>From a friend.
Anyone able to help


         Russell

FC-9801V or it's desktop cousin.

    Beep codes
    Circuits
    Unusual-fault comment.
    Container-loads of new machines in original packing.
    OR one working old one for sale?



______________


Russell,

Chris has a Japax wire-cut (EDM) machine that he acquired a year or so back
which he is now trying to get going =A0- in part so I can cut a replacement=
 for a
damaged spur gear in my Elliot shaper machine.

The CNC controls for the Japax are based on an NEC FC-9801V "factory comput=
er"
- which is a derivative of the once hugely popular (in Japan) NEC PC-9801
desktop computer. =A0NEC FC- series machines were apparently extensively us=
ed in
Japanese CNC machinery (welding and metalworking, embroidery, knitting, etc=
..)
through the 80's and 90's.

The FC-9801V in Chris's machine has developed a bizarre fault. =A0If left
unpowered for >28 hours it will boot (apparently) normally, but it will not
re-boot if switched off and powered on again within around 28 hours of swit=
ch
off, and it will not reboot if the reset button is used while powered on. =
=A0When
it fails to boot it generates a sequence of three "beeps" =A0- presumably a=
n error
indication =A0- and the processor READY signal is then found to be held low=
..

We have carefully gone over all boards and replaced any capacitors that loo=
k
suspect, but to no avail. =A0The phenomena is unaffected by temperature, wa=
s
unaffected by baking the board-set at 40C for 24 hours (so unlikely to be a
board leakage problem), and is unaffected by the presence/absence of the RA=
M/RTC
holdup batteries. =A0The time constant also seems unaffected by attempts to
provide relatively low-resistance discharge paths by holding the solder sid=
e of
boards in contact with a sheet of black conductive (anti ESD) foam with the
power off.

My guess is that the problem could be of an electro-chemical nature (possib=
ly
due to ingress of contaminants into an IC) with perhaps electrical bias dri=
ving
a chemical process in one direction which then reverses slowly over time wh=
en
bias is not applied. =A0It's a pretty "left-field" theory but I can't think=
 of
anything else that would explain such a long time constant independent of t=
he
things we have tried to influence it.

Chris received zero documentation on the machine, and despite many hours tr=
ying
Chris and I have both failed to find any relevant technical information on =
the
web (other than an unhelpful list of DIP switch functions =A0- which Google=
 does a
relatively poor job of translating from the Japanese).

IC datecodes indicate that the FC-9801V in Chris's machine was manufactured
around 1984. =A0It is based on NEC's uPD70116 (V30) processor and I believe=
 it is
very similar architecturally to the PC-9801UV series of desktop machines of=
 the
same era.

As a minimum it would be helpful to at least know what the "beep codes" for=
 this
machine are. =A0That might at least tell us where we might start looking. =
=A0It is
likely that they are the same as for it's PC-9801 desktop cousins of the sa=
me
general era. =A0Beyond that a technical manual for the PC-9801UV (we have
determined that an english-language translation was done) and/or schematics=
 of
the FC-9801V or it's desktop cousin.

Note that this is not a commercial venture =A0- so the purchase of very exp=
ensive
spares (which may be available in Japan) is not a viable option. =A0Japax
themselves went out of business quite some time ago and the machine hasn't =
been
supported for many years. =A0NEC no longer provide support for older FC/PC-=
9801
machines and (say) they can't provide any information other than the scanne=
d
(Japanese language) sales brochures posted on their legacy archive page:

http://www.nec.co.jp/fc/pdf/catalog/end/fc9801v.pdf

Any ideas ?

Regards,

Ken

--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.