What's new
What's new

program stalls and the machine spins all night?

kustomizer

Diamond
Joined
Aug 17, 2007
Location
North Fork Idaho
I have several of these Haas VMC's and one of them will do this on a very rare occasion. At the end of the day we will often times let the machine run after we leave, we set the " auto power off timer" in setting #1 to a value of 2 min. Once in a great while perhaps 2 or 3 times in a year, I will come in in the AM and the screen says "waiting" the spindle is spinning and the coolant is pumping and the program has stoped. If I push cycle start nothing happens, if I push feed hold it will function as if in single block, if I push position then program it will take off and finish the program. I find if I unplug the comm (RS232) wire off it will not do it. It thinks it is recieving something even though nothing is being sent? Haas has updateded the software twice. The machine is 5 years old now. Day before yesterday it ran all night again only this time the chip auger ran all night too and it wont start now unless I put a pipe wrench on the auger and help it start.
I called the Haas boys and asked if they had a fix for the auger motor or if I needed to replace it, he asked what happened to it and I explained, he is sending a new processor, auger motor and a service guy to change them, I reminded him the machine was 5 years old and he said it was not going to break them and they would take care of it.
So after all that wind, has anybody had the comm wire cause the program to stall? was there a cure?
thanks
mark
 
Last edited:
First, what do you mean by comm wire? Is this a single wire to pin 7 (signal common), or the entire RS232 cable which contains several individual wires?

Recently, meaning later than 2000, at some point HAAS added the capability to input serial data during CNC program execution. If your machine has this capability, then it is possible that noise thru the RS232 port caused the program to stop.

More likely it is the HAAS processor board, and maybe stimulated by a power disturbance.

Our I232 Isolator System might be of value to you.

.
 
If I unplug the entire rs232 is has never had an episode, our wire goes to a preditor hub, and is a pain to have to remember to unplug it and plug it back in to use it. The possobility of the power disturbance only after quitting time seems odd, it has never done it during business hours ( perhaps it only works 8 hours per day) unless it has to do with the hub losing power as it goes out with the lights. tell me more about this isolator thingy.
 
You may have the key right there....when the hub looses power, it could be occasionally sending a voltage spike or other disturbance down the cable. I suggest taking it off the light switch circuit and leaving it powered up.
 
kustomizer:

I think Tonytn36 has probably indicated the reason. I do not think it is a spike, meaning a large voltage, but maybe just the logical change of state occurring when power is removed. If that is the case, then the time of power off should be when the machine would hang. But you probably would not know it hung because the spindle is still running.

If the disturbance did not occur at this time, but later in the night, then it could result from electrical noise coupling into the cable more easily when the "hub" is off and the data line is in a high impedance state and near 0 V, as compared to when the "hub" is on and the typical rest state of -10 V.

Noise coupling thru the RS232 system while the machine is running is a problem in a Fadal because of the always on to receive data capability. This may be true for HAAS as well on newer machines.

For information on our I232 System visit www.beta-a2.com .

.
 
The big thing that has me wondering at this point is why one machine out of 8 and always the same machine, and why would it care if there was something going on in the hub or cable when it is running, it is not in DNC mode? this machine is 5 years old, I have 5 machines older and 2 newer ones and none of them have ever done this, perhaps it is just one of those things that make you wonder.
 
just a thought

do you lose air pressure on these occaisions? my fadal does strange things like this once in a while when it loses air. the results are upredictable at best.
 
Actualy about an hour after we quit work the compressors turn off, with the bubblers in our coolant tanks we are down to about 50 lbs of air within 2 hours of quitting time and it should have a low air alarm and shut off the machine for that reason too. It usualy has less than a half an hour to run so it should finish and shut off, the low air should make it alarm out and shut off ( the low air does alarm out unless it is in this stalled condition ). But she keeps on spinning and spitting out cooland but making no chips.
 
kustomizer:

That other machines connected in a similar fashion and operating at the same time do not exhibit the same problem implies that the problem is related to the specific machine, but the problem may be triggered from an external cause.

If you had all identical machines you might still have exactly the same result. That disconnecting the RS232 cable eliminates the problem implies that the cause or trigger to the problem event is in some fashion related to the cable.

If you could even guess at the hangup time, this time information help identify the cause.

.
 
Would It seem that the processor would be the root of the evil? I did get a new one on the UPS truck today along with a new auger motor and some paperwork indicating a warranty repair so I guess they realy are going to change them I hope it fixes it as it is a pain to keep one machine unplugged from the communication hub.
mark
 
kustomizer:

There are millions of transistors in your machine. Most of these will normally be quite stable in one of two states (on or off, 1 or 0). Sometimes one or more may be flaky and unstable caused by a defect, change in chemistry, stress (thermal or other), voltage, radiation particles, etc.

Until you substitute boards or make other measurements it can be very difficult to imagine what a problem might be.

Your machine chassis may not have a good connection going back to the main electrical service entrance causing noise into the RS232 input. But for this to have an effect on program execution would most likely mean that within the CNC "operating system" program that something was checking the UART port for data. This is not a function in any of my machines.

Yes, my guess is the processor board is a likely prospect because this contains a lot of the meat of your system. But always check that the +5, -12, +12 power supply to the processor is good.

.
 
I think the electrical is OK, as well as the machine was moved over and rewired about two years ago and the trouble moved over too. Our solar stuff makes the power much more even that it ever was before, we are so far out of town I am supprised the power or phone work at all as the lines are tangled in trees and laying on the ground in some cases( the phone wires not the power).
It was suggested to me the setting for DNC may be on and could cause this too, and if I remember right I may have used that machine in DNC mode as the memory was not quite enough for a fishing lure mold I was playing with. I will look into that a bit later when it is on again.
 
that machine is getting a signal sent to it sometime during the night, you can duplicate the problem by trying to send a program to the machine while it is running, it will show a "waiting" at the bottom of the screen, it can be cleared with "cancel", you need to find out if someone is inadvertantly sending something to the machine during the shift, before you leave.

Hope this helps!

Darrell Reid
 
reid:

What year machine and what options cause waiting to appear while you are running a program stored in HAAS memory, meaning not running in drip mode?

.
 
Nobode is here, the lights, computer, hub are all off when this occurs, it did have the DNC enabled, however there should not be any signal coming with everything. Hopefully the new processor will fix her up.
 
Check the sensors.
I had this exact problem on a Mitsui Seiki and the sensor that indicated "closed ATC hatch" didn't have a signal.
 
reply

It usually happens to me when the programmer (me) doesnt change the rotary switch and tries to send a program to a machine that is running, I have had it happen on my 2005, VF-5 and SL-30.

Strangely, it doesnt happen on my 1993 VF-0.

HMMM

Darrell
 
reid:

That is interesting and I believe that would imply a design concept defect in newer HAAS machines.

What is the newest machine that does not have this problem, possibly 2000?
What is the oldest new machine that exhibits the problem?

I believe it was some time after 2000 that HAAS introduced the capability to send data to an executing program.

In all probability HAAS uses an interrupt from the UART to the CPU, rather than poling, to read data from the UART. If the priority of this interrupt is fairly high, then a blast of data to the UART might cause the main operating system program to crash. Or a buffer overflow.

I suggest that the best way to design this type of communication would be to have a totally separate RS232 sub-system for receiving data and this would never interact with the main CNC program unless that program wanted to extract some data. As far as the running CNC program is concerned the RS232 effectively would be turned off.

If you have identified the cause of this problem, then kustomizer may have noise getting to his machine when the driver RS232 box is not on. In this case our I232 system might be of value. But it would really appear to be a HAAS problem. The other possibility is the failure does not occur at a random time but just when the RS232 box is turned off. Here our isolator might not be much of a solution, but we might be able to bias it to eliminate a change of state.

It would be very useful as background information if machine build time can narrow down which machines exhibt this problem. Then we could find out if the addition of the data receive function correlates.

.
 








 
Back
Top