Re: [CAD_CAM_EDM_DRO] Troubleshooting a BTR

Jon Elson wrote:

>vinnygenovese wrote:
>
>
>>I am trying to bring an IKEGIA FX25-N / Fanuc 5T back to life. There is a BTR
>>that I am trying to bring online. Problem is I don't have any instructions or dip
>>switch settings. I uploaded a photo to http://radiantheat.com/UCM-BTR.jpg.
>>The board has a single LED , which seems to 'blink' when i send data, but it
>>seems no signal gets through to the Fanuc 5T. (or I am sending the wrong
>>commands). I googled and googled. It seems they are out of business. Any
>>hints on the dip switch's or other hints would be very welcome.
>>
>>
>Well, I think it can be deciphered. It looks like there is no
>microprocessor there,
>I think the big chip is a 40-pin UART. It also looks like the markings
>have been
>sanded off that chip, but by tracing the power pins, you may be able to
>figure
>out which if the several 40 pin uarts it is. It is socketed, so pulling
>it and
>looking on the back may have some markings that at least identify the
>manufacturer,
>which may be all you need to narrow down the type of UART. Once you
>know this,
>you can see which dip switch is connected to which pin, and with the
>crystal frequency,
>you can figure out how to set the baud rate and transmission characteristics
>(# bits, even/odd/none parity, etc.)
>
>I could be off on this, but there is clearly no EPROM chip, and
>something about
>it just makes me think it predates one-chip micros. I note 1488 and
>1489 chips,
>and wonder if it is getting the right power to those chips, as they need
>both
>positive and negative voltages.
>
>I don't know Fanuc machines, but most systems need some switch settings
>and/or
>commands or buttons to put them in the right mode to read a program from
>tape
>or to go into DNC mode.
>
>In DNC mode, the computer becomes a file server for the CNC control, and
>doesn't send anything until the CNC control asks for it by file name.
>
>Jon
>

Hmmm.

I think the large chip actually is a microcontroller. The 8051 has been
around forever, as have the precursors to PIC line. The tipoff is that
the crystal isn't an oscillator, but a resonator. (also, there's no
reason to sand off the markings on a UART) According to this pinout (in
ascii art :) http://www.xs4all.nl/~ganswijk/chipdir/giicm/8250.txt , the
8250 (which it almost certainly would have been) takes a single clock
input - like that from a clock oscillator. Also, the clock input is on
pin 9, not pins 38+39 as this one.

Of course, it's not an 8051 either, since the crystal connects to pins
18+19 (unless the chip is rotated). I checked the Z80, and that's not
it either.

I don't remember any UARTs that weren't compatible with the 8250, though
they could certainly have existed anyway.

If you could post a higher resolution photo, as well as any information
on the underside of the large chip, that may help. Also, what
diagnostic tools do you have available? (do you have a scope?)

- Steve