I have a delta 20 on a journeyman and have been able to load and run many programs succesfully, but I now need to iron out my "handshake" issue. I have ordered the rs232 cable I was previously advised in earlier post but no mattter what I cannot get the flow to pause when the storage is full.I can slow the baud rate down to 600 and wont let it overflow but then I aill get into the complex moves and it is loading too slow. I have tried many different settings and think that thee machine is not sending the signal to laptop. I tryed to force the handshake with using the graphics and dialing down the feed but it continues to flow info. I cannot get a program to " record" at all. I use a cimco program to feed and have played with all the settings. One thing I was explaning to Dynapath that they thought was odd is how I cannot start the control box if it is atttched to the laptop. The screen will stay blank, also when I set the flow conrol to " software" it will immidiately shut down my laptop and I have to reboot the laptop. I have a good grounding rod, it works flawlessly with every other aspect, I am just sick of the "buffer imput error" when I do comlex programs causing me to do way too much babysitting. IN the manual it talks about crossing the RTS and CTS lines but im am not sure about this. I set the cimco flow control to hardware and when I start the mill just sits an waits for a signal that it never begains.Then in the hardware flow type settings I choose "RTS/CTS+DTR/DSR" and it runs but it still overloads the buffer. Any solutions for this will be greatly appreciated, I wish there is a simple affordible to increace the size of memory, I find it frusterating that it will only hold about 2000 lines of code! I know this seems garbled, but I have had to peice all the know how to run this machine from this site as it not exactly up to date cutting edge tech, but it cuts like a brandnew machine, after reading other posts I enjoy this problem, It could be much worse! Thanks
1.) If you are using software handshaking (XON/XOFF as the DynaPath calls it), the data flow to the control is stopped (in Buffered Input only) by sending an XOFF character from the control to the laptop. If "Record" does not work in sending a part program from the control to the laptop, then it's a sure bet that the software control (sending the XOFF character) won't work either.
2.) I would recommend using hardware handshaking, as it seems to work better with the newer operating systems (Windows 7, in particular).
3.) Always start the receiving device first. If going from the laptop to the control in Buffered Input (DNC), the control should be in cycle before starting to send from the laptop.
4.) Does your laptop have a serial port, or are you converting from an USB to serial? What operating system are you using on your laptop?
5.) There is more information on the DynaPath website under 'Control Support.'
6.) Consider the possibility that the transmit side of the UART chip (the actual serial port) is blown on the Delta 20 Aux Control board. DynaPath can fix it or check it out, but I know it will cost a couple hundred and the card will have to be sent in. You'll lose the machine for a couple weeks.
I am using windows 7 sometimes 32bit ,and sometimes 64 bit laptop. I have a 32bit hooked up now but If it works better I can create a "virtual computer" with win 7 and run it as a xp. I have tried 3 different cables and the one that I got that was recently recommended to me from this site works without the null modem. That may be odd in itself because when I attached the null it lost activity and then when I removed it it worked just without handshaking ability.The machine was used as a school model and I was the first to cut metal on it even though it is old. It cut wax in our class. I know it was taken care of well ut electronics do go bad I suppose. I have a rs232 jumper with a removable top that allows the ability to rewire the connection, my fear is to short something out and cause damage to circuitry. Is there a way to see that the signal is sent or received? is there code that can be written to test that or does it run behind the scenes un-seen? I will go back to the dyna site you recommended and thank you for the advice.
Actually, it's not weird at all that you remove the null modem and it works. Some models of Tree were indeed parallel, not serial. However, with the straight cable, you are missing the jumpers that enable Xon, Xoff. Easiest thing to do is to remove the DB25 connector on the machine, swap 2 and 3 wires on the plug, install your null modem and go. Also sounds like your trying to use record mode, you should be in drip mode.
From the program directory, type in rs232c, that puts the control in buffer mode.
Because Windows 7 usually only runs on newer equipment, I assume that you are using an USB to serial converter. Based on user feedback, I have to say that some work and some don't. It could explain some of the issues you are having.
Also, whether you simulate running Windows XP or not, there is no way to reliably determine how the data buffering is take place on a laptop running Windows 7. From some experimentation, it appears to me that the software buffers do not respond correctly or do not react fast enough to perform software control. This is why I recommend hardware control, since it takes the operating system out of the picture (kind of). An USB to serial converter still clouds the issue because many, if not all converters need software drivers to function.
That being said, if you were to disconnect the laptop and connect a "breakout" box with LEDs to the serial port of the DynaPath control, you could "Record" a part program and verify that the control is sending data out of the serial port.
I followed the advice and tried to send a program before pushing the start button and It will not load the program when the flow is set to Hardware or software. Only when I put the setting to "None" does it send successfully. It also will not wait and it loads the program regardless of the start button is pushed or not. I Tried every way except now I will follow the tip and convert my computer to a xp . I will also attempt the "breakout box" I assume it is the same as testing the pin coming out with a meter to see if it "tickles" as I load the program. Am I doing that correctly or is there a different "breakout box"? Thanks again for the tips it helps to narrow down the problems and saves a tremendous amount of time troubleshooting. In looking through the dnc program I see there is a million different settings for drip feeding and I just use the default settings, but I wonder if there is a special character It should be set to or if again it is all automatic regardless of the settings on the software side. I am leaning to the dynapath not sending a signal, because of how the software sits and waits without loading when I have the flow control set to that.
I am only trying the record mode as a attempt to follow the feedback from dynapath to "force the handshake" I am running the rs232 well and it works fine except with large programs that overload the buffer. I need the connection to pause the flow to avoid the error. I think that I am heading in the right direction with the wires crossed. can you please expand a little on that so I can make sure I do it right? Thanks for the help. I sure do appreciate all I get.
OK couple things:
Are you using a USB to RS232 adapter? If so there are only a few that will work.
I recommend (as I did) getting a "port replicator" or "docking station" for your laptop so you have a real serial port. I tried a converter (it didn't work), sent it back, & bought a new IBM docking station off (auction site) for $15. Works great.
The next issue is, what are the connectors, DB9 or DB25 or some combination?
If both have DB25 the pinouts are pretty simple. DB 25 to DB9 a bit more complicated, but try this article: Null modem - Wikipedia, the free encyclopedia
Third, stick with hardware handshaking, the software (XON/XOFF) creates problems because it sends a special character on the data line that must be interpreted and other controls need to be in place.
You've got the data transfer going and that's half the battle. Just need to get the handshaking working.
The "RS232 Breakout Box" (do a search) can be had for around $20. It hooks up in series with your cable (between the host computer and machine control) and allows you to try different pinouts. One I have has LED's so you can see the signal (green +, red -). Don't worry about getting it wrong, it's a ~2V signal and can't do any damage. Most common configuration is: 1 straight thru, 2&3 crossed, 4&5 crossed, & 7 straight thru (db25 TO DB25). On the breakout box you should be able to turn off the other signals. There are dip switches and if you use a jumper make sure those lines are "off". I've been using a straight thru cable & breakout box until I make the time to solder up a cable with parts I bought at radio shack.
Software settings are another subject. I'd try 1200 baud on both ends until you get it working, try a 5ms End Of Block Delay (EOB) if it's available (gives the machine some time to respond to the RTS/CTS signal). You must have parity & stop bit set correctly or you wouldn't have made it this far.
Last edited by mfgbydesign; 01-10-2012 at 09:48 PM. Reason: added detail
I see much of what you have here is what I have been looking for. I guess the good part of all this is the great experience I will have with these connections by the time I am done with this. I will post my progress as I may run into more roadblocks. Thanks for the great advice . Hopefully I will be able to pick one up today and test this out tonight. I have a "old time" electrical supplier in town he seems to have anything I can think of at reasonable pricing. I have tried several different cables and USB converters and see that they don't all work as you said. The combination I have now works well, all but the cts/rts is my assumption. Thanks again for the hand in this, I think I am close!
As I watched the LED light up when I connected to the port, I noticed the (two rows of lights, one green, one red) red #8 showed on when the control was switched on, the red 4 and 20 along with the opposing side green(RTS and DTR) woulds blink until I activated the port. Then all 4 (R4,R8,G RTS, G DTR)stayed light up. I set flow control to "None" sent prog. and red# 2 light up as it transferred until complete. I set control to "Hardware,flow cont. to RTS,CTS" and sent a prog, the computer prepared to send but held waiting for a response to send and would not transfer. I switched to " Hardware, DTR,DSR for flow" and it sent same as usual where it loaded completely and ran. I could not get it to pause with the feed rate it was all or nothing,and only loaded with the control start buttton pressed first. So If I understand correctly, I should swich the #2 and #3 wire and then add the "null adapter" and it should have the connection correct to facilitate the handshake. I will now go through all the last set of instructions from previous posts and try each step watching the led feedback and note the changes. Here is a picture of the box I used, thanks for the pointers,I hope to have continued success. This gives me a good idea whats going on in my transmitting.
You are using a tester, which does not give you the ability to try different "pinouts." If you look at this "breakout box," you have the LEDs, plus dip switches for each line, and jumper wires you can use to, for example, cross 2&3. Keep in mind if you use a jumper you need to open the dip switch for each pin (in most cases).
(I am not endorsing this specific product, just looks like the one I use)