From: "FyberOptic" <fyberoptic@...>
Jan 23, 2010
Go to (URL)
Put "pqu" into the Grantee Code fie=
ld.
Search, hit Details on each result, and prepare to possibly shit pants=
.
For those who are impatient and want to know what I found, how does "s=
chematic diagram" sound?
It's detailed documents of two versions of the Ma=
ilstation 250 and one of the Mailstation 350. Not only are there photos of=
the hardware, but also photos of the circuit boards, and then the best par=
t, full schematics. And oh yes they are full. For example, I now know tha=
t the CPU crystal is 12.288mhz. And, even better, I know the labels of eve=
ry CPU pin, including the unused ones.
From there, I can see that it can a=
ddress up to 8MB of memory. I can see that it has TX/RX serial pins. I ca=
n see that it does indeed have an additional unused device selector. I can=
see what is very likely an oscillator out pin, which could possibly be use=
d to determine the frequency currently used by the CPU after going through =
the divider. I can see that it has a whole 'nother unused 8-bit port. Man=
y of the pin labels I simply don't know the purpose of at the moment, so wh=
o knows what else is in there.
And for the record, the 350 runs at 34.4mhz=
.
So, why only the 250 and 350? Good question. I'll start at the beginni=
ng of my search and go from there in explaining what I learned.
First, I w=
ent in search of the FCC number on the back of my Mailstation: 1EBCHN-27372=
-OT-E. Turns out, this wasn't a normal FCC ID. I didn't understand why it=
was different from what they wanted. But regardless, I got nothing, so th=
at was a dead-end. So, by happenstance I dropped it into Google before giv=
ing up, and found some results. Nothing big, it looked like just some FCC =
update notices of some kind. Stuff that looks like this:
648-CX-2001 =
8145386-5
NOTICE OF CHANGE NOT AFFECTING COMPLIANCE
HAS BEEN FILE=
D FOR THE FOLLOWING
REG. NO.: 1EBCHN-27372-OT-E
GRANTEE: =
INVENTEC ELECTRONICS (NANJING) CO.,LTD.
NATURE OF APPLICATION:
EQUIPMENT=
DESCRIPTION: E-MAIL TERMINAL MODEM DEVICE W/FAX WITH CALLER ID
TRADE NAME=
: CIDCO
MODEL: FB DET1, DET1A, DET1B, DET2
I noticed the grantee name,=
"Inventec Electronics." So, back to the FCC database with that. Many res=
ults! But I started going through'em anyway, clicking Details, then one of=
their external photos, just to see what it was. Didn't take long, but the=
next thing I knew I was staring at the picture of a Mailstation, and I cou=
ldn't believe the other files I found in there with it. Short of a datashe=
et for the CPU, a schematic of the whole thing is the next best thing you c=
ould have.
So, curious of what else Mailstation-related I might find burie=
d away in there (like perhaps the older models), I searched just for grante=
es with that name, which you can do there on the left if you do an advanced=
search. I got multiple results for Inventec Electronics. One of them was=
listed as "Nanjing" though, which is in that text I pasted above that I fo=
und on Google. It has the grantee code "PQU". And sure enough, putting ju=
st that one into the search gives you only the Mailstation results. Just t=
he three, though, unfortuantely.
Eventually I found that the number on the=
back of my Mailstation isn't an FCC ID. It's an FCC Registration Number, =
which is different. The older Mailstation is classified under "Part 68" eq=
uipment, which apparently means it doesn't need extensive testing, since it=
's just a telephone network device. However..! The Mailstation 250 and up=
use wireless base stations to connect to their phone line. This means the=
y're using frequency spectrum, which means they require a full-on FCC ID. =
I probably could have saved myself a few steps if I owned a 250 or 350 and =
had their FCC ID to begin with. But hey, whatever works.
In any case, t=
here is one interesting thing of note about the FCC registration number. T=
he first three characters, 1EB, are this in their database:
1EB|INVENTEC E=
LECTRONICS (MALYASIA)|37 ANSON RD|10400 PENANG||MALYASIA|||08/29/1989|D REG=
INALD TIBBETTS|2151 CAMINO PABLO|MORAGA|CA|94556|
So, as it turns out, Inv=
entec is definitely who developed the Mailstation. Their name is on all of=
the schematics as well. And a Google search shows their name in Cidco pre=
ss releases/financial reports/whatever.
But here's the really interesting =
part: Inventec made a CPU for one of the TI calculators too, apparently. A=
nyone who knows anything about those graphing calculators know that they us=
ed Z80 CPUs for years. I found this in a Google search, of a guy asking fo=
r information on a chip:
I'm in search of data for an Inventec =
processor. The chip came from a TI-83+ graphing calculator as bares the tex=
t:
Inventec
6SI837 0A
Japan 0216EAI
PDARD
F0254ZAA
...if it helps, this i=
s a Z80-enabled CISC processor. I've tried googling and came up empty. Any =
data on pinouts, or clocking data would be appreciated.
Notice =
the similarity of the part number! The Mailstation's is 3SI176 0A. A Goog=
le search on the 6SI one gives some results, but I found that I get other r=
esults when I put in "6S1837". Somebody seems to have mistaken an I for a =
1.
So, I emailed Inventec, and I guess we'll see if my email ever gets bey=
ond the webmaster to anywhere that might possibly have information on the M=
ailstation CPU.
p.s. For anyone interested, here's the somewhat vague =
at times patent application for what I'm almost certain is the Mailstation.=
(URL)
=3DPALL&p=3D1&u=3D%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=3D1&f=3DG&l=3D50&s1=3D6=
,510,455.PN.&OS=3DPN/6,510,455&RS=3DPN/6,510,455
In case that link doesn't=
work right, the patent # is 6,510,455.
I almost debated posting anything =
here, but I'm fairly sure none of it is illegal to do so, because it's publ=
icly-accessible information.
From: Jon Kirwan <jonk@...>
Jan 23, 2010
On Sat, 23 Jan 2010 08:39:19 -0000, you wrote:
et/ea/fccid/
ls on each result, and prepare to possibly shit pants.
Very nice a=
nd thanks! I have a few 120's, 150's, and 250's
laying about here. The 25=
0 schematics are going to be a help
for those. I would certainly like to h=
ave such info on the
120 and 150.
Thanks again. I'm googling around, too.=
But I suspect that
I am finding nothing more than you've already uncovere=
d
already.
Jon
From: Brad Fayette <bfayette@...>
Jan 23, 2010
Content-Transfer-Encoding: 7bit
These devices are in the FCC database because they have radio transceivers.
On 01/23/2010 03:39 AM, FyberOptic wrote:
Content-Transfer-Encoding: 7bit
These devices are in the FCC database because they have radio
transceivers.<br>
On 01/23/2010 03:39 AM, FyberOptic wrote:
style="display: none;"> </span>
href="http://www.fcc.
">(URL)
Put "pqu" into the Grantee Code field.<br>
Search, hit Details on each result, and prepare to possibly shit pants.<br>
For those who are impatient and want to know what I found, how does
"schematic diagram" sound?<br>
It's detailed documents of two versions of the Mailstation 250 and one
of the Mailstation 350. Not only are there photos of the hardware, but
also photos of the circuit boards, and then the best part, full
schematics. And oh yes they are full. For example, I now know that the
CPU crystal is 12.288mhz. And, even better, I know the labels of every
CPU pin, including the unused ones.<br>
can see that it has TX/RX serial pins. I can see that it does indeed
have an additional unused device selector. I can see what is very
likely an oscillator out pin, which could possibly be used to determine
the frequency currently used by the CPU after going through the
divider. I can see that it has a whole 'nother unused 8-bit port. Many
of the pin labels I simply don't know the purpose of at the moment, so
who knows what else is in there.<br>
And for the record, the 350 runs at 34.4mhz.<br>
So, why only the 250 and 350? Good question. I'll start at the
beginning of my search and go from there in explaining what I learned.<br>
First, I went in search of the FCC number on the back of my
Mailstation: 1EBCHN-27372-<wbr>OT-E. Turns out, this wasn't a normal
FCC ID. I didn't understand why it was different from what they wanted.
But regardless, I got nothing, so that was a dead-end. So, by
happenstance I dropped it into Google before giving up, and found some
results. Nothing big, it looked like just some FCC update notices of
some kind. Stuff that looks like this:<br>
648-CX-2001 8145386-5<br>
NOTICE OF CHANGE NOT AFFECTING COMPLIANCE<br>
HAS BEEN FILED FOR THE FOLLOWING<br>
REG. NO.: 1EBCHN-27372-<wbr>OT-E <br>
GRANTEE: INVENTEC ELECTRONICS (NANJING) CO.,LTD.<br>
NATURE OF APPLICATION: <br>
EQUIPMENT DESCRIPTION: E-MAIL TERMINAL MODEM DEVICE W/FAX WITH CALLER ID<br>
TRADE NAME: CIDCO<br>
MODEL: FB DET1, DET1A, DET1B, DET2 <br>
I noticed the grantee name, "Inventec Electronics.<wbr>" So, back to
the FCC database with that. Many results! But I started going
through'em anyway, clicking Details, then one of their external photos,
just to see what it was. Didn't take long, but the next thing I knew I
was staring at the picture of a Mailstation, and I couldn't believe the
other files I found in there with it. Short of a datasheet for the CPU,
a schematic of the whole thing is the next best thing you could have.<br>
So, curious of what else Mailstation-<wbr>related I might find buried
away in there (like perhaps the older models), I searched just for
grantees with that name, which you can do there on the left if you do
an advanced search. I got multiple results for Inventec Electronics.
One of them was listed as "Nanjing" though, which is in that text I
pasted above that I found on Google. It has the grantee code "PQU". And
sure enough, putting just that one into the search gives you only the
Mailstation results. Just the three, though, unfortuantely.<br>
Eventually I found that the number on the back of my Mailstation isn't
an FCC ID. It's an FCC Registration Number, which is different. The
older Mailstation is classified under "Part 68" equipment, which
apparently means it doesn't need extensive testing, since it's just a
telephone network device. However..! The Mailstation 250 and up use
wireless base stations to connect to their phone line. This means
they're using frequency spectrum, which means they require a full-on
FCC ID. I probably could have saved myself a few steps if I owned a 250
or 350 and had their FCC ID to begin with. But hey, whatever works. <br>
In any case, there is one interesting thing of note about the FCC
registration number. The first three characters, 1EB, are this in their
database:<br>
1EB|INVENTEC ELECTRONICS (MALYASIA)|37 ANSON RD|10400 PENANG||MALYASIA|<wbr>||08/29/1989|<wbr>D
REGINALD TIBBETTS|2151 CAMINO PABLO|MORAGA|<wbr>CA|94556|<br>
So, as it turns out, Inventec is definitely who developed the
Mailstation. Their name is on all of the schematics as well. And a
Google search shows their name in Cidco press releases/financial
reports/whatever.<br>
But here's the really interesting part: Inventec made a CPU for one of
the TI calculators too, apparently. Anyone who knows anything about
those graphing calculators know that they used Z80 CPUs for years. I
found this in a Google search, of a guy asking for information on a
chip:<br>
I'm in search of data for an Inventec processor. The chip came from a
TI-83+ graphing calculator as bares the text:<br>
Inventec<br>
6SI837 0A<br>
Japan 0216EAI<br>
PDARD<br>
F0254ZAA<br>
...if it helps, this is a Z80-enabled CISC processor. I've tried
googling and came up empty. Any data on pinouts, or clocking data would
be appreciated.<br>
Notice the similarity of the part number! The Mailstation'<wbr>s is
3SI176 0A. A Google search on the 6SI one gives some results, but I
found that I get other results when I put in "6S1837". Somebody seems
to have mistaken an I for a 1.<br>
So, I emailed Inventec, and I guess we'll see if my email ever gets
beyond the webmaster to anywhere that might possibly have information
on the Mailstation CPU.<br>
p.s. For anyone interested, here's the somewhat vague at times patent
application for what I'm almost certain is the Mailstation.<br>
//--unintelligible--//
In case that link doesn't work right, the patent # is 6,510,455.<br>
I almost debated posting anything here, but I'm fairly sure none of it
is illegal to do so, because it's publicly-accessible information.<br>
From: "cyranojones_lalp" <cyranojones_lalp@...>
Jan 23, 2010
Hey great! Now we know the functions of the pins th=
at are
not hooked to anything!!! ;^)
The names they put on the used pins=
don't seem as useful as the
names I gave them, though.
I googled the pin =
names on unused pins 91 thru 94,
[LP FP FR SCP] and got an interesting hi=
t:
(URL)
m=3D4&source=3Dgbs_overview_r&cad=3D0#v=3Donepage&q=3D&f=3Dfalse
patent nu=
mber is 5448259 just in case.
Seems likely there is a LCD controller in it=
, but I
can't say it is useful knowledge. But who knows?
grantee name, "Inventec Electronics."
It has been known who made it for =
a long time now:
(URL)
I'm sure(???) it=
was on l-h board.>
In addition to the ti calculators, they made the ipod =
for apple.
I thought it=
was interesting when I found it, too. ;^)
ed Inventec,
You speak Chinese? :-)
Just kidding, they prolly have *som=
eone* there who
speaks English... right??? :-) :-)
Wouldn't it be a ri=
ot if they sent the datasheet. I never
even thought to ask for it!
CJ
=
From: "FyberOptic" <fyberoptic@...>
Jan 23, 2010
nice and thanks! I have a few 120's, 150's, and 250's
. The 250 schematics are going to be a help
y like to have such info on the
Somebody will have to co=
rrect me if I'm wrong, but as far as I know, the 250 and under are all prac=
tically identical in terms of guts (aside from the wireless capability), wh=
ich would make the schematics fairly useful even for the older models. It =
all looks very similar to my DET1 from what I've gone over so far. The CPU=
is likely identical in any case, which was the biggest thing I was after i=
n terms of info. But a full schematic is nice too!
googling around, too. But I suspect that
you've already uncovered
Doesn't hurt to look. Sometimes a=
random search can put you on a certain track, which is how I found that st=
uff to begin with.
From: Jon Kirwan <jonk@...>
Jan 23, 2010
On Sun, 24 Jan 2010 00:13:58 -0000, you wrote:
oups.com, Jon Kirwan <jonk@...> wrote:
a few 120's, 150's, and 250's
going to be a help
on the
but as far as
=
e schematics fairly useful even for the older
similar to my DET1 from what I've
entical in any case,
info.
I don't see the LCD display pre=
sented on the schematics,
though I suppose the connectors I see probably su=
ffice.
I am finding nothing more than you've already uncovered
sn't hurt to look. Sometimes a random search can put you
ack, which is how I found that stuff to begin
Yes. One never know=
s. And not just google, I suppose. There
are a lot of different engines ar=
ound and some of the web
spiders are better at some things than others.
In=
any case, thanks for what you found already. Big help, I
think.
Jon
From: "FyberOptic" <fyberoptic@...>
Jan 23, 2010
p@...> wrote:
re
I'd have to look at the insides ag=
ain to see just how tiny those pins were (they always seem bigger in my hea=
d), but now that we know for sure that there's a sixth device select, it mi=
ght be pretty neat to take advantage of it. Something like compact flash c=
omes to mind, since the data and address lines would be easy to tap into fr=
om the eyelets on the board.
seem as useful as the
names on unused pins 91 thru 94,
hit:
ct&zoom=3D4&source=3Dgbs_overview_r&cad=3D0#v=3Donepage&q=3D&f=3Dfalse
patent number is 5448259 just in case.
ntroller in it, but I
=
Ah nice, I was wondering about those, and have been doing some poking aroun=
d myself for possible matches. I dismissed a possible LCD at first, but af=
ter seeing that patent, I think it seems a little too similar to be a coinc=
idence.
I'm thinking BZCLK might be a buzzer? The 350 has a buzzer connec=
tion for whatever reason. I don't know how that's used since I've never us=
ed that model before. But I'm assuming one of the ports would set the freq=
uency coming out of BZCLK. Could be neat to have. A clever programmer cou=
ld probably manage music of some kind out of there.
Looks like there's an =
unused interrupt available too. There's three bits unaccounted for on P3, =
so it's surely one of those.
All I've ever really known about the 350's g=
uts is that it "has a different cpu". I never looked into it further than =
that since I don't have one. I just didn't realize how different it was. =
This thing is 16-bit based on the data lines coming out of it. It also see=
ms to have two identical sized pieces of flash (which I'm sure 350 owners w=
ho have opened theirs already know), which at first you would think is a go=
od thing, but the sector size seems to be 4KB. Not quite as easy to work w=
ith at the 256-byte sectors of the old model dataflash. At least it has mo=
re ram for holding temporary sectors. And who knows how much the CPU can a=
ddress at one time. I'd love to know what CPU architecture it's based on, =
for curiosity's sake if nothing else.
One other tidbit I found out is fro=
m a trademark search. Apparently Cidco also registered "Surfstation".
=
From: "FyberOptic" <fyberoptic@...>
Jan 23, 2010
don't see the LCD display presented on the schematics,
the connectors I see probably suffice.
I noticed that, even though the =
350 schematics includes both the LCD and the keyboard matrix. So maybe the=
y're not quite as "full" as I originally implied. But fortunately, we alre=
ady know the keyboard and LCD!
(URL)
ion/files/part%20files/mailstation_LCD.doc
(URL)
oup/mailstation/files/part%20files/mailstation_keyboard.html
case, thanks for what you found already. Big help, I
No prob=
lem, I'm just glad to contribute. I was late to the Mailstation game and l=
ack certain tools to be of more help on the hardware testing side, but I st=
ill try to see what I can muster up once in a while in other areas. Good l=
earning experience, too!