It's integral with the cabinet on the side of the machine, which looks ok, but there are two problems. First, it may not be easy/possible to mount the new controls and screen on it and have them look good and be usable. Second, the location of the controls is close enough to the table that it may interfere with building an enclosure around the table to contain flood coolant and chips.

The alternative is to replace the electronics cabinet with a wall mounted cabinet, extending the wiring from the mill as needed. This would also free up space next to the mill for the coolant system. The controls would either be on a longer arm from the mill top, or be free standing on a rolling cart of some kind. Decisions to make.

Erik

[michaelthomas]

Hi Erik,

Can you tell me what to use to transfer pgm's to the control from PC.

The control has a db25 RS232 connection, but I have no such serial connection on the PC ( I have a DB9 serial, though).

Do I need to get a DB25 serial card for the PC?

[michaelthomas]

Never mind.......lol. I found the answers in the Dynapath control forum.......which I was not formerly aware of.

[erikjgreen]

Spent one day this last weekend starting my pull of the old hardware.

I've got the wiring diagrams from related mill models posted here, and I did some measurements of the current and voltage at various points in the old layout. The big transformer near the top center of the cabinet is wired for 83 volts, which rectifies into 125, so minus circuit losses that matches the 120 volt drive voltage for the servos.

The SD1525 card cage has its own power supply, which runs off a 110v feed from the other transformer. There was also a dedicated power supply for the Dynapath delta 10 hardware.

Anyway, after about 3 hours of careful removal, I had the Dynapath card cage out, many wires removed in the cabinet, and the wires connected to the servos (tach, encoder, and power), coolant pump, spindle motor, limit switches... basically every item out on the "main" section of the mill that connects back to the control cabinet I've got the wiring isolated for.

Here's the inside of the Dynapath console box:



Note the long cable coiled inside. Also note the hole at right which goes through the hollow support for the box down to the control cabinet. All the wires ran through there. Unfortunately, the dynapath console, support, and control cabinet are all one piece sheet metal. So if I want to re-use the existing cabinets I have to put it in exactly the same place, and either mount the (bigger) LCD across the front or get a smaller one to fit the box.

Here's the opening in the top of the main cabinet that connects to the support:


It's a really convenient way to run the wires, but it's just not right for what I want to do with the mill. At the very least I'll remove the dynapath console and support arm by using a portable bandsaw to cut it off, then I'll patch the big cabinet.

But the big cabinet staying there means that whole side of the mill has a big item in the way if I want to put something there, like a tool changer. Plus it may interfere with the enclosure for flood coolant I want to build.

It's a big project to replace the main cabinet, but now's the time to do it if I am going to... I picked up a 6 foot tall 19" metal cabinet in good shape from a surplus place last week. It's got casters, screw down legs for leveling, and it's big enough to fit all the electronics. I'd have to have some kind of large cable or conduit linking it to the mill, which would have all the electrical lines in one big bundle going to it... not sure that's a good idea. Also not sure how I'd pull off bundling all the wires into a flexible package.

Ideas welcome on that.

Anyway, at the end of the time I had set aside for this, the cabinet is mostly empty:



Across the top are, left to right, the fuse/disconnect switch for the cabinet, which links to the cabinet latch, the main transformer for the servo drive voltage, the secondary transformer for the electronics, and a barrier strip for connections. The lower right is the Servo Dynamics drives and power supply plus another barrier strip.

Middle of the cabinet left and upper right are the wires going to the various servos, pump, etc. They're pretty easy to trace.

I've asked a question in another thread about Viper drives... I'm trying to decide if I should replace the SD drive setup (in hopes of getting something smaller and quieter).

SD drives, 3 axis, all working (left side of the pic is the dynapath power supply):


Once I work that out and decide which way to go with the electronics cabinet, I'll be assembling the new electronics into it.

In the meantime I'll be working on building out my control PC, which will run EMC2 and have a Mesa 5i20 card in it.

If anyone out there is looking for either SD1525 drives or a Dynapath Delta 10 setup, let me know.

Erik

[erikjgreen]

I've spent some time on and off the last week working on the computer controls. I had bought a Dell Pentium 4 desktop system that just needed a hard drive for $10. It's a fairly fast CPU, but once I installed Ubuntu and EMC2, the latency testing showed only fair to middling results (50000 jitter).

I expect latency won't be a huge problem with the Mesa board anyway, but I'll try to find ways to improve that. I still need to check the BIOS revision, and I've turned off power management and BIOS usb support. I'm installing a new video card too, rather than use the built-in video.

After installing Ubuntu and running the system upgrade I got the pre-built EMC config for Mesa/servos running and the lights on the cards looked good. Now I need to do the video card, get a driver for my touchscreen working, and test out the actual I/O of the Mesa card.

Then I'll mount the computer in the cabinet near the servo drives and start connecting wires... probably starting with the limit and home switches before I move on to the servo connections, which will need some power supply tweaking now that the old hardware is mostly out of the cabinet.

Onward and upward.

Erik

[erikjgreen]

After a three month lag between posts, time for an update.

If you saw my other posts here, you know I've been working on the power connections for this mill, plus I bought some parts for my CNC lathe project. I've also spent a good part of the last two months "organizing" my house and shop, so now most of my tools are in one place, a lot of junk got thrown out, and I'm back working on the mill (at least it's the higher priority project at the moment).

I worked out with some help in another thread that the Servo Dynamics chassis has two power feeds, one 83VAC and one 110vac. I think the 110 is just to run the fans... makes sense, you couldn't get full performance on them off of 83 volts, and powering the servo drives off 110 may be possible but makes things more complicated.

Anyway, I spent part of this weekend pulling out my phase converter, which I will re-mount in a portable form (sort of.. it's kinda heavy) for other uses. With that wall space freed, I ran conduit for 6 gauge power connections from the box that powered the converter... it's 4 gauge wire to there, so I can branch off another set of wires for eg. my air compressor from that if I want to.

I'm going to try to pick up a piece of SJOOW (flexible neoprene) cable about 7 feed long or better today at lunch, to run from the wall box to the mill. The existing cable is 10 gauge, and I have picked (but not yet bought) a Hitachi WJ200 AC drive to use for my spindle, which is 5 hp/three phase. Using Hitachi's derating document, it needs to have 6 gauge wire run to it if it's powered off of single phase, which I'll be doing. It's a nice drive, I'll have sensorless vector mode plus braking and all the other bells and whistles, and I can put an encoder on the spindle to get accurate positioning.

I'm not sure yet what I'll do with the spindle brake on the mill... I could use it for E-stops, or wire it in so the control PC can use it for doing a fast reverse, but I don't know if it's needed, especially with the AC drive I'll use.

My next steps at this point are getting the servo drive chassis powered up, tuning the drives in torque mode per the manual for a similar servo dynamics drive, and using a 1.5 volt battery to test moving the axes. Once that's done I'll be looking at hooking up the Mesa cards for control and encoder feedback.

Erik

[erikjgreen]

I've been working on wiring, and also on getting the computer to fit in the Mill's electronics box.

I thought about it and decided that the existing CNC control console is both too small and badly positioned for use, especially if I enclose the table (dunno if I'll manage it, but I'd like to build a full enclosure suitable for keeping flood coolant sprayed at medium pressure on the tool bit).

So, I'll be cutting off the "arm" and the monitor from the mill's electronics cabinet. The existing hole for wire routing I'll either use for a vent or (since an upward pointing vent isn't the greatest idea in a potentially wet environment) plug it with sheet metal or possibly a status pole (with lights) and put the vent on the side of the box.

The computer and controller boards from Mesa are going to go in the electronics box, and the keyboard/monitor/mouse etc will be on a separate rolling stand.

Here's a pic of the cut-down chassis for the computer. It's a dell Optiplex desktop that I built up with EMC2. It turned out to be a nice base for this, since the motherboard is mounted on a sled that snaps into the case. I could have just mounted the motherboard loose in the electronics cabinet, but I wanted support for the mesa cards other than the card slot connectors.



The silver metal chassis in the lower left of the mill's electronics box is the cut-down sheet metal from the Dell. It will be housing the "sled" for the motherboard as well as the power supply. Here's a pic of it on the floor with the motherboard and power supply installed, but not connected to anything.



Sorry for the fisheye effect, I happened to use my wide angle video camera to take these.

Erik

[erikjgreen]

Well, I got a few things done. The PC is mounted in the cabinet and connected to the Mesa boards, and it boots up fine. I had to go to plan B on mounting the Mesa boards.

I had gotten some nylon stand-offs, and I was going to put the boards on them standing clear of the metal backing plate in the cabinet, but I didn't have the right combination of screw sizes and drill holes to make it both fit and be solid enough to trust.

So I took a spare piece of white polycarbonate I had, cut it to size, and screwed that to the backing plate, then mounted the Mesa boards to that, being careful not to warp them with the screws. I may need to re-do the screws with a washer or some other spacer under the board corners if there are any problems with heat.

Using a combination of the diagrams posted in this thread from the Lagun CNC 310 and some free-form notes I found on the net from a CNC repairman that gave the pinout for the encoder cable connector on the Dynapath Delta 10, I found the appropriate wires and shield for the Z-axis encoder and connected it to the axis 2 connectors on the servo interface card (the connectors are numbered 0-3).

I started up Ubuntu and EMC, verified that the Mesa card firmware loaded, and tried to use hal_scope to read the encoder by manually moving the ball screw. No luck.

Going back to RTFM, I read through the hal_scope instructions and found out I'd been missing some trigger settings. Fixing these, still no luck

So I went back to the Mesa manual and found out that since my encoders were differential and the default type was TTL, I had to move jumpers. So I shut down, moved jumpers, rebooted.

Then I tried again. Still no luck with hal_scope, but hal_meter works. Also, the main AXIS display works, showing the Z axis position changing when I move the ball screw. Very cool.

So, today's next step (if I get time) is to figure out the appropriate values for my ini file, encoder counts and ballscrew pitch, to make sure distance measurement is accurate. The servos all drive the axes via timing belts, so I have a gear reduction there to think about, plus quadrature encoders.

Probably the encoders are 1000 count, so with a 2:1 belt reduction that should be 8000 counts per turn of the ballscrew. (1000 * 4 * 2). I haven't measured my ball screws yet (or the pulleys for that matter), but a rough guess puts me at .2 pitch (5 tpi), so that's about 40,000 counts per inch of ballscrew movement or 0.000025 inch per count of resolution.

So as mentioned earlier in this thread by another Erik, I'm probably limited by the backlash in the ballscrews and mounts more than the encoder set up.

Assuming I can get the numbers in EMC2 set up tonight I'll tweak the Z axis wiring and put it in a more permanent position, then connect the other two axes and test them, putting the numbers for them in EMC2 as well.

Then it's on to the limit and home switches, which should be pretty quick to get working.

Erik

[erikjgreen]

Forgot to mention, for future reference in case anyone else converts one of these: The encoders are differential "three wire" type, meaning they have A, B, and index outputs.

So each encoders has 8 physical wires connected.... A and -A, B and -B, Z and -Z (or, in the lagun diagrams, "marker") and GND and Vcc (+5v).

I'll post a pic plus the wire color code later, once I get the other axes' encoders wired up.

Erik

[erikjgreen]

Got the other two axes' encoders hooked up, and neatened up my wiring a bit. No limit switches yet, ran out of time.

I'm going to order a Hitachi WJ-200 VFD today, and I've been thinking about the spindle pulleys.

Does anyone know how much/if the spindle pulleys on a variable speed mill head will create backlash for operations where the direction of the spindle reverses?

The existing head has a speed range from several hundred up to 4000 rpm depending on which gear range I use, selected by a crank that connects to a chain in the head, which adjusts the speed pulley diameter. With the VFD, I won't generally need to adjust that, and I'm wondering if I should replace it with a timing belt setup.

The spindle encoder I'm planning will have an index for absolute positioning, and I'd like to "know" exactly what position it's in with as little backlash as possible. Anyone who has done this and replaced that belt/pulley?

Erik

[erikjgreen]

I didn't get any more wiring done last night, but I did verify (via counting rotations) that all three axes are 2:1 ratios. Assuming the servos can get up to close to their no-load speed (2500rpm) with a load, then I should get (2500/2)/5=250 inches per minute rapids max.

I am not seeing as much lubricant as I would think normal getting onto the ways from the Bijur lube system, so I've got to check the tubing and whatever the end fittings are called. The spindle lube works fine.

The hitachi VFD I ordered can use an external braking resistor for higher braking performance, although it has a built-in one too. I want to be able to insta-stop the spindle for tapping among other things, so I'm going to see if I can find a surplus 250watt or so resistor at a local surplus place. I also want to install AC line reactors to protect the vfd electronics and help the spindle motor run cool.

I'm also going to pick up a few solid state relays so I can power on and off the coolant pump, task lighting, the vfd, servo amps, and compressed air from the Mesa I/O card. Basically, powering on the mill with the cabinet lock will boot up the computer, which can then use the Mesa to power up everything else. That also means I can use e-stop to cut power to all those items if I want.

Erik

[mungaro]

Eric

Hi, all;

I'm about to start cleaning up, testing, and then updating a Lagunmatic CNC 250, a 3 axis CNC knee mill I bought almost a year ago at an auction.

It's currently got a Delta Dynapath 10 controller, a 5 HP spindle motor, and the original electronics in a side cabinet. I don't have a pic yet of my own mill (I just finished moving it into my shop space) but here's a generic pic of the same model, to give you an idea:



I've almost gotten my phase converter assembled, and once that's done I'll be able to power the mill up and test it.

Longer term if the servos and drives are good, I'm going to replace the rest of the electronics (everything but the drives and power supply parts) and run the machine with emc2 and a VFD on the spindle. I want to be able to run some rather large programs on it as well as add a 4th axis later.

I do have a few questions at this point:

1) Is there anything I should be aware of as far as "gotchas" for starting this thing up the first time? I'd hate to miss something that caused damage. At this point I'm planning on making sure there's lube getting to the ways and spindle, then trying manual moves of the table, then spinning up the spindle under power.

2) It looks like I have an auto-lube system for the ways and spindle (2 Bijur lube systems) both have oil that looks good. Should I do anything to recommission these? Add more oil? Test?

3) Mostly at this point the mill needs clean up and a bit of de-rusting.. there's no serious rust on most of it, just one spot on the table where there might be a little pitting. I was going to clean the rust with naval jelly, sand with 320 grit, wipe with xylene, then a thin coat of grease or way oil. Should I avoid this method for the surface rust on the Z dovetail? The other parts of the ways just have grease and dirt on them, but the Z has some rust. I have hard chromed ways, I think.

Thanks for any help you can offer, I'll get pics of my own setup up soon.

Erik

Eric:

I have a the same machine with all the books and manuals. When you start it up make sure that everything is in the off position. The machine needs to be zeroed first or it will not take off and work.
mark