Urgent: Dyna DM4400 / DM4000 measurements

Yeah, I hadn't considered the possibility of a set of holes in the column casting sides through which one could insert a rod that would (also) go through an I-bolt in the weight. That would be a nice way to stow a weight to detension the chain. The downside to this approach is that it would make the dimensions for the head block far more sensitive which is a problem for me (not next to the machine with a saw and blocking at hand).

Anyhow, I look forward to the pictures, but it sounds like I'm going down a path in at least some semblance of the right direction.

Hi Alan:

I got home today and couldn't get the machine to respond so what I have is limited but I think it should help. The machine was left with the head up, so the weight was down at the bottom. Here's a pic of that:

Attachment 425774

The four cap screws are holding a flat plate to the top of the weight with a gap underneath. I managed to reach in and get the eyebolt loosened up and pulled out. Here's that:

Attachment 425776

As you can see they didn't just tap into the weight itself but used a nut underneath in the gap. It's a 12mm thread but of course the thread doesn't matter as long as your eyebolt will go in the hole in the plate, so maybe a 3/8" thread if you don't have the metric. There's only a slight amount of room bigger than the thickness of the nut so the thread can't be extra long.

Fantastic! This is making a lot more sense to me now. I have to laugh though. I have no idea how you reached down into the column and removed that eye. It looks like the weight is 3 feet down or something and the hole is only 6x6 inches. Maybe it is just an illusion. But this does introduce a few questions though:

1. What is the size of the opening?
2. how big is the eye?

I'm guessing that there is no reason to remove the eye, so it is probably on my machine. I had envisioned some type of hook contraption that would be removed during setup and inevitably become lost. Granted, whatever hooked onto that eye is probably long gone, but I don't really care. I can lay a 1" bar across the top of that hole, tie a rope off between the eye and the bar, then slowly lower the head until the chains are de-tensioned, finally block the head. That should work fine. Question there though... do you really think blocking the head is necessary?

Great photos! Thanks again for being willing to help out.
Alan

Can you identify this photo. It was sent to me but I can't quite make it out. That seems to be the table below the weights, which would make the weights above the spindle. That doesn't make sense to me... Plus, the metal on the side looks like sheet metal, whereas the other photos show more of a casting (which is what I had been expecting). Finally, there are two eye bolts (which I asked about and was told that one was for lifting the machine with a hoist, along with the bolts on the plenum).

I'm all discombobulated with the photo and can't get a bearing. It seems like the photographer was standing on the Y axis way cover and shooting the photo facing the rear of the machine (and downwards, of course). But it doesn't quite add up and is definitely confusing me.Attachment 425780

Quote:

---

Originally Posted by MrMetric

Fantastic! This is making a lot more sense to me now. I have to laugh though. I have no idea how you reached down into the column and removed that eye. It looks like the weight is 3 feet down or something and the hole is only 6x6 inches. Maybe it is just an illusion. But this does introduce a few questions though:

1. What is the size of the opening?
2. how big is the eye?

I'm guessing that there is no reason to remove the eye, so it is probably on my machine. I had envisioned some type of hook contraption that would be removed during setup and inevitably become lost. Granted, whatever hooked onto that eye is probably long gone, but I don't really care. I can lay a 1" bar across the top of that hole, tie a rope off between the eye and the bar, then slowly lower the head until the chains are de-tensioned, finally block the head. That should work fine. Question there though... do you really think blocking the head is necessary?

Great photos! Thanks again for being willing to help out.
Alan

---

Definitely an illusion. The opening is roughly 5 x 7 inches, taking into account the Z axis motor housing which is partially over the opening. Eye bolt was just a little less than arm's length down, I was able to reach down with a small ball peen hammer to loosen it up. The eye is about 1-1/8" opening.

The photo in your other post is definitely standing where I did, at the front of the machine looking down, but the weight/opening is out of view at the back (top) of the photo. I recognize the same size table down at the bottom with the X motor on the left and the ball screw end bearing housing sticking out the right side. I would say the two eye bolts are attached to the drawbar mechanism at the front and the spindle motor housing at the back. At least that's what's in the same place on mine but it's different. You can just make out the chains on the right and left at the back of the photo.

If I could get the head down on mine I could show you the same view but that's not gonna happen until I figure out what's keeping mine from starting. Last weekend I started in on converting it from LinuxCNC to UCCNC--I put in a different hard drive in the controller pc and left the Linux installation alone in case I needed it before I got the other system running (like I thought I would do today) but for some reason my signalling over the parallel port is not getting through, so nothing responds. Not sure what I did.....

Great. Thank you for confirming the size. The weight seems to be fairly snug in the cavity, with little more than a half inch or so around the edges. It almost seems like wedging this during transit isn't as big a concern as I originally thought (?).

I'm still baffled by the photo I posted. Those look distinctly like weights to me, but are in the wrong place. They'd be adding to additional need for counterweight, not reducing it... But, I guess it is academic. The other photo clarifies what I need to capture.

I'm curious why you are changing from LinuxCNC to UCCNC. I never completed my LinuxCNC conversion, but I may have to do the Dyna someday. I had assumed it would be to LinuxCNC again..... The Dyna is working with the Mitsubishi Meldas M3 control so I'm not seeing a driving reason to retrofit it right now. I'd certainly welcome people's opinions on why I should reconsider that position though. The main reason I'm avoiding an immediate retrofit is that I want to have some fun! I've been converting for years and I'm a little burned out. And, I'm not too sure if I could reuse the Mitsu drivers. They are all digital and I don't know if I could interface these to LinuxCNC. So a retrofit quickly gets out of control... Do I need to replace the AC servos too? That would be even more money... You get the idea. But we'll see. I will learn more over time about how the Mitsu M3 works and whether I like it, but I'm going to stick with it for awhile (one argh is that it is 3 phase only... I kind of like being able to run without a phase converter if possible).

This looks like it might be a suitable bolt:
https://www.amazon.com/Swpeet-Stainl...1&s=hi&sr=1-12

Quote:

---

Originally Posted by MrMetric

Great. Thank you for confirming the size. The weight seems to be fairly snug in the cavity, with little more than a half inch or so around the edges. It almost seems like wedging this during transit isn't as big a concern as I originally thought (?).

I'm still baffled by the photo I posted. Those look distinctly like weights to me, but are in the wrong place. They'd be adding to additional need for counterweight, not reducing it... But, I guess it is academic. The other photo clarifies what I need to capture.

I'm curious why you are changing from LinuxCNC to UCCNC. I never completed my LinuxCNC conversion, but I may have to do the Dyna someday. I had assumed it would be to LinuxCNC again..... The Dyna is working with the Mitsubishi Meldas M3 control so I'm not seeing a driving reason to retrofit it right now. I'd certainly welcome people's opinions on why I should reconsider that position though. The main reason I'm avoiding an immediate retrofit is that I want to have some fun! I've been converting for years and I'm a little burned out. And, I'm not too sure if I could reuse the Mitsu drivers. They are all digital and I don't know if I could interface these to LinuxCNC. So a retrofit quickly gets out of control... Do I need to replace the AC servos too? That would be even more money... You get the idea. But we'll see. I will learn more over time about how the Mitsu M3 works and whether I like it, but I'm going to stick with it for awhile (one argh is that it is 3 phase only... I kind of like being able to run without a phase converter if possible).

This looks like it might be a suitable bolt:
https://www.amazon.com/Swpeet-Stainl...1&s=hi&sr=1-12

---

Yeah, that's the right eye bolt. Assuming yours is the same. Don't know about the lock nut that comes with it though, it might be pretty difficult getting a wrench in to try to tighten that on the bolt and I wouldn't want to just finger tighten it. Edit: I just re-read this and remembered you'd be able to bring the weight to the top so getting a wrench in shouldn't be difficult.

As for LinuxCNC, I have several reasons for switching over, not least of which is just for the fun of it (yes, I'm an unapologetic nerd) but for technical reasons as well. I first started using it about 10 years ago when I built a cnc router--I looked at Mach3 at the time, but since Linux was (is) free and I had a resident Linux expert (my son) to help me out, I got all the basics working and was able to apply my experience with it when I acquired the DM4400's. It's a good system and really very capable if you have the expertise in understanding how to integrate all the electronics/sensors, etc. I only have a very limited background in electronics but I was able to get all the basics working, just not the ATC, that was (and still is) way out of my league. I've been a bit frustrated at times trying to find my way around Linux--I've tried a couple of times to add a second parallel port to gain more I/O's but was never able to get it working so even if I had the ATC expertise I would still be at a standstill. I know there are other ways of adding hardware to Linux but all in all I'd rather just switch to something Windows based with the proper hardware for it. I had been looking at UCCNC off and on over the years and just recently found out they had a special offer until the end of August to get a free seat of the software when buying their entry UC100 dongle, so I just ordered that and waiting for it to arrive while I prepare the PC for it. I would really prefer their higher end boards for the additional I/O's, but figured for the low cost it would be a good starting point. When I (hopefully) retire in the next year or three or four, I hope to be able to really dig in to the ATC and see if I can get somewhere with it. UCCNC looks like it offers a great deal of support for machine hardware/sensors etc. so when I'm ready for it I will upgrade to the better board then.

I don't see any compelling reason for you to switch the 4400M over to Linux as long as the Mits controller is working good. It's an old system now but I've run several Mits controlled machines over the years and I've always been pretty impressed with their interface and reliability. I'm sure LinuxCNC would be able to handle all of it but I don't think I would be inclined to do it unless I was supremely bored (which happens all too often).

I ran the DM4400 for quite a while on single phase in the garage. The only piece of hardware that ran on 3 phase is the spindle (servo) motor so I used it on single phase for some time, but I found I was severely limited in how much material I could cut for roughing. Using a 2" diameter 4 flute index mill I couldn't cut tool steel at more than 0.020 depth with half the cutter width, and when the inserts got the slightest bit dull the motor stalled easily. I broke a lot of inserts that way. I just recently acquired through a trade a nice Phase-Craft 5 HP rpc and got it hooked up--what a huge difference it made! I made some test cuts under the same conditions, I was able to run the cutter down .060 depth with 3/4 of the cutter width with no sign of giving up, and since that's about the max I would cut tool steel at anyway I called it good.

Dan

You read my mind. The Eye shouldn't be too hard to put in with a lock nut if I've got the head down (so weight up). I'll see the machine later this week for a run through and I'll look then. I hope the mini-training goes well. The machine was jogged to the -Y limit and I cannot jog it off that now. It is kind of strange. Not sure what is going on there, but I'm definitely not familiar with the MITS so perhaps there is some mode that needs to be set? Hopefully the person that ran the machine will know. But, the servo amp definitely works, etc. No damage... I MPG'd to the limit very slowly; this is definitely a logical error.

I'm an electrical and software engineer, so LinuxCNC, control systems, wiring, etc don't really bother me. Oddly, however, some of the servo tuning stuff was very frustrating to me. Ditto for getting some of the configuration parameters right for LinuxCNC. The manuals are on the weak side, in my opinion, but the support in the forums is excellent. I did a nice job on the Hurco I was converting. Sadly, it looks like I'm going to be carting that to the dump to make room for the Dyna. I don't have room for the Hurco but it is a nice machine. sniff sniff. I will just drop the electrical box, though, so I'll have all the components ready for the Dyna if I want to convert that sometime.

I spent far too long on the conversion (start/stop). The really frustrating thing was, as mentioned, the tuning. I don't look forward to that if I do decide to retrofit the Dyna sometime. But right now, that control is working great so I see no reason not to just have some real working time. I have a backlog of CNC projects I'd like to tackle! The really big reason why I'd consider a retrofit is to get away from the 3-phase. I'll contact Mitsubishi at some point soon, but a tech told me that the M3 control requires 3-phase. It is more than just the VFD for the spindle; all the amps have 3-phase inputs too (AC servos). Internally, all of these drives are manufacturing their own motor voltages and frequencies (essentially they are all VFDs of a sort) using a high voltage DC bus. So those 3 phase inputs don't really do anything but get rectified made into DC (well, I don't really know AC servo technology, but I'm guessing it is the same as a VFD). But, that doesn't mean that the 3-phase isn't monitored and, therefore, necessary. It is all how they designed the system. So, if a tech says that the Mitsubishi control requires 3 phase, then I'm inclined to believe him. For a variety of reasons, I've got a PhasePerfect 10HP converter, which is kind of a Cadillac. I should be OK.

Alan--

You being an EE maybe I could pick your brain over something. I burnt out one of the small circuit boards on the servo amp for the spindle and I'm attempting to replicate it. Even though I know little about circuit designing I've been pretty successful at duplicating and creating simple signal distribution boards. I use Eagle and PCB-Gcode for making boards on the mill. Anyway, this is a pretty simple shunt regulator board for which I actually was able to acquire a schematic from Servo Dynamics and I've got most of it worked out and I'm ready to order components but I'm a little puzzled by a couple of things. Got time to answer a few questions about it?

Dan

Sure.... I'll send you a PM with my email address. I've moved entirely into software now in my professional life, but I still do a little EE stuff for fun. Its funny though. Engineering is like any specialty. I have a family member that is has a specialty in ENT surgery, but some people thinks he knows about prostate stuff (or whatever). Granted, he may know more than the average person, but there are probably a lot of well read lay people out there that know more. :) The same applies to engineering. There are a gazillion specialties out there and very few (none?) are masters of all. Anyhow, I'll try to help you out if I can.

Alan

The following two drawings are *very* rough representations of a Dyna DM4400M (Mitsubishi control) with and without the enclosure. They are for reference only as a means of figuring out space within a shop. Tolerance is literally and inch or more, so don't place a whole lot of stock in them! But, it is better than what the manual provides which is, well, nothing! :)

One drawing is the front and side view. The other is the top view in both clothed and unclothed states.
Attachment 425890
Attachment 425892

I just managed to snag a SKIP manual and they have a real line drawing. That is far better than my cheesy thing. The drawing, below, doesn't have the control panel. On a Mitsubishi control, it is roughly 2' x 2'.
Attachment 425928