FP*NC control update project, again.

Hi Folks, I thought bounce the idea of starting a FP*NC machine/control interface thread up. The idea is to end up with a resonably specific idea what will be required of a new control to run a FP*NC.

It would be great to have sub-forum for this but perhaps formating the subject line would work, like "FP*NC CONTROL-<topic>. (speed, hydro, scale...)

Here are some starter topics if anyone is interested:

*does the gundig/siemens machine interface differ?
*speed changing system.
*hydralic system control
*scale interface
*servo drives

-TomP. aka Dr. Frankendeckel

tomp:
I will give you my thoughts here. Don't think a sub-forum is in the cards. Don has been very cool to having too many different forum's on the site!
For me i think putting this in the Deckel group is fine, like you have done.
I am very interested in this as a project. Let me outline my personal prefferences:
I have looked at the control market and for my money i am inclined to choose the Heidenhain TNC 310. This is more expensieve than some but it has the following advantages:
Can use the output of the linear scales directly, no interface necessary just cables.
Has integral PLC that is very powerful (1700 PLC commands, and 23 assignable imputs and 23 asssignable outputs.
Operator interface is very similar to the existing Dialog4.....Motion control of axis moves with arranged push buttons just like the original Deckel setup. This is very important to me because i use my machine for proto type and repair work.
Control will opreate the original Bosch servo amps without conversion, so that original servo's could be used. (the Siemans servos are reputed to be very good).
The Heidenhain control is small and self contained... just needs an enclosure to cover the back.
Issues for any conversion:
Largest issue is the gear shift. Programming the PLC to change the gears on one of these machines is a real challenge.....Sean has stated that the gearshift is accomplished via a seperate board. I think any control retro-fit should address the shifting. Maybe an inverter drive with a 2 speed chamge to give good low end torque. ( i need my low end)
Second largest issue: The servo drives. This is not a problem of how, but money. You can get amplifiers made that will drive the Semens servos, you just have to pay.
I am not looking at this as the cheap hobby fix, so maybe i am out of step here. I don't need this to be dirt cheap, just reliable, and functional without jumping through hoops or making the machine less friendly to use.
Please keep me posted on your thoughts either here or through my e-mail.
Cheers Ross
Has

Well, I was thinking of seperating the job of control replacement into two sections.
1st is the machine interface, like scales,
gear change and such. 2cnd is discussions of specific controls.
On the gear change issue I know the my siemens 3m FP3NC does all the work. It drives a relay for each motor and senses the eight micro switches along with jogging the motor.
I have already replaced all the motor control relays resistors and such with a mitsubishi VFD. The control outputs jog, low, high and direction. The VFD signals "motor running" wich looked for by the control. (bad to unclamp if motor is spinning!).
Whats the deal on the scales? Arent they just differential quadrature signals?

Servo amps: They are old school analogue units. Modern amps like AMCs analogue units are every bit as good, are very quiet run cool and have lots of protection circutry. I was working on a droppin replacement panel for the bosch amps using AMC amps when my control flaked out.

For what its worth most new controls actually do speed the machine up some because they know when a tight move is comming up and reduce the feedrate enough to maintain acceptable following error. This allows a higher average feedrate.

Lastly, I wonder if the machine itself differ depending whether its dialog or sinumerik equipped? Do you have the wiring diagrams for the dialog equipped machine?
-Tom Phillips.

Tom:
First let me qualify my responsed. First i am not a "Wire Head" but i do understand electronics if i get enough information.
Yes i have the full electronic/electrical documentation for the Dialog4 control. Problem is that it is only in German.
I believe the machine side is athe same for either Dialog or 3M.
Your work on the gear change is very interesting, i would like to know more..maybe e-mail me the details.
The Heidenhain scales produce a sine wave i believe. For some controls that is a problem and needs to be converted to a digital signal. (i think).
I would like to know more about the servo amp solution you are working on. Like i said i am not afraid to pay for getting this machine to be more reliable in the future. My one hope is that a newer control could make 3-D contouring more efficient. With the Dialog the processing speed is slow enough that with many point to point moves the effective feed rate realy drops.
Cheers Ross

OK...hold on a sec...

I have both Dialog and Siemens controls.

Before you guys get along too far, it is important to address that Ross and Tom should be looking at conversions from two different places...

Here's why...

Tom, since you've never had a Grundig (dialog) controlled Deckel, you have to realize that symantics aside, you already have a "converted" machine.
The 3M is a Generic control...or I should say "universal".
Those Heidenhain converter boxes you have from your scale outputs can be considered "adapters"...the Dialog/Bosch proprietary system doesn't use them.

The conversion that Ross is speaking of is...in essence...converting to something similar to what you already have Tom...that is...cheaper parts...a little more capability...a little more mainstream.
Although his choice is a HH 310, he is pondering to get where you more or less already are.

Now, the Dialog control is a special control...working on it has some charms and Ross would like to get as close to that as possible...which the 3M isn't, but that is a matter of preference.

I also prefer the Dialog, but I purchased my 3M because it is already 75% "converted" and I figured I would change out the front end to a later model Siemens (810?).
Since the Siemens system is already "universal" and is programmed/adapted to each application, I figured it would not be a wild step to program a more modern front end to work seemlessly with the amps(etc) already installed by the factory with the 3M...especially since the critical factory parameters could be transferred directly.

I cannot see any reason why you would start replacing amps and the machine side of the control system away from what you have since it is old enough that used parts can be had, popular enough that used parts can be found, but new enough that it can already be programmed to tell the machine to do things that the machine cannot physically do...and new enough that it should still integrate with an even newer front end that has more bells and whistles.

Although I have not done it yet, I will have to guess that a direct Siemens/Siemens *upgrade* is the way to go and not a retrofit from scratch.
Anyway, that is why I have one.

Now, for the gear changes...

Ross, I am guessing about the proprietary board taking care of the gear changes and simply reporting "completed" to the control.
There is of course an upside and downside if this is true.
The upside is that obviously that board can interface with a "universal" control with no problem (since it's done on the 3M), and very little programming would be required.
The downside is that you have to keep a proprietary board with your retrofit...which could eventually fail and cost.
If this chore is done by the control itself being harwired into umpteen relays...well there's a bright side to that to...it would already have been done again on the 3M and would have a known set of parameters to copy/translate into whatever control you needed as well as a known wiring scheme.

Ok, so point is that the 3M is a conversion for all intensive purposes. There is nothing I can find that is proprietary about it and the Deckel...just programming and wiring. So it has been done and can be used as an upgrade platform for someone like myself or Tom, and as a model for someone like Ross.

Thoughts/Comments?

Sean

Maybe it is deep in the group and I just haven't read it (i.e. sorry about the possibly redundant request), but can someone explain the complexities of the gear shift. Ultimately it must be controlled by a computer already and hence has isolated (and computer compatible) control inputs. The PLC should be used for manipulating this. Not having looked at a machine for retrofit, however, I fully recognize that my question may just be plain foolish. Still, I keep hearing the comment but no real description.

Maybe I really do need to schedule a trip to come and see you Ross. This might be what gets me off my rear!

Thanks,
Alan

I can explain the gear change on the siemens equipped machine. Its run entirely by PLC in the 3M. Each of the three motors is controlled by relay directly from the siemens control. Nine microswitches, three for each motor to detect shaft positions. For any given speed the contol knows what positions (switches) are needed. It activates motors one at a time I think until the desired switch activates. The motors rotate in one direction so that simplyfies things. The motor runs on reduced voltage so if the gear "hangs" it wont break anything. The controls also jogs the main motor back and forth to facilitate gear engagement. I have the chart showing what switches correspond to each speed. The VFD I installed works well with the gear change system and replaced a ton of relay logic in the cabinet. Note that the last gear motor only has two positions so one switch is unused.

So Sean, do the heidenhain boxes output conventional logic level quadrature signals to the siemens? Thats a major dark area for me.

On the servo amps, no reason to change them if your happy with them. Mine are noisy, I can hear them singing and growling a bit.They also put out a fair amount of heat under load. New amps are silent, cool running, very small and still take the same input signals. They wont improve the machine response of course but they are probably cheaper then getting a bosch board replaced. They do have to be setup for the servo as do the bosch units of course.

Tomp,
Based on your description, I don't quite see why the speed change should be such a big deal. If the PLC is halfway decent, it should be capable of programming in the information fine..... But maybe I'm missing something. (Am curious about your being able to use a VFD tough... Usually VFDs don't like loads being switched on and off...)

As for the Heidenhain... I guess it depends on the exact model, but typically the Heidenhains are *not* phase quadrature. They are resolver based (basically two sine waves that are 90 degrees out of phase with one another). Theoretically more accurate than a phase quadrature, more of a pain to deal with.

--Alan

Ok, I'm lost...
Tom, are you saying that your 3M uses Bosch servo amps?

My 3M is completely Siemens.

Sean

just a few notes to make the information about the gear change accurate. First the small gear change motors (3) aare controlled by teh control as stated. They have power applied when the selector shaft is needed to be shifted. While the power is applied to the shift motors the main motor and brake is pulsed at reduced voltage to allow small movements of the spindle.
The motor jogs for a discrete number of pulses, and then reverses direction and pulses some more . It will reverse a third time if the gear is not engaged on the first reversal. While this is happening the gear change motors also change direction, but they do not do this in direct correlation with the main drive. The change motor will only change in direction once before the whole thing times out and generates a fault.
On the later Dialog4 controls the gear shift motors run together at the same time. Ealier Dialog4 units run them one at a time.
You might not ever see the reversal of the shift motors if your machine is one that shifts quickly. On mine i had trouble with the shift giving a fault for nor shifting so i gor a chance to see all this work for the full cycle at times. The cure by the way was to clean all the contacs of all the main motor contactors. The jog voltage is low enough that if the contacs are dirty the main motor will not jog properly.
Tomp: I would like to know how you use a VFD to accomplish these tasks.
So now for my education: I see that the Heidenhain (just an example) has imput for 23 switchable lines. Now with the programming in the control does that mean that you need 1 imput for each micro switch contac (5) Travel limit switches (4 pre axis min)= (12) Low lube oil (1) and possibly (3) tacho imputs? Is there anything else that you need to tell the control? am i missing somthing
Cheers Ross

Hey Alan:
Why don't we arrange a weekend field trip. I would love to meet you. You just name the time and i'll be there. Bring your wife if you like, there is a very nice shopping center with good lunch spots and a Nordstroms just down from the shop....We are not far from Muir Beach if that is your fancy, or Mt Tam for a picnic.
Cheers Ross

[This message has been edited by AlfaGTA (edited 08-15-2004).]

Is that 23 line of input OR output? Remember that you have to control the switching motors (output). I assume you also have some microswitches on each travel of the gears (2 each for 6 total, inputs). Limit switches, as you indicated (inputs). Plus other stuff you've probably forgotten about.

The tachs would not go into the PLC. They should already be going back to the servo amps (right?). Encoders probably also have special places where they go. If not, that could eat up a good number of signals (3 each, input).

Yes, we definitely need to meet. If I come up to Muir Beach, however, I'm going to have to go there with the kids. We have a very close family and they don't like exploring without dad. Nordstroms... my wife will like that. Then again, she might rationalize it and discover that the Nordstrom's at Stanford is closer. If she is going to spend 2 hours in a car, plus a few hours shopping, why not just spend all that shopping locally!!!

Alan

Ok, now Im confused about the scales. Are the heidenhain scales on the FP*NC resolver based or optical? Resolvers need excitation dont they? Sean, do you know if that is what the box on the siemens does, convert analogue resolver to ttl level? The reason I ask is that resolver interfaces are usually an extra option on most controls Ive seen.

Sean, my seimens control has bosch servo amps, model TR20 I believe. Thats one of the nice features of the old analogue velocity mode amps we all have. They are fairly generic in application, as long as it has enough voltage/current.

Ross on the VFD fittment, the VFD itself is always on if the system is on. The pulsing/reversal is controlled via input signals to the VFD. The siemens outputs
jog, low, high and direction from opto-isolation modules. The VFD has jog, speed1, speed2, and direction so it was fairly easy to connect. Also, on my machine there is a little box under the relays whose job it is to monitor the spindle motor power lines and indicate if the motor is running.Not sure how it does that but the VFD has a running signal I used in place of that box. There are a few other details like braking I can provide. How do you upload jpgs, I can post a picture.

On the I/O stuff, most of the controls Ive been looking at seperate out the limit/home switches and encoders from the general PLC logic. Is there anything related to the hydralic system that needs to be included on the i/o count?

On the i/o count dont forget low lube signal.
There are also some circuts that tell the control if various magnetic breakers trip.

Tom, yes pictures would help.
There is a FAQ section which describes the process for posting.

You are asking the wrong person (me) about those scales as well.
I know that they are analog and work directly with the Grundig/Dialog control, but on my 3M they are converted via 3 little Heidenhain inline boxes...I've assumed to standard digital quadrature.

I'm surprised your machine has a 3M control with Bosch amps. I think the TR20 was the standard Bosch unit in Dialog run machines.

Most people want to get away from the Bosch drives since they are very pricey and hard to come by.
My 3M has (unless I've lost my mind), all Siemens drives/power..etc. I cannot open the cabinet right now so I cannot check, but it is something along the lines of the S7 series. The installation looks factory to me.
I'll be unpacking and hooking up my 3NC soon as it's high on my list of things to do (gotta get power to it since I'm afraid of the parameter backup battery going south).

It is definetely interesting that we now have controls in configurations of:

Grundig/Bosch
Siemens/Bosch
Siemens/Siemens
(dynapaths and hurcos may be a different critter).

We'll all have to keep at this and maybe we can actually figure out a nice clean upgrade/retrofit path.

Sean

The FP2NC with Dynapath Delta 20 that I have has Bosch Servo amp and HH scales. Not sure if has the inline converters for the scales that Sean mentions.

Sean:
according to Matt the FP2NC he has : Delta 20 control, with Bosch amps and Siemas servos.
Alan:
"Is that 23 line of input OR output? Remember that you have to control the switching motors (output). I assume you also have some microswitches on each travel of the gears (2 each for 6 total, inputs). Limit switches, as you indicated (inputs). Plus other stuff you've probably forgotten about."

That is 23 imputs adn 23 outputs from the PLC. The imputs for the spindle and the imputs for the scales alomg with the 24 VDC are additional.

[This message has been edited by AlfaGTA (edited 08-16-2004).]

Ross, would you be trying to keep your toolchanging pendant? There's a couple more.

Hmmm...toolchanging pendant... glad you are a patient man .

Sean

Well i haden't thought about the tool pendant in the retro fit....i haven't forgotten about the other one, but i am just now getting the broken support casting finished and painted..hope to put it on the machine tonight maybe.
Cheers Ross

While you're doing your I/O counts, here is one thing you should be aware of. Not rocket science, but somthing you should think about... Don't add any type of E-stop into the I/O as the sole source of stopping the machine. You can have it in there for monitoring purposes, but an e-stop chain should directly remove power from the servo and spindle relays. That would be a certified installation on a new machine. Obviously this is a safety thing!

--Alan