ADVANTEST R9211B

[stfjohn]

Hi everyone. My Advantest R9211B FFT analyzer suddenly presents this message when turned on: "Fatal Switch - illegal state 0 in main - not installed!" . Can anyone help me?
Thanks.

[m k]

Just a guess

Different models have different options that are so close to each others that there are switches selecting what is present.
Now one of those switches has failed, so wrong option is indicated.

[stfjohn]

Thanks for your input. Unfortunately the device doesn't just not install an option: it just doesn't work! That message appears on the monitor and everything stays off!

[m k]

Hardware option.
Since one switch has a wrong value it doesn't recognize its guts.

During a boot procedure it tries to initialize all parts it has.
Wrong switch value indicates that it includes a part X.
But initialization fails because the actual part is Y.

[stfjohn]

Ok. Since I don't have the wiring diagrams (I asked Advantest, but they told me they are not available to provide them, not even for payment.... 40 year old device: will it be a state affair? Mah!!), what can I do concretely?

[m k]

You open it and start guessing.
Few pictures wont do any harm.

You can also try deciphering the message.
Not installed is clear.
Fatal switch can be a hardware thing, but combined with "illegal state 0 in main" it can be a software thing.

Hardware switch is simple.
Software switch, a parameter of main() resulting state 0, that is illegal, can be trickier.

But first thing first.
The machine must have all necessary voltages present.
Reseating all pinheader style connectors are not bad either.
Socketed chips can also be pushed up a bit by their sockets and so should be bushed back in.

[stfjohn]

The voltages are all present. I have already deoxidized all the connectors present and the sockets of the various chips (It's the first thing I did..). Unfortunately I really believe that it is a software switch, because the only hardware switch present works perfectly.

[stfjohn]

Here screen image...

[m k]

Now the message is different, you missed the underline, it's important.

So the main program has a parameter that is not allowed to be zero.
Unfortunately it doesn't say what is not installed.
But it clearly is a failed initialization of something.

Maybe connections at the back can give out some boot info.

Next you can check that all modules actually have their voltages.
There can be local regulators.

Bad capacitor can also pull down a small regulator, or prevent reset signal going off.
Top side bulky elcos are easy, but they can also be very evasive if popped from under side.

[stfjohn]

I tried removing the cards of the 3 options present: the result does not change. I think I can see a couple of reed relays on the motherboard: I will try to disassemble everything (which is quite complicated...) to see what components they are. The voltages indicated on the various printed circuits are correct, however.

[stfjohn]

Here some pictures of the motherboard.

[stfjohn]

Idem.

[stfjohn]

The battery is good!

[m k]

It needs more permanent memory.
Can't see what the gold top is, NS32 something?

Upper side is floppy and CRT.
Lower left has keyboard, controller has also an LCD style display part, beside it is 8k ROM. it's not enough.
Lower right is GPIB stuff.
Center square is not visible, but not needed either.
Center triplet under CPU is so center that the thing probably wouldn't display anything if there is a fault.

Since keyboard controller has that display part it can put out very primitive stuff, if implemented.
How is that lower left connected, both connectors in use?

In front of front euro connector are bus terminals.
Is there anything displayed if you operate the machine without any of to end side bus cards?

It should be possible to put any card to any slot, but who knows.
You can try different combinations.
Start filling from crystal end, if possible.

LS244 2nd left from upper crystal has some texture issues, some picture stuff only?
Those used to be less robust back then.

[stfjohn]

Thanks again for your kindness. The golden part is NS32081D-10J. The central square is TMS32025GBL. The Eprom board (Mitsubishi M5M27C101k-2) is connected to the J9 connector at the bottom left: 16 in total. The switch on the board in question works well. I tried to move the various contacts: normally 2 and 5 are ON. Putting them all OFF also makes the writing on the screen disappear, which reappears by putting 5 in the ON position. I tried to remove and put back all the boards, but nothing changes. Removing the Eprom board, however, all the LEDs on the front light up and everything remains still, obviously... In fact the LS244 and also the 74157 underneath seem a bit shabby...
Best regards.

[m k]

LS244 is 4+4 bus driver, F157 is 2to1 x4 data selector, easily connected together.
8 bus connectors are also easily 4+4.
But those chips are also constantly in use, so easily picking up some wear over time.

Good to know that missing ROM card has some visible effects.
Means that it's at least partially present and available.

So the gold top is a Floating Point Unit.
Its location is a bit odd, something like not central.
Is a connector above it in use?

Topmost bus connector seems to be different.
(is there a better name for eurocard connector)
Between it and next connector are more bus drivers.
Maybe ROM card and this topmost is a minimal installation.
Possibly some switch changes are also needed for that.

Very difficult to continue if only info is that earlier text or not.
I'm quite sure some primitive console type boot info is available somewhere.

Back has some trigger connectors above GPIB, are they from J19?

There is another pinheader set near by, maybe J17, is it in use?
It's not far from JTAG.

[stfjohn]

The connector above NS32081 is used to connect the optional printer. A (working) stabilizer board is connected to bus J8 which supplies power (+15V -15V -24V) to all bus below. J17 connects a video output on the back. J19 connects SMPLG CLK and TRIG inputs and outputs, always on the back.
I ordered the LS244 and F157 ICs. I will try to replace them.

[stfjohn]

I changed the LS244 and 74157 ICs, but, unfortunately, nothing changed: always the same message on the CRT.

[squadchannel]

There is a lack of information on R9211 on the Internet. (Advantest is also reluctant to actively publicize resources)

Currently, a maintenance manual is listed on eBay. It can be called a service manual as it also has schematics.

Can you purchase it and scan the PDF for future use?

eBay auction: #116111410582

[m k]

Would you donate a part of cost?

[stfjohn]

I don't see any service manuals with wiring diagrams on eBay! Only operating manuals! These can be downloaded easily from the Internet!

[stfjohn]

Sorry, I saw: in reality it is the R9211 not the R9211B: the power supply diagram already does not match: the 9211B uses switching power supplies: I would not want the rest to be different!

[squadchannel]

Sorry, I saw: in reality it is the R9211 not the R9211B: the power supply diagram already does not match: the 9211B uses switching power supplies: I would not want the rest to be different!

Really? In the second image, the title of the figure includes the letter "B".
changed? hmmm...

[stfjohn]

My device has n. 3 switching power supply "Nemic Lambda" !

[m k]

Since power is different all other parts can be so also.
You should contact the seller and ask some more pictures, component layouts should be enough.

Seems that letter versions are from the beginning, but actual timing is still a bit hazy.
One brochure includes 4 versions and F is missing.
If F is newer, SMPS can come with it, and be fitted to all machines later, for what ever reason.