Macro commands not working over DNC

[mflux_gamblej]

I heard it from David de Caussin (now of i-t-s.com / fadalvmcparts) himself this morning that macros over DNC was a standard included software at a certain moment - but he doesn't recall specifically which version was the first to have it. He said that the -5 cpu + the software that came with it would definitely work, and that maybe older versions would have that feature too. So, it seems that the software upgrade will help. I have three software upgrade boards on the way to me right now. one or two of them are version 96, and one of the is version 101.

I didn't find a version 103.5 board for cheap yet, but it is possible that i'll bite the bullet and pay full price for one at about $400 if it is necessary. that software version is evidently compatible with the -4 cards, and came standard (minimum version) with the -5 cards.

[mflux_gamblej]

the software upgrade did it. I bought a different 1610-1 board which had software version 101 (from 2001) .... my machine got a decade's worth of thinking in control software crammed in. totally great. now if someone would just give us all the codes / procedure to unlock advanced feed forward.

I can DNC macro code just fine now and that means I can use my probe systems effectively. change fixture offsets, measure tool diameters, check for broken tools - the lot.

I did notice that over DNC it doesn't do the PRINT / INPUT commands in a great way - the messages dont really get displayed, but the control does wait for input and you have to press enter to keep going.

but, no more error messages, and no more in-action.

ran the probing cycles - used all the #V, #FX, and R statements that I wanted to, all went off without a hitch. checked out the fixture offsets after and they were updated by 0.0001" or so each time I ran the program over DNC.

[neilw20]

Great news!!

[rbjem]

Thanks for keeping us updated. Glad to know what the fix is when I want/need some macros in my programs.

now if someone would just give us all the codes / procedure to unlock advanced feed forward.

You need a board to be able to use AFF. There is no code or password to unlock it. My tech let me in on a little secret: the board for AFF is the same one that is used for a side-mount tool changer The machine he was referring to when he said this had a 32 tool chain style, not the circular, but i'm not sure if there is a difference in the electronics or not between the two. I would think not, since FADAL liked to keep things simple

So if you have a machine wtih the side-mount changer, you got yourself a free upgrade to AFF.

[mflux_gamblej]

Thanks for keeping us updated. Glad to know what the fix is when I want/need some macros in my programs.



You need a board to be able to use AFF. There is no code or password to unlock it. My tech let me in on a little secret: the board for AFF is the same one that is used for a side-mount tool changer The machine he was referring to when he said this had a 32 tool chain style, not the circular, but i'm not sure if there is a difference in the electronics or not between the two. I would think not, since FADAL liked to keep things simple

So if you have a machine wtih the side-mount changer, you got yourself a free upgrade to AFF.

So, more on AFF, looks like you need a PCB-0191 / 1470 board. this is the 'option security' board. and you would also need a 4-port CPU bus board, 1730-5 because it needs a connection to the CPU board - so this 4-port of bridge board would replace the 3-port one what is there by default.

looks like some of the fadal vendors have these available - about $1700 worth all said and done. not sure if AFF is worth that to me. I really just want to mess around with it - my parts are OK as is really.

[neilw20]

M94 is AFF. Similar to G61 and G64 in Mach3. Fancy version of exact stop and CV mode.
To achieve similar without AFF, reduce the feed rates near the corners.

[mflux_gamblej]

M94 is AFF. Similar to G61 and G64 in Mach3. Fancy version of exact stop and CV mode.
To achieve similar without AFF, reduce the feed rates near the corners.

well, it's not quite that simple. I do use a feed rate / helical ramp angle ratio look up table in my post processor, along with G8, G9 to reduce 'booming' and violent reversals when I run troichoidal tool paths. that's just modifying feeds. AFF does it differently - it will modify the servo gain using a look up table, in addition to feed rate ramping so that you don't have to change the feed rates as much during high speed machining. its pretty clever as far as I can tell.

If it turns out that I am unwilling to spend the $ on the AFF boards, i'll just tune up my post processor even more - maybe using the 'low' 'medium' 'high' M-codes for the gain - that'd be like a very crude aff with no ramps.

[neilw20]

As I said, FANCY - pretty neat stuff, and you would expect it at that price!! The wonderful world of servos.

[rbjem]

I get pretty good results with M94.1

Not a big fan of M94. I tried it on some mold finishing passes, and ended up with some gouge marks.

[mflux_gamblej]

I get pretty good results with M94.1

Not a big fan of M94. I tried it on some mold finishing passes, and ended up with some gouge marks.

that means that you have AFF I believe - fancy.

I was able to locate the boards to unlock AFF on my machine too. I'll test it out next weekend or so - may take me quite a while to work out how I want to use it. I plan to use look up tables and other things in my post processor to choose the best AFF parameters depending on what's going on.

The pcbs are: 'Option Security' PCB-0191, and '4-position CPU Bus' PCB-0189.

I couldn't tell you if the option security board is multi purpose or if it is one board = one option. It is probably the case that its multi function since there aren't other bridge boards bigger than the 4 position. so if it were the case that the option security board 'unlocks' more than AFF, then thats cool - maybe it does have something to do with the dual arm board, I have no idea.

Hopefully AFF will be a good addition to high speed tool paths. It looks like it will come in handy, and might make the machine last longer since it will smooth out the motion.

I probably won't start another thread for AFF stuff, since it will be mostly programming and stuff that's just custom to the way that I do things.

[rbjem]

Well one machine at my work does have AFF (G94.2) Ive never used it though. I know 94.1 works and then I dont have to worry if I need to run the program on a different machine.

About the boards being dual purpose: Thats what I remember my tech saying about 2 years ago but i could be wrong. Maybe it opens up more options too. He would probably know. When I see him next ill double check.

Sent from my SPH-L720 using Tapatalk 2

[mflux_gamblej]

I guess this is the Advanced Feed Forward (AFF) thread now..

Here's a picture of the 1470 'option security' board installed, with the 4-position cpu bridge board. So, it seems that AFF can only be unlocked by physically installing this board and the board isn't totally simple. it has a CPU of some form on it and it has some memory chips, so its not like you can wire in some caps or something and make it work as far as I can tell.

More to come on whether or not AFF makes any difference to me in terms of cycle times, part quality, or control usability - e.g. not getting 'helical ramp too short' type errors without reducing the feed rate to a very low value.

Of course I did the check on MDI... it used to be that if I programmed M94.2, I'd get a 'Security Violation' error - now there is no error, and I can call adder codes like M94.2 R2+12 to tell the control to use the AFF parameters for 'tool 12' in the AFF table. M95.2 cancels AFF, M91 restores servo gains to the initial values.

[mflux_gamblej]

Thanks for keeping us updated. Glad to know what the fix is when I want/need some macros in my programs.



You need a board to be able to use AFF. There is no code or password to unlock it. My tech let me in on a little secret: the board for AFF is the same one that is used for a side-mount tool changer The machine he was referring to when he said this had a 32 tool chain style, not the circular, but i'm not sure if there is a difference in the electronics or not between the two. I would think not, since FADAL liked to keep things simple

So if you have a machine wtih the side-mount changer, you got yourself a free upgrade to AFF.

I have it from a good source that the processor on the AFF / option security board does Not participate at all in the AFF calculations - it is indeed there for the dual arm (side mount) tool changer. this means that it may be possible that AFF is enabled by something more basic, like a resistor between two of the pins on the board. or maybe not? I don't know, but it seems more likely that it is something simple like that than not.

I guess after fadal got bought, the sales department went crazy thinking they'd sell AFF for $6000 - so few bought it. That's why the 'security board' though.

[pinch]

I have tested with my V94 software, DNC macros work, but as you say its having some problems with PRINT/INPUT commands, but it stops and waits for user input.
Just wanted to keep folks updated what versions that work.

[mflux_gamblej]

I have version V94.1 on my VMC-20.

I just installed probes on that system and went to run my code on it. It does not throw errors over DNC, but it does not work. It won't store anything in variables V1-V(N)

It just skips those parts! very dangerous.

It does run OK when you load the programs into memory then run them. That being said, I use probing variables, and those didn't work with my wireless spindle probe code even through the memory. So just be careful, the macros don't all work even from memory, and even more don't over DNC.

I have a version 96 that i'll put in, otherwise I'm going to just copy my version 101 eproms off of my other machine and run that.

[mflux_gamblej]

OK, so putting in that version 96 board on my (new to me) VMC-20 that I have wasn't gonna fly, since it happens to be in Dutch language! wtf!

owell. So, I wired up a lot of wires to some 0.1" PCB headers, 2 sets of 14 wires, 1 set for each side of an eprom. I then, one eprom at a time, soldered my wires onto an eprom off of my vmc 4020 / the machine donating its software, which has version 101.4. I then used a minipro programmer, from ebay / autoelectric.cn model TL866CS to read the data off of the wired up eprom.

To get another set of eproms erased, you put them in a UV light box - thats available on ebay too.

Once you have the .BIN files off the version you want, you then put them onto your erased eproms that you put in the UV light box (obviously don't erase the ones that you want to read, you need another set of eproms... or maybe your existing eproms from the machine being upgraded if you are brave).

desoldering the thru hole eproms is a hell of a chore, I have a solder sucker / vacuum desoldering workstation that sucks with a pump while it applies heat. that worked but it took a long time. don't even try if you have to do it by hand, maybe just wire up your headers that you built earlier. those will work for programming too.

after programming the 6 eproms, I soldered them back onto the 1016 board.


prior to swapping system software take pictures of the following screens:

all SETP screens
BL screen
DF
DT (if you care to)
SV (just in case, and for all axis)

then power down your machine, pop the new system software board on your cpu card and fire up the machine.

you can CS it, but don't move it around. just let it stay at CS pos, then enter the SETP menu and put in the parameters. there may be new parameters you didnt have before, dont guess!!!!! RTFM first. you can really break some sh*t if you just enter any old settings here.

once you're SETP is back how it should be, then you can CS the machine and move it around if you so desire.

at this point you should be done and should not have lost any DF DT SV BL or anything accept SETP. that being said, i read somewhere that its good practice to erase all memory and just put it all back in. maybe that was for installing internal memory though, I don't remember.

at any rate. best of luck out there. attached here is a zip file which has some BIN files in it. not sure what they might contain *wink* but there are 6 files, and they may have something to do with one O one Dot four.. or something.

btw, my machine works great and can probe just like the other 4020 machine etc. a software upgrade is a good thing WRT fadal. it is backwards compatible. go fadal! just maybe check to be sure though. I have all -4 cards in both of my machines.

[wozzwinkl]

Hi. Thanks for posting all of this really useful information about your probe set ups. I have a 1994 VMC4020, and I have wanted to get a probe system on it for a while.

I just got a great deal on a fairly complete, basically unused Renishaw probe system set up from eBay- TS27R, MP10, OMM, MI12, MI8 (I understand that I can eliminate this), SSM (Skip Sharing Module; not sure what this one is used for- sharing inputs with multiple probe types, I think?), 2 extra stylii and manuals / software. I also ordered the Fadal 1590 probe interface.

I have many questions, but my main one is this- I have a 1400-4 CPU with V94.1 software; do I need to upgrade to V96, or V99, V101.4 in order to use the macros? I guess I am not sure what the exact threshold is for macro functionality. I can get up to V103.5 from ITSCNC.com, but that will run me another $400, and I don't even know if that version is backwards compatible enough to work on my sad old machine.

I am going to try to avoid asking you questions for anything but the absolute most difficult things that come up because I know you already put a ton of time into your posts on here about the whole process.

So, thank you, and hopefully I can get this thing working.

-Jon

I still

[mflux_gamblej]

Hi. Thanks for posting all of this really useful information about your probe set ups. I have a 1994 VMC4020, and I have wanted to get a probe system on it for a while.

I just got a great deal on a fairly complete, basically unused Renishaw probe system set up from eBay- TS27R, MP10, OMM, MI12, MI8 (I understand that I can eliminate this), SSM (Skip Sharing Module; not sure what this one is used for- sharing inputs with multiple probe types, I think?), 2 extra stylii and manuals / software. I also ordered the Fadal 1590 probe interface.

I have many questions, but my main one is this- I have a 1400-4 CPU with V94.1 software; do I need to upgrade to V96, or V99, V101.4 in order to use the macros? I guess I am not sure what the exact threshold is for macro functionality. I can get up to V103.5 from ITSCNC.com, but that will run me another $400, and I don't even know if that version is backwards compatible enough to work on my sad old machine.

I am going to try to avoid asking you questions for anything but the absolute most difficult things that come up because I know you already put a ton of time into your posts on here about the whole process.

So, thank you, and hopefully I can get this thing working.

-Jon

I still

yeah, u need newer software. i just did a probe install on my vmc-20, which had v94.1 - that didn't work. it actually was a very bad thing with that particular version because it didn't give errors, it just skipped the code and executed the next line. holy crap! that nearly crashed my tool setter.

I know that I use v101.4 because thats the newest i have, and it works fine. read below on that...

as far as the skip sharing thing - that refers to the old fanuc nomenclature. G31 is called a skip step or something like that, and it uses the probe input as an interrupt to stop the motion. its the same on the fadals.

sorry to let u know u may need to spend more money, but another few hundred is WELL WORTH it to have probes.

otherwise, i bet you can find some cheaper ways to upgrade your software? you need eeproms, or just use the ones u already have for example. thats enuf info.

[wozzwinkl]

OK, that is helpful to know. I will go ahead and get the software upgrade. It is worth the money, I just didn't want to spend it if it wasn't necessary.

On that note, do you think the V103.5 software will work? I suppose if it doesn't I can always reprogram it to V101.4....

Thanks again! I'm sure I will have more questions once all the hardware shows up!

-Jon

[wozzwinkl]

Well, V103.5 officially works on my 1994 VMC4020. Boy are there a lot more options...

One note about transferring info; I am fairly sure that the axis survey data are stored on the axis controller cards themselves. I accidentally zeroed my Y-axis survey trying to write it down (don't ask...), and after the new software install, I had Y-axis survey data again. Pretty glad for that, though I doubt it would have been that big of a loss.

-Jon