my Multicam project

[Boltz]

As I've posted before, I'm a new Multicam owner. Former sheet metal fabricator, ran a Strippit CNC turret punch press for 20 years, and then my department was phased out (now they're paying about 2-1/2 times my fully burdened cost to buy the parts from an outside vendor!). Still looking for work in sheet metal, but at 50 years old, the odds are not in my favor, so this is "plan B".

It weighs about 5000 Lbs., and I only paid a little more than a dollar a pound for it! I figured at that price I could probably part it out and at least break even.

Haven't ruled out a complete motion control retrofit, but for right now I'm working with what I've got, which is the Extratech M24, Electro-Craft BRU-series drives, and Minarik motors. One of the X-axis servos throws a following error fault if you run it at full rapid for it's full travel, but works okay for cutting. I sent an extra junk box motor to the folks at NC Servo, and they're working on it.

The Extratech controller looks to be a pretty good DSP unit, with an Intel i960 commanding assorted FPGA's and whatnot from from top name manufacturers like Analog Devices and Cypress. I got a pretty good set of schematics with it, along with a bunch of source code for it's functions. Still doing detective work on how to update it's NVRAM. When I talk to Multicam they say it's old as the hills (completed and shipped in Dec. 1999, as far as I can tell), and they don't have much info on it. Extratech says they can't release any information without approval from Multicam.

Here's a picture of the dust foot I just built and installed. It's inspired by the Thermwoods with "dual exhaust". The clear junction box looks a little goofy,
but I like being able to monitor the dust flow, and I had some small pieces of half-inch polycarb laying around from making custom base plates for hand held routers. I looked for a long time and couldn't find a suitable "Y" fitting.

Thinking about adding a hole in the j-box with a rubber diaphragm which would give me a quick visual check on dust collector suction, and thus filter condition.

Unfortunately, when I stand back and look at that 5HP Colombo in the middle of it all, it looks pretty small... That thick steel gantry could haul around a 10HP spindle pretty easy. Looks like 3/4" shank is standard for a lot of the good tooling.

Also, the Z-axis lead screw (acme 1/2-10) has a bit of play, it looks like the Z was worked pretty hard at it's last job. Contemplating whether to just try a new nut and screw, or go for a ball screw upgrade, which would involve changing the pulleys from 1:1 to 2:1, since the slowest ball screw I could mount would be a .200 pitch. The servo would have to be retuned also. Open to ideas on this, also wondering if air cylinders would be superior to springs for keeping the backlash tight. Certainly, they'd be linear (unlike springs), but would it be worth the trouble? They sure look cooler!

One of the things this machine seems like it would be well suited for is custom carved door panels, both cabinet and entry. I'm hoping that the combination of heavy weight and DSP servo control will enable me to do stuff that aluminum gantry machines with steppers can't, or do it at speeds they can't touch.

More later...

[cabnet636]

looks like a great project to me!!

jim

[Boltz]

Jim, thanks for stopping by! I gotta say it totally amazes me that you can keep on top of so many forums, and get so much work through your shop.

I've googled myself silly looking for a Multicam forum and haven't found one, so I guess I'll just plant myself here and see who shows up.

I'm gonna see how far I can get with the original controller on this machine, because I've been really impressed with what I've gotten out of it so far. I guess that's from my roots in sheetmetal running a standalone Fanuc 6M-B. I've seen windows do so much flaky stuff over the years that I just don't trust it, even if the controller is on it's own board.

Keep up the good work, and I'll probably see you over at the Vectric forums!

regards,
-Jim Hart

[sodfarmer]

These routers are built like tanks. I've programmed and ran one for about six years now and they will cut about anything. Good luck on them having any schematics. They weren't very good at documenting routers at this time. I have found the best thing to do is contact your local Multicam office. I have dealt with the Chicago office and their router guy knows his stuff. If you adjust the motors on the x axis they will run fine. The main thing is to adjust your homing blocks on the x axis. This will save you alot of headaches. If you have any other questions I'll try to answer them.

[Boltz]

Yeah, I don't think Multi's are quite "Big Iron" like Thermwood or Motionmaster, I guess they're "Big Iron Lite"... no worries about flex in the table and gantry on a machine like this. The prices are pretty good in the used market, and if a guy is willing to take some time to "crack the code", and figure out how the controller works, you can get one helluva machine for the money!

Fortunately I've got a halfway decent set of schematics, I was lucky enough to get the PC which ran the machine before. It's got about 10 installations of MX32 on it, too! Looking for advice on a compact rugged basic PC to serve files to it.

Had several chats with the woman who answers the phone at MC Chicago. They act like my machine dates back to when dinosaurs roamed the earth, even though it's barely 10 1/2 years old! Like, do they expect you to shell out $70,000 for one of these, run it for 5 years, and then scrap it out?



I did get a good tip from the folks at MC Chicago, though. I was getting an error message at the control station which said:

POWER FAULT
POWER GLITCHES 7

Although they couldn't rummage up a schematic for my "ancient" (1999) machine, I did get an email suggesting that I carefully adjust the 5v power supply which supplies the controller board as close to 5v as possible. It was reading 4.72. Since they said I could blow my controller board if I went too high, I took it to 4.97 and tested, haven't had that problem since.

Can you clarify what you said about adjusting the motors on the x-axis? Are you talking about adjusting two things, the motors, and the "homing blocks"?
Or just about adjusting the homing sensors? It looks like that's how one squares the gantry: by loosening the carriage mount bolts and the sensor mounts, then sliding the mounts a little and re-homing it until you get it square.

Or... something else?

Thanks again, for stopping by! I'll be away from the shop for 2 weeks, but I'll be checking in here.

Keep it coming, former and current Multicam guys!

Here's a pic of me and "old reliable" after the feeling of "what the *hell* did I just do?" subsided a bit. I've since taken the big risers out.

[sodfarmer]

The one I run has a stepper system so I don't know if its the same. On ours you have aluminum blocks with slots, we unplug our motors and move the gantryto where the homing switches begin to be lit then unlit. We then just adjust our blocks to get it dialed in. We usually never adjust the switches, just the blocks. Also have your camera ready when you need to order parts, it helps. When you figure out the controller, you will like it. There are many "tricks" that can help.

[MCPservice]

Morning....the name is Chris and I work on Multicam machines, that is to say I repair, install and train, and the problems I have read through so far, none are that bad...SodFarmer stop moving the target block they were not meant to be moved to get the machine to find home, it is easyer to have the gantry square itself. Call me or email...and I will explain how to do this.

There are old schemetics if you can provide me with your serial numbers, I just need the last 4 digits of your machines and actually....the 5 volt power supply should be adjusted to 5.07-5.09 DC, the control board is powered by the 5 volt PS, and the control board requires at the very least 4.9 anything lower will cause some odd behavior of the machine and at 5.0 volts you are only .1 away for low voltage error.

MX32 software can be upgraded to PSS software for your DNC files. With a M24 board and as long as the second control board( there are two board ontop of one another) as longs as the second board is a J or K board, you will be able to run XP or higher with PSS.

The backlash in your Z screw is 2 or 3 things...the nut at the bottom of the z-screw is either lose or missing...so you can physically pick up on the head and feel the head lift about an 1/8" or the supernut on the z is worn, and yes it is called a supernut, or the belt/pulleys are loose and if you are using a lead screw right now, a upgrade to a ball screw is available.

If you have any question...please give me a call or email me...

Coliver@mcpservice.com
678-548-1483

[Boltz]

Like I said in my previous post I've got a good set of schematics which came with the machine. Thanks for the advice on the logic PS, zero faults so far since I adjusted it, but maybe I'll go back in there.

I'm starting to get pretty comfortable with MX32. Although the host PC is an ancient win2K box, I've tried installing MX32 on an XP computer and it seems to run just fine. What would I gain with PSS?

Thanks for the tip on the Z-axis... it's obviously not Nook, so my next guess would have been Thompson. I'd love upgrade to a 10HP spindle, sill studying what would be required, particularly whether the existing Minarik motor would be up to the task of hauling up a 75 Lb. motor. I'm thinking 3/4" ballscrew would be ideal.

Thanks for stopping by my thread!

-Jim Hart

[sodfarmer]

I told him to move the target blocks because I watched a tech from Multicam do this. He worked for Multicam and this is what he said to do. They have slots in them for a reason.

[MCPservice]

You watch the tech do that because the machine was soooo far out of being square, once the gantry is square you never and I say never move those blocks again. Yes they have a slot, for adjustment, that is because the old machines locating block, bolt holes were not drilled as precisly as the new machines are now, the new ones don't have that adjustment.

Once the blocks are in the right location, and the next time when you have problems with the machine finding home....you loosen the bolts that attach the gantry on to the bearing cars.

Loosen all bolts that attach the gantry onto the bearing cars, should be 4 bolts on each car, loosen until you can turn the bolt by hand. 16 bolts total/ maybe 8 bolts depends on the machine.

Now that you have loosen the bolts. the gantry can now pivot a
little.

Have the machine find home. shift key green0/0

After it finds home

tighten one bolt on one side of the machine, then walk to the other side and tighten the diagonal opposite, just like you were tightening a tire.

You have now tighten 2 bolts

Move the gantry back a couple of inches.

Find home again and repeat the process.

You have now tighten 4 bolts.

Move the gantry back again.

Find home again

This time tighten all the remaining bolts.

It should be square at this point, easy way to check after you have found home, put your hand on the gantry....and hit E-stop, if the gantry is not square you will feel the gantry kick.

i know it kind of long....but I hope this will help.

[Boltz]

Hey, thanks for the gantry squaring procedure! I was pretty much thinking that was how it's done, until I was reading the user's manual (I've built a somewhat successful career on being the one guy who's weird enough to read the manual), and it tells you to change the "XSQUARE" value in the machine parameters menu.

My machine has non-moveable bolts screwed right into the racks, and the prox sensors are mounted on slotted brackets, as shown in the picture.

[Boltz]

Well, the picture in my last post showed that prox sensor in it's last days. It now looks like the photo below.

While swapping the motors on the x-axis to troubleshoot my servo problem I discovered a loose pinion. Oddly, it was on the motor that seemed to be working properly. Being an experienced troubleshooter, I know that it's best to just change one thing at a time, so I tightened the pinion and put it all back together.

Evidently something malfunctioned during the homing process, causing the motors to get out of synch and smash the proximity sensor into the bolt which is normally it's target. This is probably the thing I like least about the Multicam so far.