Calling Fluke 9100 Users

guddler

Busting vectors like it's 1982!
vacBacker
Feedback
10 (100%)
Credits
4,058CR
Was just wondering if there are many of us here and how much you've done with your unit.

I've just been looking at a Tempest PCB for the museum and took the opportunity to redo one of my scripts and it just got me wondering how to do some more complex stuff.

I tend to do this stuff in fits and starts about 2 or 4 years apart, so not often
smiley36.gif
 

guddler

Busting vectors like it's 1982!
vacBacker
Feedback
10 (100%)
Credits
4,058CR
smiley36.gif
Well, ordinarily I'm not really doing anything any more than you would be able to do with the 9010. The one thing that I found really handy when I first got it, and it will sound such a simple thing... you can apply a mask to a ROM or RAM test.

The consequences of that run deeper than it initially sounds. On the 9010, say you have a pair of 2114's. You have to test them as a pair since you cannot apply a mask. It's not a big deal as they're simple enough to diagnose anyway but with the 9100 you can test the lower 4 bits or upper 4 bits individually and therefore individual 2114's. Of course I'm pretty sure that if you script it on the 9010 then you can split them down anyway.

I can't remember if it is Centipede or Missile Command offhand but one of them (maybe both?) has the data striped across a bank of RAM in a pretty funky way and that mask really comes into it's own there. Think it's Centipede.

Anyhoo, I just went scanning through the TL/1 reference manual and I'm 99% sure Ive got a way rattling around in my head now that I can parameterise ROM and RAM blocks into config files to make the definition of UUTs far simpler to set up using MAME code. Will have to have a play with it before I go off the boil again!
 

Equites

Chief Sheesher®
vacBacker
Feedback
35 (100%)
Credits
3,329CR
guddler said:
smiley36.gif
Well, ordinarily I'm not really doing anything any more than you would be able to do with the 9010. The one thing that I found really handy when I first got it, and it will sound such a simple thing... you can apply a mask to a ROM or RAM test.

The consequences of that run deeper than it initially sounds. On the 9010, say you have a pair of 2114's. You have to test them as a pair since you cannot apply a mask. It's not a big deal as they're simple enough to diagnose anyway but with the 9100 you can test the lower 4 bits or upper 4 bits individually and therefore individual 2114's. Of course I'm pretty sure that if you script it on the 9010 then you can split them down anyway.

I can't remember if it is Centipede or Missile Command offhand but one of them (maybe both?) has the data striped across a bank of RAM in a pretty funky way and that mask really comes into it's own there. Think it's Centipede.

Anyhoo, I just went scanning through the TL/1 reference manual and I'm 99% sure Ive got a way rattling around in my head now that I can parameterise ROM and RAM blocks into config files to make the definition of UUTs far simpler to set up using MAME code. Will have to have a play with it before I go off the boil again!

Interesting stuff, it is Centipede!
 

backflipper

Active member
Feedback
4 (100%)
Credits
774CR
I haven't used mine in ages. I really want to get back into it though.

I will be firing mine up next February. You thinking of pooling information on here? If so I would love to contribute.
 

guddler

Busting vectors like it's 1982!
vacBacker
Feedback
10 (100%)
Credits
4,058CR
Not sure what I was thinking really! Pooling information is a difficult one. Oll mentioned it last weekend. I'm cool with it as long as it's a two way thing but the problem is that there will always be leechers that never contribute and some of them could even use it for commercial gain and I don't really see why we should fund that.

Which is a good point actually, on the flip side of the coin, I couldn't ever rule out having to do commercial repairs to supplement my income should my circumstances change so I guess the above comment could be equally applicable to me!

One thing I'd like to know is if there's any way to restore system settings from a UUT programatically, or most notably set the screen colours programmatically. I've not found a way to do it yet and have to do it manually each time it boots because I hate the default white on blue background of the colour monitor. I'm guessing you'll be using your B/W one though?

I think I'm right in saying that the video card, when jumped for B/W expects a 5V DC input on the RGB connector so power some chips on the card. It's a seriously bizarre thing for them to have done and makes it nigh on impossible to use a colour monitor in B/W mode (or a non Fluke B/W monitor). Well, that's what I assume and only if my memory serves me right anyway.

guddler2014-11-23 12:48:16
 

backflipper

Active member
Feedback
4 (100%)
Credits
774CR
That is strange. You woukd expect that once you set it it woukd be saved in non volatile memory.

I can't remember who but I know that someone(in the US) is booting their fluke directly into os9 which may open up more possibilities. There is loads more info on the tech tools list about the os9 operating system.
TTL Fluke

Also I have seen someone had a real time clock running on a plug in card but I can't remember if that is standard for all flukes 9100a/ft.

Pman is a super user of this system and has done some really cool stuff on it.backflipper2014-11-23 13:24:06
 

guddler

Busting vectors like it's 1982!
vacBacker
Feedback
10 (100%)
Credits
4,058CR
The realtime clock works on mine. I had to replace the RTC chip when I got it as the internal battery had long since expired. I noticed the other day that it was about 18 minutes out but frankly that ain't bad for 6.5 years! I think the clock is standard on 9100's, but since mine is the FT I can't vouch for the others.

Booting direct to OS9 sounds interesting. I think!!

Anything to make working on scripts remotely easier would be very welcome. You can do it and transfer the files via serial but I seem to remember it's a bit on the clunky side. I have notes somewhere, in the same folder that the backups of my CF card are I think - wherever that is
smiley36.gif


I left the Flippers list in the end. There was not much on there and something irritated me enough to throw the towel in (or someone). Who knows. I'm a miserable bugger at times.
 

backflipper

Active member
Feedback
4 (100%)
Credits
774CR
Yeah I was transferring to and from the fluke via rs232 using a terminal program on my laptop.

I need to order an aztermonster cf-scsi adapter and then I will be able to start easily messing about with different disk images.
 

DanP

Administrator
Staff member
vacBacker
Feedback
5 (100%)
Credits
2,232CR
guddler said:
smiley36.gif
Well, ordinarily I'm not really doing anything any more than you would be able to do with the 9010. The one thing that I found really handy when I first got it, and it will sound such a simple thing... you can apply a mask to a ROM or RAM test.

The consequences of that run deeper than it initially sounds. On the 9010, say you have a pair of 2114's. You have to test them as a pair since you cannot apply a mask. It's not a big deal as they're simple enough to diagnose anyway but with the 9100 you can test the lower 4 bits or upper 4 bits individually and therefore individual 2114's. Of course I'm pretty sure that if you script it on the 9010 then you can split them down anyway.

You can apply a mask to the ram test on the 9010, you just have to write your own test. I gave up on ever owning a decent 9100 or ever having the time to learn how to use one. I use the 9010 (well used to) and wrote a script for ROTJ which has some single 2114's used 4bit as you describe. Below is a excerpt from the script which hopefully will show what I did to test them. For new 9010 users everything after a ! is a comment :)

This section is one of the controlling tests (usually user selectable). It just set's the memory range of the 2114 we're looking at and calls the testing routine further down in the script. For 8bit ram we would normally just run the "ram short" command with the address range.

*****************************************************************************

program RAM3

dpy RAM test @ 2400-27FF IC 7N

aux RAM test @ 2400-27FF IC 7N

REG8 = 2400

REG9 = 2800

execute lowbit

*****************************************************************************

This bit is the mask apply and go up down the range selected.

*****************************************************************************

program lowbit

!aux Low Test Started

REG1 = F0 ! mask

REG2 = 0 ! up or down 0 = up, 1 = down

REG3 = 0 ! times thru loop

REG4 = REG8 ! Store passed start address

REG5 = REG9 ! Store passed stop address

loopit:

REG8 = REG1 ! Store mask to use

REG9 = REG4 ! Store address to be tested

execute go ! Engage test

INC REG4 ! Increment address

if REG4 = REG5 goto exitloop ! Test if we've hit the stop address

!if REG3 = 2 goto exitloop ! Uncomment to limit runs thru up and down loops (for script testing only)

IF REG2 = 0 goto upit ! Are we going up?

IF REG2 = 1 goto downit ! Are we going down?

upit: INC REG1 ! Increment mask

IF REG1 = FF GOTO upREG2 ! If we've hit top of mask range set to go down range next run

goto loopit ! back to start loop

downit: DEC REG1 ! Decrement mask

IF REG1 = F0 GOTO dnREG2 ! If we've hit bottom of mask range set to go up next run

goto loopit ! back to start loop

upREG2: INC REG2 ! Increment direction indicator register

INC REG3 ! Used for limiter - used for testing only

GOTO loopit ! back to start loop

dnREG2: DEC REG2 ! Decrement direction indicator register

INC REG3 ! Used for limiter - used for testing only

GOTO loopit ! back to start loop

exitloop: ! Return to sender

!aux Lowbit Test Complete

program go

REG1 = REG8 ! Store passed bit mask

WRITE @ REG9 = REG8 ! Write bit mask at target address

READ @ REG8 ! Read back target address

IF REG1 = REG8 goto exitexec ! Compare stored mask with what was read

dpy TEST failed $8 @ $9 not REG1

aux TEST failed $8 @ $9 not REG1

stop ! if we hit here there was a problem - halt the program and display error address

exitexec: ! Exit routine and display success message on rs232 terminal

!aux TEST $8 @ $9 Succesfull

! Uncomment if you want to see progress

*****************************************************************************

I'm sure the 9100 has much more comprehensive in built test available but this is all I have :)

Cheers,

Dan
 

guddler

Busting vectors like it's 1982!
vacBacker
Feedback
10 (100%)
Credits
4,058CR
It does - but the bottom line is, you only have to write what you pasted above once and that's it, it's done then. Nice one. I figured it had to be possible.
 

backflipper

Active member
Feedback
4 (100%)
Credits
774CR
I also need to make an RGBiS(CGA TTL) to RGBS adapter so I can connect my fluke up and get some wonderful colour.

Well actually I have a very small space to create my work bench and I was thinking of utilising my test monitor for both the fluke and my jamma rig.

I do like the green phosphor fluke monitor though...
backflipper2014-11-23 23:36:28
 

guddler

Busting vectors like it's 1982!
vacBacker
Feedback
10 (100%)
Credits
4,058CR
I've just got mine wired up direct to a Hantarex Polo if that helps...

Is the video card in the Fluke TTL? I'm beginning to wonder if I modified mine. I do remember spending a lot of time studying the schematics when I got it.
 

backflipper

Active member
Feedback
4 (100%)
Credits
774CR
guddler said:
I've just got mine wired up direct to a Hantarex Polo if that helps...

Is the video card in the Fluke TTL? I'm beginning to wonder if I modified mine. I do remember spending a lot of time studying the schematics when I got it.

I always thought it was, it's been a while so I will have to check again...
 
Top