[Verell]

Hello,
Yesterday I lucked into a SuperMax YMC-40 CNC mill in near mint condition. The COMPUMATIC MicroPath Series 40 controller is flaky. At what I paid if I can't get the controller fixed I'll replace it with a Mach3 setup & still be $$$ ahead.

It'll be a week or so before I receive it as I have to make space in the garage & schedule the riggers to deliver it.

The spindle is a 3phase 3hp motor & I've got the usual 1 phase 220V. Will this VFD/Phase Converter handle the spindle motor: Hitachi VFD. 3HP 200-240 volt, Rated 2.9Amps/10A overload , three-phase. Part#X200-022NFU.
http://www.hitachi-america.us/produc...200/index.html

NEED DOCUMENTATION:
The system came w/o any documentation. Any chance that someone might have documentation on the MicroPath controller?

[radioactive]

That's the same convertor I bought for my mill as well... it should work great.

I should have my YCM-40 here this weekend to start my conversion. Going to Mach 3, getting rid of the Anilam controls...

[Verell]

The riggers just left. Still a lot to do before it's usable, but now I can start!

INTERSTATE RIGGING ROCKS!!!
It was delivered by Ed Wallace, owner of Interstate Rigging LLC in North Andover, MA. Very professional & well equipped outfit. The delivery & placement went extremely smoothly, the team did it smoothly, & quickly due to no waste motion. Everytime one guy needed something, the other had it ready to hand to him.

They suggested a couple of small position changes to avoid me having problems down the road. Made sense, glad I took their advice. They placed the machine by eye & when we measured for final positioning it was only 1/2" out of square, a small tweak with a prybar & it was perfectly positioned!!! I highly reccommend them if you're moving machinery in MA/Southern NH/Southern ME!!!

Next steps:
- Install motor & draw bar
- Install VFD/3 Phase converter, s/b delivered tomorrow
- Run a power line from the breaker panel.
- Begin debugging the controller.

Some pix:

[helixcnc]

Hi Verell,

I am happy to see that the riggers got the machine in with no problems. See I told you they were the best. I called an old friend today to see about info on the control and he said he would talk to the engineer to see if he has or knows anything about the MicroPath control. If I find any information or find any manuals I will let you know. Good luck and I hope you get it to run.

[Verell]

Thanks Tom,
Anything you can come up with will be a BIG help.

[Verell]

I've been poking around & looking at things before powering it up.

It turns out the main control PC card is a PMAC of some kind (I think a PMAC-Lite) built by Delta Tau Data Systems. This card handles all the servos, taking the encoder outputs & generating the input signals to the servo amplifiers!

Delta Tau is still around & from their web site they're still making varients of the PMAC cards!!! From their descriptions, the new cards are still heavily derived from their initial designs. This is extremely encouraging as it means that there's a good chance that if the card is bad I can either get them to repair it, or I can locate a good used one!!!

[Verell]

Today I reinstalled the spindle motor & power drawbar that were removed for shipping & delivery.

The spindle has a Bridgeport style V-belt variable speed drive. I was expecting more of a battle with the spindle V-belt, but it actualy wasnt hard at all. I tilted the motor a bit & slid it forward so I could get one side of the V-belt up into the pulley. Then with 1 hand I pulled down on the bottom part of the pulley opening it up & pushed the belt up into place. Once I had the belt in the pulley I just moved the motor around untill I could get 1 bolt in 3 or 4 threads, then could shift the motor until the other bolt went in. Torqued the motor bolts to 18 ft-lbs & that was done.

The Kurt draw bar was easy, just oiled the mating gears, slipped it into place, installed the 3 hex cap screws & the air lines. Verified that when the drawbar mechanism was down it engaged the mating shaft, tightened the 3 cap screws & it was done.

Somehow the 3 screws for the cover got misslaid. They're not metric threads, probably 10-32, so a trip to the local HW store is in order.

It turns out that the entire control mechanism has been disconnected from the 110 volt transformer output & a dedicated AC cord installed. Apparently this was done so a UPS could keep the controller running once it came up working. After supper I'm going to plug it in, turn the controller on & begin troubleshooting...

Anyone know if the PMAC firmware normally gets automagicly gets reloaded from the PC if the PMAC's backup battery is discharged(which I'm pretty sure is the case)?

[Verell]

I took pix of all the cable connections, then pulled the PMAC cards tonight.
It's a PMAC-Lite #602399-100 card with the companion DPRAM card. There's an EPROM with a white label that reads:
PMAC
v1.15G
1-12-95
DELTA TAU
I presume this is the permanant onboard firmware EPROM with V1.15G firmware.

The backup battery is definitely gone, not surprising as the mill was stored for about 3 years.

What is the battery's purpose? What happens when the battery dies? Is firmware or settings lost, or will the host just reload it?

I powered up the controller tonight, but not the servos & spindle. The controller is Win32 based, came up & seemed to be happy. Does this mean that the PMAC subsystem is also OK, even with the battery dead?

Tomorrow I'm going to disconect the spindle motor & connect 120V single phase to the transformer that provides 120VAC to the controller & 65VAC to the servo subsystem.

[jalessi]

Verell,

Here is some home work for you.

See attached files.

Jeff...

[vger]

I pulled this out of the manual for mine...

Option 16: Battery-Backed Parameter Memory
The contents of the main flash-backed memory (components U7, U10, and U16) of the PMAC1.5-Lite are not retained through a power-down or reset unless they have been saved to flash memory first. Option 16 provides supplemental battery-backed RAM for real-time parameter storage that is ideal for holding machine-state parameters in case of an unexpected power-down. It can only be used when the main memory is flash-backed. Option 16 provides a 16k x 24 bank of battery-backed parameter RAM in components U6, U9, U15 (smaller than the full footprint), with the battery in BT1.

Steve

[Verell]

I'm sorry, but I've already read those documents, they aren't too helpful, they assume at least a PMAC 1.5 lite, which has flash and also they assume you know what's being stored in the 'active memory'.

My board is an 1995 vintage 602399-100 which does NOT have flash memory. I've found a HDW manual for the 1992 vintage 602274-102 which I hope is pretty close to the '602399 board. The '274 manual states that it has:

ROM: 128 Kbytes EPROM for master control program

RAM: 384 Kbytes(128k 24-bit words) for active memory battery -backed-- Est. battery life 2-3 years.

EAROM: 2kBytes EEPROM for setup parameter storage.

I am NOT trying to program this card, just trying to find out what the 'active memory' is generally used for, and if it's commonly reloaded when the Windows based control program is started up & finds that the battery backed up memory has lost data.


Unfortunately, I've been unsuccessful, but am still trying to locate some Compumatic manuals that might help me.

[vger]

Ok, The PMAC is a computer in itself (albeit a very dedicated one) and has the program code that runs it stored in ROM. The code (program) in the ROM is executed when power is applied does it's houskeeping, allocating a section of active memory for it's own use and the remainder for whatever program (gcode or other) is to be run. The active memory is used for anything that may be changing while the machine is running... like...
Current postition of axies (in encoder counts), Spindle speed and wheather it is on or off, Pointer to the address in active memory containing the current block of code being executed, and lots more. Part of the active memory will be reloaded when you power on the controler. More will be filled when you transfer a program to it. And some will be altered when you home an axis. Part of the reason for the battery backup may be so that you can stop at the end of the day, turn off the machine, and come back the next day and find it still knows where it is. If the 1992 vintage board used EEPROM for parameter storage then yours probably does too, or at least something that does not require a battery to retain it's values. There are utilities to allow you to read the parameters, and a cursory look at a few will tell you if they are intact.

Steve

[Verell]

... The active memory is used for anything that may be changing while the machine is running... like...
Current postition of axies (in encoder counts), Spindle speed and wheather it is on or off, Pointer to the address in active memory containing the current block of code being executed, and lots more. Part of the active memory will be reloaded when you power on the controler. More will be filled when you transfer a program to it. And some will be altered when you home an axis.


Part of the reason for the battery backup may be so that you can stop at the end of the day, turn off the machine, and come back the next day and find it still knows where it is. If the 1992 vintage board used EEPROM for parameter storage then yours probably does too, or at least something that does not require a battery to retain it's values.

There are utilities to allow you to read the parameters, and a cursory look at a few will tell you if they are intact.

Steve

Thanks Steve,
That's a BIG help. It's what I was hoping the active memory was typically used for.

Sounds like there's a good chance that the controller + PMAC firmware together should be able to deal with the active memory not being preserved. I've got a pair of batteries on order, but am going to proceed with check-out.

Any suggestion as to the image names for those active memory parameter display utilities so I can see if they're on the controller's Win95 C: drive?

Are they likely to be DOS based as there aren't any utility icons that show up in the program manager MicroPath window with the controller program's icon.

Next step is to power up the servo system as well as the controller & see how far I can get.

BTW, my background is computer systems design, electrical & logic during the 1970s, then many years of designing/writing kernel mode code for the VMS operating system.

I know my way around computers in general. I just don't have the specific info I need to help me get going on this system.

[vger]

Verell,

Sounds like you got into computers about the same time I did. My first programing experiance was with a Bendix computer (used on the F-15 and the first Space Shuttles) using the front panel. You know... toggle in the address in binary, toggle in the instruction in binary, hit load button, address increments, toggle in the next instruction/data..... and so on.

There is one jumper on the PMAC you might look at. Jumper E51 determins where the pmac loads the parameters from. And it differs in it's setting based on card version. The NON default setting is ON for the PMAC Lite. If it's in the non default setting then it loads the factory default parameter settings from ROM rather than the machine specific settings from EEPROM.

Ok, the utilities, take a look for a DOS executable named PS.EXE . Pretty intuitive and starts you up in a tutor mode. It should let you do all the damage you want to do May be in a sub folder /bin/baldor not sure.

You may have it pretty easy, I have to reprogram everything in mine because I'm using different motors than the robotic system it came out of.

Steve

PS my card has

P1.5L
V1.16G
DELTA TAU
08-24-99

on the PROM. And no battery on board.

[Verell]

..You know... toggle in the address in binary, toggle in the instruction in binary, hit load button, address increments, toggle in the next instruction/data..... and so on...

Been There, Done that, was a DEC PDP-8 in SMU's semiconductor research lab in 1968-1969. In 1970 I graduated & went to work for DEC in their Industrial Products group working on the PDP-14 - the FIRST PLC!

...

Ok, the utilities, take a look for a DOS executable named PS.EXE .

...

P1.5L
V1.16G
DELTA TAU
08-24-99

Thanks for the E51 & PS.EXE info, will look for it this evening.

The monitor on my controller is on it's last legs, I can't read the DOS text. The controller window is usable but not exactly sharp. I've got to lug one up from the basement & find some place for it to sit untill I can replace the one in the controller.

Was surprised that your firmware was only 1 rev. higher than mine while being 5 years newer. I'm guessing that implies the firmware's been VERY stable!

[Verell]

I'm trying to ensure I understand & have cleanly disabled the 3 phase spindle control subsystem before I apply 220VAC to 2 of the 3 phases to power the controller & servos.

Things were looking pretty good, Identified 2 NO relays that both have the 3 phases in, & their outputs are connected together, however, Phases 1 & 3 of the output of one contactor are wired to Phases 3 & 1 of the other contactor. Obviously spindle rotation will be fwd or reverse depending on which contactor is selected.

There's a 4th contact on the fwd & reverse relays, and a pair of contacts sandwitched in between them. I suspect these are wired to avoid simultaneously enabling the FWD & REV contactors.

There's an adjustable overload relay built onto the output of the fwd & reverse relays. Thought it was a timer, but some research revealed it's purpose.
All well & to be expected.

Traced the switched phase wires over to a terminal block where they connect to the motor cord's 3 phases.

However, there are 3 light gage wires also connected to the terminal block. Haven't figured out where they go/what they do yet. PITA as they go into a wiring channel & I can't find where they emerge...

A wiring diagram of the spindle power subsystem would be great to have about now!!!

[Verell]

Finished the 220V hookup. Disconnected the 3 phase wires that were the input to the motor control contactors.

Took a deep breath & pushed the START button. Not much happened. Controller didn't power up... Some quick troubleshooting revealed that while I'd reconnected the controller hot line that the PO had disconnected to use a dedicated 110V line for the controller, I'd overlooked that he'd also disconnected the neutral line as well. Wire was buried right next to the terminal it had been removed from, capped off with a twist insulator.

Hooked up wire & controller PC powered up. However, servo subsystem wasn't powered up. Accidently discovered that pressing the START button a 2nd time caused the servo system to power up!!!

Tried the JOG button on all 3 axis. X-axis servo started very slowly & quietly drifting to the right. Nothing I could do changed the drift. Couldn't get Y or Z axis to respond at all(sigh). Noticed that the X & Z axis servo amps have their green 'OK' LED lit, but the Y axis amp doesn't.

These symptoms match what I was told was the behavior when I bought it, so I think I'm now back to where the seller described the system.

I tried the PMAC reset button. It got the controller Windows application unhappy, had to stop & restart it. Nothing else changed.

I'm going to have to see if the windows system has that PS.exe somewhere. It should tell me if the PC is talking to the PMAC which will be the 1st big step. Would have been nice if the PMAC had some LEDs indicating it was responding to the PC, and some other useful things such as X, Y, Z, etc. servos active...

The controller's X position display remained zeros while the servo was drifting. Either the servo's encoder isn't working, or else the signals aren't getting passed back to the PC for display.

My suspicion is that it's a problem common to all 3 axes, either the PMAC controller, or else a common power problem.

Guess I'll start by looking to see if the servo's phase encoders are producing pulses. Now just where did I put my Tektronix logic probe??? & will it still be good after 30+ years???

[Verell]

Thanks to GOOGLE, I found an instruction manual for MIT students on how to use the mill's MicroPath controller & printed it out:
http://www.media.mit.edu/physics/ped...-tutorial.html

GOOD NEWS:
RTFM always helps. In this case, there's a fairly complex (& IMHO non-intuitive) startup sequence that must be followed to get the Windows control program to come up synchronized with the PMAC machine control board.

Among other things:
"You start by setting the EMERGENCY STOP which locks down" BEFORE pressing the controller START button!!! At a certain point further on, you pull out the EMERGENCY STOP & press the START button a 2nd time!!! Bizzare to say the least!

Anyway, THINGS STARTED WORKING when I followed it!!!

At that point, the X,Y,&Z position displays showed something other than 0s & when I tried to manually change the X or Z position by hand, the servo motor gave me heavy resistance, and if I managed to overcome the resistance momentarily, the servo would immediately restore the original position.

Next I successfully followed the procedure for setting the machine's zero reference position. This involved using the JOG (manual move under power) knob to line up reference marks on the spindle (Z-axis), and also the X-axis(table left-right position). There's a LOW-MED-HIGH speed control for jogging, as well as a dial for fine adjusting within the selected range. I started with low speed jogging, then tried medium & high. Pretty impressive to see that big table zipping left & right!

Anyway, once I had X & Z zeroed, I moved to the procedure for actually using the JOG feature. You can select JOG distances of 0.0001" to 1.0000", and the table (or spindle) will move the specified distance, at the dialed in speed!!!

This tells me that an extremely high % of the control systeim is working!!!

Y is "motionally challenged"
Notice I haven't mentioned using the Y axis. This is because the Y axis servo driver's READY light is OFF. The Y axis knob can be easily moved by hand. Not a surprise as I was told about the Y axis READY problem when I bought the mill. Still, troubleshooting a single servo loop is a lot less daunting than not having any of the servos working!!!

BOTTOM LINE:
I believe that once I find & fix the Y axis READY problem, I'll have a working machine!!! Hopefully the Y axis problem will turn out to be something like a bad connection, or easily locatable component...


More good news: helixcnc sent me eMAIL today saying that he'll have a set of manuals & prints for me later this week. The set was 'too thick to send by mail'!!!

I'm having trouble figuring out how the spindle speed control works, or if the only control is fwd & reverse & the manual speed. Hopefully the prints will make this clear. Once I understand it, I'll be able to hook up the VFD/phase converter to run the spindle!!

I'm getting really psyched up to say the least.

[Verell]

Went out after supper & the specific distance eg; 1.0000" to 0.0001" jogs are not working. Left them working when I came in to fix supper.
Went completely thru the startup & setup 3-4 times & just couldn't get them to work. Jogs holding the JOG dial down do work. Go figure... I must be doing something slightly different but darned if I know what I'm skipping. It's also possible that this is the 'the controller only comes up properly every 5th or 6th time' historic problem(sigh).

Tomorrow I'm going to install the new CMOS batteries in the PMAC. Can't hurt, might help.

Also, I'm going to read up on the servo amplifiers and see why the Y axis servo amplifier's READY light is staying off.

[Verell]

I just got eMAIL from Tom saying he was able to track down copies of a full set of prints & the operator's manual for my mill!!!

This great post-sale follow-up service is an example of the service Helix Automation provides. You'll never regret buying from them, or having them service your CNC equipment!!

They're at:
http://www.helixcnc.com/