Siemens 810M Communication. help Needed

I'm New here. Lots of good info here. Look forward to learning. Did some searching for help with a problem I have but did not find anything yet.

Using a Siemens 810M under the General Numeric Brand Name. About 17 years old hooked to a Motion Masters CNC Router. Using BeachCam to upload/download data. (Old DOS based but it works)

About 16 months ago we lost the ability to download from the controller to the computer. Everything else worked fine. Now the Upload does not want to work. (Everything else appears to still work just fine.) Literally I left one evening with it working and the next evening it did not. Machine was shut off the time in-between.

Any tips on troubleshooting this? We have used the same cable, same software and same communication specs all along. I have already rung out the cable along with its jumpers. I am pretty sure the computer's I/O port is working. My only positive clue that the controller may be talking to the computer is at one point i had the controller in "data in" mode (DIO) and shut the computer off. I immediately received an alarm at the controller.

I do have a call into Siemens but so far nothing from them.

Any ideas? I plan to get a Serial port checker to be 100% certain on the computer end. I have no problem getting into the computer to ring things out. Will ring out the 25 Pin D-Sub next. Have access to electrical test equipment if there are any ideas out there. If need be I can post the communications specs on the PC end as well as the controller end.

Thank you for the help.

Quote:

---

Originally Posted by mjhjenks

I immediately received an alarm at the controller.

Any ideas? I plan to get a Serial port checker
Thank you for the help.

---

The alarm was probabally because the hardware handshake was lost.
If you get a checker, you can do a loop-back test to check the PC in and out.
The majority of programs use XON/XOFF software handshaking, so you can usually use the three wire cable with the HW handshake jumpered out at both ends.
With a checker, you can also confirm the data is comming out of the controller.
Al.

Some more info. Pulled the controller apart last night. Verified that i have continuity in the RS-232 line from the PC into the controller into the RS-232 port on the CPU card. (6FX1125-8AB01)

Spoke with Siemens. They say it is common for this card to loose the port. Even like mine did having output go down and input work just fine for a while. (There is one chip for input and one for output.) They suggested that i switch to the second RS-232 port on the CPU card which has its own set of chips. (Also need to changing the internal settings. SD50xx bits or something like that) That is the next test. If that does not work then the damage is below where these two ports are linked on the card. (Not just a chip) Time for a new card or have it repaired.

Question. Anyone have experience with this CPU? 6FX1125-8AB01. Again it is a 810M controller. Sounds like a chip went down. I know i can pay to get it repaired but i can it be diagnosed and repaired at home? Had to ask.

Matthew

080821-0726 EST USA

mjhjenks:

If standard 1488 and 1489 interface chips are used, then these are extremely reliable if you do not exceed their ratings. However, lots of these fail because direct connection between one system and another may allow for excessive voltage to be applied between the two communication ends. Lightning, electrical shorts to ground, and welding are some reasons for the high voltage difference. 25 to 30 V will likely cause problems.

That is why we make and sell an isolation system, I232, to minimized the possibility of excessive voltage being applied to either an input or output.

Now to troubleshoot your problem.

If I understand your current problem it is as follows:
1. You send a program from the computer to the CNC and the CNC does not receive this.
2.You have checked your wiring and think it is OK.
3. I do not know if your old problem of sending from the CNC to the computer
was corrected or not, and if corrected how was the problem solved?

It is quite likely that at the CNC you have a 25 pin D connector, and that pin 2 is RxD and this comes from TxD at the computer. If this assumption is correct, then pin 3 at the CNC would be its TxD, and pin 7 signal common. Pin 1 is probably tied directly to the CNC machine chassis and used for connection of a cable shield. Many CNC machines and almost all computers have pin 7 also connected directly to machine chassis. HAAS goes thru a 100 ohm resistor.

Use the the machine chassis, pin 1, or pin 7 as your voltage reference point. With your RS232 cable disconnected from the CNC measure on the CNC connector the voltage from pin 2 to 7 and 3 to 7. If pin 2 is RxD I expect it to read about 0, and pin 3 should be in the range of -5 to -15 V.

If this is the case, then pin 2 to 7 at the end of your cable from the computer should be in the range of -5 to -15 V.

Make these measurements and report back.

For information on our I232 and background information visit
www.beta-a2.com
First read the web page on NOISE and GROUNDING and then I232.

.

Did not get much done last night but some more info.

Gar.

I looked the CPU card over but did not see any chips with 1488 or 1489 on them. Put the controller back together but did not get much farther then that.

To answer your questions.
1. Programs sent from my PC to the controller does not get there. Usually when i hit the "DIO" key i will see the Kb indicator scroll down as the program uploads.
2. I am very certain that the cable from the back of the PC to where it enters the RS232 port on the CPU is good. (I rung it out from end to end)
3. We never fixed the ability to download from the controller to the PC. I rarely did that so it was not fixed.

I am not sure what RxD & TxD means. Three wires run from the PC to the controller.
Pin 2 @ PC goes to pin 3 @ controller (Red)
Pin 3 @ PC goes to pin 2 @ controller (White)
Pin 7 @ PC goes to pin & @ controller (Black
At the PC 25 pin D-sub, Pin 4&5 are jumpered as well as pin 5,6 & 21.
At the controller 25 pin D-sub, pin 5,6 & 21 are jumpered.
This is the same cable we have used for 10+ years and it rings out OK.

I did not get to meter the cable when a program is being sent to check for voltage. Will do that soon and report back.

080822-2115 EST USA

mjhjenks:

RxD and TxD are data signal names. TxD is the transmit signal and connects to an RxD which is the receiver. TxD supplies the voltage and current to the cable that goes to the destination where there is a receiver. You never connect one TxD to another TxD. It is always TxD to RxD.

Binary data is transferred by using two signal levels relative to a common. A minus voltage represents a binary 1 and a positive voltage a binary 0. Also there is a defined dead band about 0 V of approximately +/- 3 V. A typical signal level is +/-10 V generated from a typical +/-12 V supply. Some newer levels are around +/- 5 V.

The normal rest state when nothing is being sent is a logic 1 and is a negative voltage. Stop bits are a logic 1. A transition from 1 to 0 forms the start bit. The start bit is one clock period of the 0 state. If there has been no transmission for a while, then it is quite obvious that a change from 1 to 0 is a start bit.

The receiving UART has a priori information on the timing and structure of a data word. X baud rate is the timing information. There is always one start bit, next N data bits, a parity bit if parity is used, and K stop bits. Suppose only 1 stop bit is being used. Stop bits are a logic 1 which is the opposite of a start bit. If everything stays in synchronization, then after the stop bit has been sent and received the receiver is ready for a new start bit and this can occur any time after 1 clock period after the beginning of the first stop bit.

If both connectors are 25 pin then 2 to 3, 3 to 2, and 7 to 7 may be correct, and since it previously worked it is correct. The other pins connected together relate to hardware handshaking and are probably OK.

If the CNC connector is wired like HAAS and Fanuc, then pin 3 is RxD. Therefore this pin on the CNC with nothing else connected should be near 0 V relative to pin 7. Pin 3 at the CNC cable end should be a negative voltage of -5 to -15 V relative to pin 7 of that cable when no data is being sent from the computer.

.

Quote:

---

Originally Posted by mjhjenks

3. We never fixed the ability to download from the controller to the PC. I rarely did that so it was not fixed.

---

might possibly be because pins 4 & 5 at the CNC are not connected

Some more work last night. To compound my RS232 port issues somehow my controllers memory got scrambled. (Turn on the controller and no screen comes up although the red LED on the CPU is on and the fans are working) In talking with Siemens, they recommended that I clear the memory on the Controller and start over. Did that last night. Screen is back up and I worked thru setting data up to loading my TEA1 data. Had to stop there because I still can not upload to the controller from the #1 RS232 port. I switched to the #2 port but still got nowhere. Not 100% sure I configured that port right though. Will talk to Siemens again today for that.

I did take some voltage readings on the RS232 line as suggested by gar. Here are the results.

From the RS232 line at the controller end. Cable connected to the PC but disconnected from the controller. Pin 2 to 7 has +10.50v. Pin 3 to 7 has -.51v

From the RS232 line at the controller end. No cable, just checking the 25 pin D-sub on the controller. Pin 2 to 7 has -.17v. Pin 3 to 7 has +10.44v.

Sorry gar, still not 100% clear on what this means. Seams like it means that both ports are working and my problem may be deeper into the controller. One question though is does this mean my PC's port is working?

Thank you.

Some more info. Looks like I did not configure Port #2 correct. Will change that and re-try the data transfer. Based on my results so far Siemens suspects that the CPU is dead as far as I/O is concerned.

So... Repair/replacement may be in my future. Still interested in any info or places who work on these.

Matthew

080827-2129 EST USA

mjhjenks:

At rest, no data being sent, pin 2 should be about -10 V relative to pin 7. Did you have the meter connected backwards?

Assuming that you had -10 V at rest, then while sending data from the computer you should see the voltage change to something in the range of +/-1 V, and it will jump around.

Same is true of pin 3 to 7 on the CNC connector when sending data from the CNC.

Suppose your rest voltage from TxD is about -10 V, and while sending data the voltage fluctuated around -5 V, then you have a bad driver or a bad +12 V power supply at the source that is sending.

Look at my web site for some troubleshooting information. The schematic for a cable does not show the wiring for an IBM PC 9 pin connector. However, I describe the PC connections.
http://beta-a2.com/e232_photo.html

To avoid future communication problems you should consider our I232 Isolator system.

.

OK.

CPU is dead. That is the communication part of it. Otherwise it is working just fine as verified by Siemens Technical support. We were able to successfully re-set it but it does not want to send or receive data.

Anyone have any other options for repairs aside from Siemens? They can fix/swap itout for a pretty penny. Any experience with buying used units. Seams like all the independant repair places in am finding are in the UK.

Any advice.

Thanks

dear all
i am trying to use free dnc software FROM CDEM DNCNETLIT4.0 for 0M 0T FANUCCONTROLS .
PLEASE HELP ME THE CONNECTION DETAILSFROM PC TO CNC
AND PARAMETERS TO BE SET ON

VENKATESH REDDY

A quick update. Still working thru this. Was quoted between $600 and $1800 to have the CPU fixed so i am motivated to try and fix it myself if i can.

Right now i am certain that the PC is sending data and that it is getting thru to the RS232 ports on the controller. From there it goes nowhere. We have verified this by sending data from the controller to the PC. The controller believes it sent the data but it does not get into the RS232 data cable.

I did get a hold of the CPU RS232 schematic. We did locate both 1488 & 1489 chips and have replaced them. (4 chips total and we added sockets at the same time) Things still are not working. I am also looking at the D8251AFC chips too as the 1488 & 1489 feed into them. Any comments on that? All i can say right now is they are not cheap chips.

Beyond that there is also some uncertainty on what to set my 5000 series communication bits to. This is what we have so far.

5010 & 5018 to 00000001 for Xon-Xoff
5011 & 5019 to 11000100 for 1200 Baud
5012 & 5020 to 00000001 for Xon-Xoff
5013 & 5021 to 11000100 for 1200 Baud
5016 & 5024 to 00101000

Siemens is recommending i try 10101000 in 5016 & 5024 to use Xon-Xoff. We are also operating under the assumption that @ 1200 Baud my controller can clear the buffer w/o stopping the data flow and therefore have kept 5014, 5015, 5022 & 5023 at 00000000. (I am using a program called Beach Cam for uploading to the controller. If using even parady, 7 data bits & 2 stop bits. 1200 Baud using Xon-Xoff.

Thanks again for any advice.

Some progress.

I re-verified my PC is working by connecting it to another PC with the same software and sent a file from one to the other.

Next i replaced both 8251 chips. Found them new on Ebay. I believe that was the trick. After that and again re-setting all the data i was able to upload my TEA1 & TEA2 data as well as do a test download of the PLC program.

Next problem is my PCP program would not load. I have 2 copies and neither worked. The DIO instantly shut down once i started to transmit. Siemens says that is usually from a corrupt PCP file.

Can anyone tell me what the PCP file does? It looks to be a long string of alpha-numeric characters. How are they developed and are they standard or ???

Thanks

Last night i figured out the PCP problem. When you re-boot after loading the TEA1 & TEA2 files it tells you to re-load your setting data info. I glossed over that and it looked to be right but it was not. Once these were set the PCP went in just fine.

The machine is now back up and all i have to overcome are my table limits. They need to be re-set which of course like everything else is going to be hard as i do not know where that is located either. We are looking....

hola soy nuevo aca tengo una punzonadora CAPRI con sinumerik 810M el monitor no muestra nada, aparentemente por mis conocimientos el monitor esta encendido pero no tiene ninguna señal entrante, la placa de donde sale el flex del monitor sera posible conseguirla o hay otra forma de verificar su funcionamiento? si me pueden ayudar se lo agradezco muchas gracias

Mi Espaniol is muy mal. No compredo todos. Hablas english? Sorry.