[jmccracken]

I'm completely lost ...

I continue to have trouble drip feeding my Heidenhain 151B. Note that I have no problem communicating from PC to controller, and vice-versa, when the controller / TNCServer on the PC are configured as follows:

* Heidenhain 151B controller is set to EXT mode (baud rate has always been set to 2400)
* TNCServer (running on PC) is set to FE (see attached picture of TNCServer set to FE mode)
* Machine Parameter:
M71: 515
MP218: 17736
MP219: 16712
MP220: 279
MP221: 5382
MP222: 168
MP223: 1
MP224: 4

Also, I have no trouble communicating when both the Heidenhain controller and TNCServer are both in ME mode.

My symptoms are that I get a "BLOCK FORMAT INCORRECT" error (see attached picture) when drip feeding a large program, screen says TRANSFER BLOCKWISE, a large program. Specifically it gives this error on line 1221 every single time regardless of whether I am running the program during transfer or not.

Running a program of less than 1221 lines behaves fabulously.

From reading other posts, I see that the Heidenhain controller should be set to FE mode. When I do this, the TNCServer simply says "stopped" as soon as I attempt to either upload, from the PC running TNCServer, or download from the Heidenhain controller. In other words, FE mode (reached by pressing the MOD key) on the Heidenhain controller doesn't seem to work.

What in the world is it doing? What should it do? What should I do to drip feed this (and other) large programs?

Thanks again!

[awemawson]

What cable wiring are you using? Sounds like a handshaking issue. MP222 = 168 equates to software handshaking so in theory it should work with a 3 wire link. I assume it is getting to the point where in internal buffer overflows and gives you the error, however the handshaking should stop it before that. I seem to remember that the software hankshaking uses 'ack' and 'nak' however the values that represent 'ack' and 'nak' are alterable in the parameters (sillly really as they are standardised ascii characters)

[jmccracken]

Thanks for responding. My cable checks out as follows:

(25pin TNC) - <9pin PC>
(2) - <2>
(3) - <3>
(4) - <8>
(5) - <7>
(6) - <4>
(7) - <5>
(20) - <6>

I've confirmed this by ringing out my cable with a multimeter.

Also note that, within TNCServer, I have an ability to "short circuit hardware handshake" (see attached picture). I have tried both with and without this box checked with no difference.

Can you suggest the appropriate machine parameter settings?

[awemawson]

Sadly I no longer have my Interact so I can't just have a looksee and am remembering from about 4 years ago!

You need to download a copy of the TNC151 Interface Description from the Heidenhain site and look at the parameter descriptions. On page 60 it refers to all the ascii control characters being able to be changed by parameter, and I suspect that may be your issue. I do have a copy of my parameters from my TNC151B that I saved years ago if you want to bang me an email at andrew(at)mawson(dot)org(dot)uk I'll send you a copy but remember the machine side of things will differ although the comms side should be the same.

[John Bennett]

Cable should be (DB9-PC to DB25-Control) 2 - 2, 3 - 3 and 5 - 7 (only three wires are needed but you can also have a shield ground if required on one end of cable only). You can look at Cadem.com under DNC for wiring and settings, but they do not support drip feed under cademlite.

[jmccracken]

I wanted to try to give a definitive guide to setting up drip-feeding a Heidenhain 151B based on my (epic) saga of the last few days. I can't say that there aren't other ways to make this work, but the following steps work on my machine.

1. Communications pertinent machine parameter

M71: 515
MP218: 17736
MP219: 16712
MP220: 279
MP221: 5382
MP222: 168
MP223: 1
MP224: 4

2. Cable between PC (9-pin serial / DB9) and Heidenhain 151B controller (25-pin serial / DB25). This equates to a DB9-DB25 null modem cable.

(25pin TNC) - <9pin PC>
(2) - <2>
(3) - <3>
(4) - <8>
(5) - <7>
(6) - <4>
(7) - <5>
(20) - <6>

3. Install TNCServer from Heidenhain website. Setup TNCServer to 'FE' mode and 9600bps (baud). See attached screenshots:

4. Press the MOD key on the Heidenhain 151B controller. By pressing the up or down arrow, find "RS-232-C-INTERFACE" and set it to 'FE' mode. See attached screenshot:

5. While still in the MOD menu, press the up or down arrow, find "BAUD RATE". On my controller, my only choice was 2400. Leave it or set it to 2400. Note that this is the non-sequitur ... and what took me so long to figure out (specifically that the Heidenhain controller is set to 2400 baud while TNCServer is set to 9600 baud). WTF, one might ask ...

6. Turn on the PC and connect to the Heidenhain 151B controller which is in any normal operating mode (manual, automatic run, etc.). TNCServer should say "Ready" in the bottom left corner.

7. In TNCServer, set the "folder" to the correct folder having your programs. Note that ISO programs (what I'm using) have to be named the 'program number' <dot> 'I'. In this case, program number '1' must be a text file named "1.I". I'm told that Heidenhain programming language uses 'program number' <dot> 'H'.

Now, you're ready to drip-feed and run programs. Some notes:

* My controller loaded around 1700-lines of code before it stopped and waited for the controller to ask for more lines of code.
* I was getting a "BLOCK FORMAT INCORRECT" error at around line 1225. I spent a long time thinking that it was some sort of error associated with the TRANSFER BLOCKWISE (aka drip feed) function of the controller. Turns out that this error is generated when there is a bad dimension in a G02 or G03 command (in either I, J, X, or Y entries). In other words the controller is smart enough to be testing each line of code as it is entered to see if there are moves which cannot be made ... thus, by getting this error, it was saying that there was a bad line of code in the program rather than a fault in the communication between PC and controller.

I hope this helps someone fighting through this challenge ...

[John Bennett]

Good write-up!

[awemawson]

Presumably TNCServer auto senses baud rate and adjusts accordingly. My current TNC355 allows 9600 to be set on the control.

Glad you found the error.

[gus]

it does not on this level control

[xsa1]

Hi this may sound like a Silly Question but which file has the TNCServer in it cant seem to find the right one on the heidenhain website.

I to will be trying to link this into our TNC151 control.

thanks

JAck

[gus]

it is part of tncremote

[moonman5]

Hi jmccracken,

Brilliant post, helped me out a lot! I was having trouble setting up my Tnc 151for drip feeding and found your post, followed your on screen instructions and got a result! all works fine now!

So thought I would just say thanks!

Moonman5

[jmccracken]

Fantastic! I'm so glad it was helpful. It sure took a long time to figure out ...

[staffordshirech]

Thanks from me also. I have altered all my settings and parameters as above, checked the cable pin-out but no good..
Unfortunately, my machine is still not speaking to the PC.....

The TNC server shows 'ready' but the TNC just says 'in-/output not ready'
every time I try to upload or download anything.
The machine was in use up to two weeks ago and otherwise works fine.
I just feel I am missing one little setting somewhere!!

I will make enquiries with it's previous owners whom I know were feeding it remotely, to see how it was done there.

Meanwhile if anyone has any wonder cures? Please post them!

Les

[staffordshirech]

Sorted!
Following some advice on the Cambam forum, the problem was traced to a pin cross-over.
Now my machine too works fine on the above settings.

Les

[Ludde85]

Hello,

I need some help with this as well... I have a TNC155-controller, don't know if it is A/B/P/Q...

However, I'm using TNCserver, and have some problems with Drip-Feed.

My cable is connected as jmccracken wrote, and also the parameters are the same. Baudrate is set to 2400 on both the controller and in TNCserver. On my controller, I have no possibility to change the "RS232 INTERFACE" if I press the MOD button, so I have no idea what it is set to. Or where to find this setting...

I choose "FE", Baud 2400, COM1, and the correct folder in TNCserver. In the status field it says "Ready".
I choose "program run - blockwise" on the controller, press EXT and I get a message that says "Blockwise transfer" and "Program number =". I choose program number 1 (as the filename is 1.H) and press ENT. I can see the lines transfering in the TNCserver field, and so far so good.
The problem came when line 1119 was transfering (or atleast this is the last line showing in the TNCserver. Then the TNCserver status field says "Stopped" and in the TNC I get the message "Fel programdata", in english it will be something like "Faulty program data". The message you can see below in the picture is just the old "blockwise transfer" message that's still there, so this has nothing to do with it.

My first thought was that it must be something wrong in the Block (like jmccracken's problem), but I've checked the program-file, and nothing is wrong there...

I hope you can help me out with some hints or ideas, as I have no idea what to do next...

That's right, forgot a quite important thing... When the fault message "FEL PROGRAMDATA" showed up, Also the two lines below appeard;

001 C X-27,376 Y-9,568
113 DR+ R F M

Also, in my code, the block 1113 look like this;

1113 C X-27,376 Y-9,568 DR+ R

So it kind of look like the controller made two blocks (001 and 113) of my line "1113".

I guess this is the cause of the problem, but WHY did the controller do that on that block, and not on the first four-digit blocks..?

Attachment 217094
Attachment 217092

[staffordshirech]

Hi,
The numbering system of two lines is normal. Seems like it can only do 3 digits before it automatically continues on the next line.

I have had your problem several times. Each time it is because there genuinely is a fault with the code. The machine is very unhelpful really. It says 'problem' but does not or cannot tell you what!
Try sending a very short program that you know is correct.
I use CamBam for most of my code generation and it does sometimes generate a glitch on complex operations.

Download this useful emulator from TNC Emulator You don't need to buy it, the trial version works well enough for testing existing programs.

I can email you a known-to-work program if you want one.

Les

[mecanico]

Did you get it to work?
Greetings

[janisz]

I was following the advice here to get the RS232 working on my mill with TNC 155P control and ran into a problem - using oscope I probed both Tx and Rx pins on the connecting cable using signal ground, and both pins go down to around -12V level, when sending data from the PC to TNC the signal level reaches +12V as expected, and using terminal software on the PC to send random strings I can get the TNC to complain about wrong data received, but when I try to send something from the TNC to PC, the signal level voltage never reaches -3V, the definition of the signal looks good, no distortions etc, just that the voltage level of the signal doesn't reach the minimum +3V for the receiver to sense it, like an op amp at the output that should have had aprox -/+15V power only has -15V and 0V, hence not being able to pull the pin up to the required +12V

To test if my connecting cable is sound, I used the cable to make a PC-PC connection and upon scoping the Tx/Rx pins the signal level was bouncing between -12V to +12V just as expected.

So I'm guessing there is some sort of hardware issue on the TNC side, I haven't yet got around to opening it up and checking the RS232 circuitry, maybe someone has encountered this problem and could give some hints? Having hard time finding any schematics of the TNC 155 as well...

[machinehop5]

Is this the correct Manual for your control?
http://www2.i-logic.com/manuals/Heid...-interface.pdf