1997 HAAS VF-2 Mill Will Not Drip Feed Large Files

It's been a while,... There could be a number of things effecting it, larger programs with drip feed, more chance of interference.

Double check your communications Parameter (hand shake) , don't think that's the problem if your smaller programs run

memory caching may not set correctly (check that) even though drip feed, as I recall, it reads a block at a time... but that was a long time ago.

I am guessing that it has never worked? True?

I had what I believe is a similar problem with an old HP plotter with a RS232 interface.

Or, and I am guessing, but it could be that the USB to RS232 interface you have is not correctly connected and USB being a streaming communication, its "over running" the data buffer eventually.

Buffering is helping for small files, but then the buffer over runs and s**t happens!

Also, certain USB to RS232 connectors, especially cheap Chinese ones from many sources, are not hand shaking correctly, thereby causing the exact same problems....

The company that sells PICAXE chips in the UK had this problem some years ago and they now supply for a small amount of money, a USB to RS232 converter that apparently works correctly. They only guarantee this one with their chips. But it has an unconventional connector at one end though...there may be better answers around.

My fix years ago, was to throttle down the output of the PC, so that it was sending data slightly slower than the HP plotter could handle it. A cheap fix, but it worked for many years without problems!!

Its a common problem to my mind. One that is simple to fix, ONCE YOU KNOW WHAT THE PROBLEM IS!!

Try talking to these firms and reading on the websites about the problem:-

http://www.kksystems.com/usb-converters?gclid=CjwKEAiAmo_CBRC9qbGQssjqi28SJABYTgZxOLGuf3jQk-wcRwa-UMK8BIHcC_B8G3P83AQhV-2KFhoCxLjw_wcB

or

https://www.lammertbies.nl/comm/info/RS-232-usb.html

Reading these may help you in getting the RS232 end correctly connected, which it may not be:-

http://www.windmill.co.uk/rs232-communication.html

http://electronics.stackexchange.com/questions/24190/serial-to-usb-adaptors-types-and-problems

There is loads of websites for this problem, I simply searched on "usb to rs232 with correct handshaking".

Remember, it might be just one or two pins on the RS232 side that have been ignored, or incorrectly placed, it could be as simple as 5 minutes with a soldering iron and correct knowledge!!

A breakout box with LEDs, will give you a clear demonstration as to what is gong wrong I feel and are not expensive. Years ago I would add one in to a system and simply leave it connected forever!!! It showed me immediately if everything was working correctly!!

Something like this maybe:-

http://www.ebay.com/itm/Pocket-Sized-RS-232-Serial-Break-Out-Box-CablesOnline-TS-BOB-/280852483299

OK i am wild guessing. Are the communication parameters correctly set?

Some times only a miss typed number can disturb the communication 9600 would be correct but 9060 bd will not run!

Another recommendation if you run at 115200 bd it might be slightly too fast for an ageing PC so go one step down in the baud rate or try 56600 bd - in most cases fast enough and working.

Or check the quartz crystal in your machine if it says 16MHz and due to aging - possible as these are cheap crystals or even ceramic resonators it is down to 15MHz or when their frequency is off from the marked value the calculations to figure out the baud rate could leave the permissible tolerances and then you get errors in the communication.

Use a serial communication spy scope able to decode the serial communication (some cheap scopes can read the communication (e.g. Pico scope 2205A can decode the RS232 communication - some times if if you the see any special characters ); $; besides 0 to 9 and ABCDEF or sense making strings (G codes and such) then something can be wrong with your communication.

But that is just some guessing what gan go wrong.

Best regards Uli

專業的雷射切割機製造商，需購買機器可以聯繫我

So... I have a conclusion to this issue, and I thought that it is worthy to share because this can happen to anyone who has one of these older HAAS controllers.

We realized that this alarm code 248 out of number range refers to the "N sequence" lines (i.e. "N100, N102, N104...)

Once the n sequence lines got up to the N100000 (6 numerical spots) the controller would quit reading. Apparently, the controller was not designed to handle more than 100,000 lines of code.

We were able to successfully update and modify our post process through or CAM program. We removed the "N-lines" completely and now we are running large drip feed programs with no problems!!

Strange isn't it? Reminds me of the 2K bug where all the computers weren't designed to go past the year 1999.

So if anyone else runs into problems running large drip feed programs on an old 90s vintage HAAS controller, you may fix this by simply removing N-sequence lines from your G-code.

it did sound like a memory issue at first in this case..., but now it sounded like a syntax issue for the block ID,...ie N??????

Thanks for sharing