What's new
What's new

Options on my D11 now that I got SPS: 'Selbsttest gestoppt: Zu viele Fehler'?

Vitran

Aluminum
Joined
Mar 31, 2016
Deckel FP50/CC

The move from shop 1 to shop 2 was good. I had a great time learning how to run the machine and learning the D11 interface. Extra care was given to the PLC box to make sure it was handled carefully. The machine was unpowered for about 1 month and put back together in the new shop. All electrical wires are connected.

Now that it is has power and is on (along with my Hardinge Superslant!) I am getting a dreaded SPS: 0000 Selbsttest gestoppt: Zu viele Fehler or Selftest Stopped: Too many errors. From looking around, Wiggo is the only person who as done something like this (not 100% up to speed on that).

My current plan is to:
1. Look inside the PLC to see if there is any obvious damage. Maybe a loose wire or blow cap or something like that would be a miracle. Sometimes take apart and put back together works (please be so).
2. Check all batteries and see if they are dead. Maybe the loss of power caused it to lose BIOS info. I have a copy of the settings, but that is all. May need to flash the BIOS, which would also be scary. Not sure how to do that either.
3. Last plan of action. MESA card with LinuxCNC and a PC. I have experience getting home built 3D printers, and lunux cnc on a laser cutter, and two XYZ cnc router tables, but I know that three stepper motors and end stop buttons are far easier than a Deckel. I am leaving my current work so will have much more time in the shop so I could do it. Spend a few months testing each wire, finding the IOs, reading the manuals, writing scripts to change tools, speed, spindles, ect. Is or has anyone else done this? Also, is there a value in a broken D11 PLC?

Another note on this machine. There are new servos and servo drivers installed in 2002 I believe from the notes I have.


Does anyone have any first pointers of what I should do in this situation?

Also, dumb side note, the Hardinge Superslant that I bought from my work now says 'X axis locked' and wont let me start up the pump. The jerbs I have to do to get things running are piling up.:sulk:
 
Deckel FP50/CC


My current plan is to:
1. Look inside the PLC to see if there is any obvious damage. Maybe a loose wire or blow cap or something like that would be a miracle. Sometimes take apart and put back together works (please be so).
2. Check all batteries and see if they are dead. Maybe the loss of power caused it to lose BIOS info. I have a copy of the settings, but that is all. May need to flash the BIOS, which would also be scary. Not sure how to do that either.
3. Last plan of action. MESA card with LinuxCNC and a PC. I have experience getting home built 3D printers, and lunux cnc on a laser cutter, and two XYZ cnc router tables, but I know that three stepper motors and end stop buttons are far easier than a Deckel. I am leaving my current work so will have much more time in the shop so I could do it. Spend a few months testing each wire, finding the IOs, reading the manuals, writing scripts to change tools, speed, spindles, ect. Is or has anyone else done this? Also, is there a value in a broken D11 PLC?

Another note on this machine. There are new servos and servo drivers installed in 2002 I believe from the notes I have.


Does anyone have any first pointers of what I should do in this situation?

Also, dumb side note, the Hardinge Superslant that I bought from my work now says 'X axis locked' and wont let me start up the pump. The jerbs I have to do to get things running are piling up.:sulk:
There is a 3 battery pack in the SPS module (the one all the way to right)....if voltage gets low there that will cause hangup in boot sequence. Although these were NICad rechargeable in my experience regular Akaline AA batteries work fine. Note the first time you charge or change batteries you will still get the error...so you have to shut down and boot up twice basically before error clears.

You option no. 3 is insane btw,,,,ok for a Bridgeport type mill but for Deckel way too many complications. People come on here for years talking of doing such to a Deckel but they never actually do it.
 
We've had that error a number of times on our D11 and most of the time it was just the battery. If the main red switch is left off for some period of weeks or months it gets low. Leaving the red switch on for a number of hours gives it time to recharge up to an operable level again and you're back in action. I don't know but it appears the battery isn't just a backup, it has to have a decent charge in order for the system to function. It's constantly being charged while the red switch is on so it will last a dozen years or more no problem. I leave the red switch on all the time now, using only the grey switch for daily shutdown, and never have that error any more.

Nicad cells are 1.2 volts and alkaline are 1.5V, with a different discharge characteristic, and dangerous to be recharging. I would not recommend switching types.

Dan
 
Nicad cells are 1.2 volts and alkaline are 1.5V, with a different discharge characteristic, and dangerous to be recharging. I would not recommend switching types.

Dan
To be absolutely safe long term, I agree, but for short term "getter done" situations, I've never had any trouble substituting 3 alkalines in that battery pack.

And boy have I ever had D11 troubles...some of the problems so strange no one would believe me....but they were all prior to my ownership of the machines. No troubles on my watch after repairs were made.
 
I checked the battery pack on the PLC. It is a 3.6V Ni-mh recharable with 800mAh. It has a voltage of 0.8V. I thought the machine was supposed to charge those?

Off to the battery store I go. Any chance that the battery being dead will have caused all the settings to go away?
 
Any chance that the battery being dead will have caused all the settings to go away?
No, you will be fine....the parameters are on another board (I forget which module) with rectangular SAFT battery soldered to the board. Hopefully you have the parameters backed up on electronic media or laptop in case that one ever gets weak.

Remember if you still get the boot up error with new batteries, turn machine off and try again as it sometimes takes two boot up attempts to clear the error.
 
I got the new battery and no change, still faulting. I turned off the PLC off and on a few times.

I did find something related and odd. One of the boards was physically protruding out from the PLC rack about 1 cm and out of its hold at the back. I turned the PLC off, slid it home, screwed it in place (most of the boards were not screwed in place) but no change on the screen.

The board that was disconnected was the one on the right with the reset button and stack of LED's (not my photo)
http://www.practicalmachinist.com/v...g-11-hangs-after-cmos-checksum-ok-afb0310.jpg

What options do I have now to go forward? Is there another battery or something else?
 
I got the new battery and no change, still faulting. I turned off the PLC off and on a few times.

I did find something related and odd. One of the boards was physically protruding out from the PLC rack about 1 cm and out of its hold at the back. I turned the PLC off, slid it home, screwed it in place (most of the boards were not screwed in place) but no change on the screen.

The board that was disconnected was the one on the right with the reset button and stack of LED's (not my photo)
http://www.practicalmachinist.com/v...g-11-hangs-after-cmos-checksum-ok-afb0310.jpg

What options do I have now to go forward? Is there another battery or something else?
You've probably already done this, but make absolutely sure you have aprox. 3.5 volts where the battery pack wires are soldered to the board. If you do, next step is to reseat or apply Stabilant to all socketed IC legs..

If that doesn't work you may be in for quite a project without a second machine to swap modules with. Even then it can be bad.....I once had similar happen, narrowed it down to one of he 20 static ram IC's gone bad....each IC had 20 legs... found new replacements via the obsolete market....bought a used IC tester as I wanted to know which one was bad, hoping it would be one of the first ones of the 20.

Naturally it was one of the last ones*....so I went ahead and replaced all 20 of them. 20 x 20= 400 desoldering**.. (and 400 resoldering of new sockets...so I could pull bad ones out in the future by hand and also not worry about putting too much heat on the new IC's). and every desolder had to be nearly perfect to remove an IC intact without damaging a board trace. But it worked....IC no. 17 was bad according to my tester....and sure enough that fixed the machine...went from doing nothing (hanging up at SPS boot) to working perfectly.

===============

*Keeping in mind, for all I knew more than one was bad....but if one of the first ones was bad obviously I would have stopped there to test the board in the machine to see.

**This was done at home in office at night in relaxed situation over a couple of weeks.... desoldering just 2 or 3 at a time....you'd go nuts trying to do all at once....unless you lucked out and found the bad one right away.

Model 57A, Linear IC Tester - B&K Precision
 
Milacron,

Is there some reading I can get to figure out what may be wrong and how to narrow it down?

This is an assumption that it was one of the static ram chips. I have used an IC tester for some circuits my dad makes, but I do not know what type of chip we are dealing with or what I am testing it against. Do you have some reading on what tester and/or the chips I should look at?
 
Naturally it was one of the last ones*....so I went ahead and replaced all 20 of them. 20 x 20= 400 desoldering**..

**This was done at home in office at night in relaxed situation over a couple of weeks.... desoldering just 2 or 3 at a time....you'd go nuts trying to do all at once.

You just need the right tools to make this easy, for example a Hakko 850 or the poor-man's version, a Weller 16SWG. (In a pinch you can easily make one of these on your mill, I suggest attaching it to a soldering gun rather than a soldering iron.)
 
I love soldering but figuring out ICs has always been my hurdle. I'm wondering about what chips to remove, where they are, what they do, ect. I have kept the PLC as a black box that works and don't touch what works.
 
Milacron,

Is there some reading I can get to figure out what may be wrong and how to narrow it down?

This is an assumption that it was one of the static ram chips. I have used an IC tester for some circuits my dad makes, but I do not know what type of chip we are dealing with or what I am testing it against. Do you have some reading on what tester and/or the chips I should look at?
I can't find my old thread on the subject...possible I deleted it for fear whoever bought the machine might find the thread and worry unnecessarily. Sorry but I can't even remember which module the 20 static ram chips are in. The BK IC testers have a "library" in memory of how an IC should react to various tests and this IC was in the library of the tester.

But you really really need another (working) D11 machine with same software version to make absolutely sure which board is bad, is that is the case.

Ballen, I used a Hakko desolder system with built in vacuum. It was critical to have exactly the right size tip and to be sure it was always clear and clean...plus a technique of wiggling the leg back and forth during vacuum. What you show looks faster but does it leave proper holes for resolder of the new IC inserts ?

fr-4102.jpg
 
I remember now, the problematic static ram IC's are in the NPC module.

65445d1356211367-dialog-11-hangs-after-cmos-checksum-ok-afb0310.jpg


npc2.jpg


Middle group of IC's, lower board. Except looks like I was having a false memory on the number of them... I count 18 in this photo, not 20. I remember now I purchased 20 new old stock ones, but only needed 18 (only needed 1 really...but replaced them all anyway) Was right about number of legs per IC however...20 each.

I'm thinking Wiggo had similar problem, got NPC module from DMG in Europe with option to return, that fixed his machine, but he decided to return it anyway due to cost or whatever, and maybe even his fantasies of re-making the entire D11 boards ! :nutter: But at least that confirmed where the problem was.
 
No good news really.

I have been digging through the PLC and found the other battery on one of the other cards and it is at full 4VDC power. I also found the chips and a source. The chip:

Mhs
hm3-65262c-5
8905

I found two sources of the chips!
Aliexpress.com : Buy HM3 65262C 5 DIP HOT OFFER IC from Reliable ic 73 suppliers on Professional Group Technology
HM3/65262/5 : MATRA HARRIS IC Memory on Electols

I also found something interesting in the PLC. There are patch jobs to the boards. Some of the boards have wires soldered onto them, one spot has a pair of resistors linked in series as the person didn't have the right resistor, a few chips on some of the boards had wires soldered to their leads to link to other parts of the board. One 8 pin chip had four resistors linked to the leads and linked in a X on top of the chip itself. This was all done by someone not at Deckel it seems. Someone who knew what they were doing but did not have the best of skills at soldering did this.

Lastly, calling out to Wiggo likely, I need need a copy of the settings for a Deckel FP 50 CC/T. Failing that, does anyone know where I can find how to set up the config file? The machine came with four floppies that were soaked in oil so I doubt I can read them now.
 
Good news! Cleaned all the boards with isopropyl alcohol and the machine boots up! Yay!

Bad news, it says it lost parameters. Anyone have a copy of the parameters and/or anyone have info on how I can write them? The machine did come with floppy disks but they were dripping in oil. I have not tested them, but I don't really want to test them either.
 
Good news! Cleaned all the boards with isopropyl alcohol and the machine boots up! Yay!

Bad news, it says it lost parameters. Anyone have a copy of the parameters and/or anyone have info on how I can write them? The machine did come with floppy disks but they were dripping in oil. I have not tested them, but I don't really want to test them either.
What, you just hosed them down with alcohol from the garden hose ? But seriously, you cleaned the contacts where the modules insert, the socketed IC legs or both ?

i probably still have D11 parameters on my old IBM Thinkpad but mine was an FP2A and probably a different software version so might not do you any good. I'd start a topic in the General Form about how to de-oil your floppies. Also if they are original you may need a computer of close vintage to read them.

As an aside, the oily floppies sounds like something that would have happened to me, since it seemed like every D11 machine I ever owned came from previous owners that were either insane or complete idiots..LOL...but amazingly I never had that particular problem at least.
 
I hosed it down. Took a box of Q tips, eyedropper, and a bottle of 99% isopropyl and washed down the boards. The cotton had yellow tinge afterwards of material that was removed. I didn't to anything to the board contacts. I did have changes occur on the PLC as I was replacing the boards and booting up. First board I cleaned it booted up to the main screen then had a fault that stopped the whole system. Resetting the machine brought me back to the old settings. Cleaning other boards were having interesting effects on the errors. The SPS error went away to be replaced with another. I would clean a board, place it in the warm of the heater for half an hour, then insert it.

I knew something was happening while cleaning them. Why cleaning the boards does anything is still kind of a mystery to me.

As for the oil on the floppies, I have read that isopropyl alcohol cleans oil off of floppies. I'll get a bottle of 99.993% that I have at my dad's house for that.
 
As for the oil on the floppies, I have read that isopropyl alcohol cleans oil off of floppies. I'll get a bottle of 99.993% that I have at my dad's house for that.
Interesting...you may be the first member here to clean the entire boards. You probably lost parameters when you removed the module with the SAFT rectangular battery.

Is there any printing on the floppy labels (assuming they even have labels) to indicate their age ? My FP2A was 1989 with original floppies and even my relatively old Windows XP computer could not retrieve the data. But my even older IBM Thinkpad w/ Windows 95 worked like a charm to read the disc. I'm not sure it's a function of the Windows version so much as the disc drive itself.

Deckel was at least "on the cutting edge" even having parameter floppies back then.... CNC Mahos were still using punched tape even as late as 1991 !

You of course need a DNC software program for retrieving the disc data into computer memory and for parameter upload* to the D11 control... OneCNC used to have one you could download for free.....I think this is it -

CAD/CAM CNC | NCLink Registration | OneCNC CAD/CAM Software


=============


*You could of course print them out and enter each one by hand at the D11 keyboard...but it would be very tedious, subject to making an error, and takes hours. Upload via laptop ensures no errors and takes maybe twenty seconds.

I have entered Maho "constants" as they called them, many times by hand over the years but the Deckel D11 control has many more of them as I recall, so would take much longer.
 
I'll certainly try that other DNC program.

When I first got the machining running I had it using the free version of NCnet Lite from CadEM which worked quite well.

At work I was told by the guy who fixed our machine that Heidenhian releases a free DNC called TNC Cremo. It can communicate with the PLC as if the computer is an external hard drive which is super nice. The Hedienhian 151C we have at work can hold a max of 32 programs and about 3,000 blocks (I think), so having more programming space is super nice. It also does standard DNC communication, but I have not set it up for that yet. It should be possible to link to a Dialog, but I haven't tried it yet.

TNCremo - PC Software

I do believe I did download the settings at one time when we first got the machine up and running with NCnet Lite. I think we download the parameters and had that running before we figured out how to upload/download programs to the machine. Being the fool I am, I didn't think that was important at the time as the machine was running. And yes, I have searched my hard drive like mad but I keep a clean machine. It an't there.

Lastly, the machine was first purchased in 1991 for $325,000 USD, so the disks are newer than that.
 
Haza! Got the files off of the floppies and it is labeled as "Parameters"

We have a 1.2MB of data and on the outside of the diskette it says UC50SK03.D11 is the parameter file.

Of the windows we found when the machine is on RUN you can change the language to English. If I recall correctly in the SPS mode we could upload the data. We got the program to transfer to the machine, but it was in the wrong name format. We tried a few other files (there are about 30 files) on the disk. Some of them seemed more accurate as to what the parameter files should be. We found that the machine only accepted files with a *.IO file format, and even then, did not accept the files.

What am I missing in order to upload the parameters to the PLC?
 








 
Back
Top