9114a open pics & strange error msg



#5

Hi,

Following Tony's post, I have found a place to order some picofuses (and have compiled the available information on the forum with regards to those as well, which will go into my summary of this 'repairing the 9114' story)

In the meantime I have built a no-battery power source, using the information from some old posts about a suitable Radio-Shack charger. Carefully observing the polarity, I built a nice new plug directly into the 9114. And, it works!!! Sort of...

1) I do have a 9114b as well, which I had not opened and cleaned as the eject seemed to work just fine. Using this direct power-source I was able to initialize a disk (using brand new 2DD, not HD).

2) However, the 9114a, the one I started working on, is not quite working. Here is what happens:
a) when started, it moves the head a bit and then the 'fault' light goes off. If a disk is inserted on power-on, it takes a tad longer, but eventually the 'fault' light goes off as well
b) now I can talk from my 71b to the drive just fine. Things like assign, spoll, devaid, devaid$, etc all work just fine.
c) however, when I try to initialize a diskette, it starts spinning, I can see the head moving a couple of steps and then it stops with an error.
Error Nr: 255026
Err MSG: Invalid Medium - Data Checksum Error detected. The suggested resolution is to Initialize the disk. Dooh! That's what I am trying to do...

So I have posted a few pictures of the open drive, with and without disk inserted in case I put the thing back together in the wrong way. Also I did some close-ups of the drive head. I can not compare it to the 9114b and I do not have another 9114a to check if my drive head is bent or somewhat out of shape. This would explain that I can talk to the drive as long as I do not want anything from the disk as the head can not read.

Please let me know of any ideas.

Until then

Your faithful student

Peter

Pic 1
Pic 2
Pic 3
Pic 4
Pic 5
Pic 6
Pic 7
Pic 8
Pic 9
Pic 10


#6

Do you still have problems if you have the metal screening cover on the drive (as you mentioned in the other post)? If so, we need to sort that out first.

Jamming the head movement shouldn't cause the thing to draw excessive current -- the head motor is (as expected) a stepper motor that draws much the same current whether it's turning or not. Jamming the spindle motor would cause its current to go up, though.

Is it possible you've trapped a wire somewhere, maybe one of the ones going to the stepper motor or the head load solenoid? a short to earth there would cause excessive current.

The fact that the non-disk commands (DEVID$, etc) work find suggests that most of the controller board is working properly. At least the HPIL interface, processor, ROM, RAM and so on.

I don't see anything wrong with the way the disk holder is assembled, or with the head. If the upper head was slightly out of position, you'd have problems reading disks written on other drives, but it should be able to format its own disks just fine. So I don't think that's the problem.

Another thing to try is to format a 'scratch disk' in the 9114B, write some files to it, and try CAT'ing it in the 9114A and reading the files back. If it fails, try the disk again in the 9114B (I want to see if it's corrupting disks, for example if the write current is always on).

After that, it's time to get into the more complicated stuff. Somewhere I have details of a cable to link the drive mechanism to a PC parallel port, and an MS-DOS program (turbo pascal IIRC) that lets you move the head around, check the track 0 sensor, etc. I used it to do a head alignment (along with a special CE disk) before I got a proper drive exerciser. I can see if I can find that. You'll need a 'scope or a logic probe to do much more on the drive, though.


#7

If that unit is drawing excessive supply current for some reason, it might be loading your homebrew PSU too much, causing the voltage to drop (or it might be loading one of the internal regulators too much and cause its output to drop). That might cause all sorts of problems.

If you have an ammeter, connect it in series with the power supply (an easy way to do this is to connect it across the power switch on the back of the 9114, with the switch turned off. The circuit is then completed via the ammeter). A good drive will draw a little over 1A when running.

Also check the supply voltage testpoints I mentioned in the first thread when the unit is trying to initialise a disk. Maybe one is low for some reason.


#8

Tony,

Your patience is truly amazing and I can't tell you how much I appreciate it. I'm learning a ton here! Okay, I did all that you suggested, please find the results below, it seems we found something that is not as it should be :-)

1) I measured the current the unit is drawing on power-up and selftest and it is a bit less than 1A (BTW - this caused me to buy a better DMM. Fluke was a bit out of my range but I went for a Sperry DM-4400A and it immediately improved my handling of measurements and the like. And it has an amp-meter which my prior 5.95 USD pocket DMM did not have) So that is okay, if a bit low according to your expectations

2) I found a Service manual of the 9114a which has a few self-tests in there. I tried to set the jumpers correctly and execute the various self-tests (1-7). The first one, which tests the RAM, ROM, FDC abd PIA, worked flawlessly. This confirms what you suspected already based on the good execution of the standard HP-IL commands

3) Then I also found the TP with the help of your schemata and the PCB (which is nicely labeled I must say). I can only try to imagine how much painstaking work it must have been to draw those schematics!

a) In the post which described the home-made PSU with a charger from Radio-Shack, it was mentioned to run the charger at 6.5V. I did the measurements below at both 6V setting and at 6.5V setting. The values in brackets relate to the test with the 6.5V setting.

a) I tested VC on TP8, which was at 4.97V (4.97V)

b) I tested +5V on TP 10 which was 5.02V (5.02V)

c) ! I tested 12V on TP9. And this one only showed 5.76V (6.22V) !
So there seems to be clearly something wrong here!

d) I also did the test you suggested with 1) writing something on the disk with the 9114, then 2) run a test, startup, initialize etc with the 9114a and 3) try to read the disk in the 9114B again. This test worked. cat :1 showed the very same files I had copied onto it.

So, it seems that something with the 12V circuit is broken. Is there any chance that we can figure out what without the scope/logic probe? It so happens that I was able to find a HP-IL 1631D Logic Analyzer and Digital Scope for rather cheap (I believe), yet i have not used a Scope in ages and never was very good at it at all. This was just the last standard HP-IL device I did not have yet and it was a good bargain.

Anyway, as always this late at night I leave my test-bed here with great excitement about the things I have learned and with great anticipation of the new wonders TD will (hopefully) send me next :-)

All the best

Cheers

Peter


#9

Quote:
1) I measured the current the unit is drawing on power-up and selftest and it is a bit less than 1A (BTW - this caused me to buy

Well that's interesting. The picofuse in the HP battery is rated at 2A (early battery packs) and 5A (later ones). So something is causing this unit to draw too much supply current only under specific circumstances.

I assume you were measuring the current while the drive was spinning and/or the head moving during the self-test. That's when the unit draws the highest current

Quote:
a better DMM. Fluke was a bit out of my range but I went for a Sperry DM-4400A and it immediately improved my handling of measurements and the like. And it has an amp-meter which my prior 5.95 USD pocket DMM did not have) So that is okay, if a bit low according to your expectations

I must admit I tend to go totally OTT when it comes to tools and test gear.
Quote:

3) Then I also found the TP with the help of your schemata and the PCB (which is nicely labeled I must say). I can only try to imagine how much painstaking work it must have been to draw those schematics!


Actually, the 9114 schematics didn't take that long. The HP9100B was a lot worse. I'll explain. In the 9114, you have a number of 'large chips' that have well-defined functions. There's a 68x09 processor. It can only be used in one way. It needs a clock input, it needs reset, it has address, data, and control buses that go to the other chips on the board, and so on. The only 'inventive' part is the PSU. OK, there were a couple of little bits of logic to sort out, but most of it could realistically have only been wired one way, and was wired that way.

Now think of a board of simple logic chips -- say a board from a 9810. You come across a 7400. It's a quad 2 input NAND gate. Now those NAND gates could be used in many ways, they needn't all be used in the same part of the circuit. Maybe 2 are cross-coupled to make an SR flip-flop. Maybe one is cross-coupled with a NAND gate in another chip to make an SR flip-flop. Get the idea. A lot more to work out.

And now think of the 9100. No chips at all in the logic section. Just boards of transistors and a big board of mostly diodes. Now a transistor could be almost anything -- part of a flip-flop, part of a buffer circuit, even part of a power supply. You've got to do a lot of work to have some idea of the overall stuff on the board.

Quote:

a) In the post which described the home-made PSU with a charger from Radio-Shack, it was mentioned to run the charger at 6.5V. I did the measurements below at both 6V setting and at 6.5V setting. The values in brackets relate to the test with the 6.5V setting.

a) I tested VC on TP8, which was at 4.97V (4.97V)

b) I tested +5V on TP 10 which was 5.02V (5.02V)

c) ! I tested 12V on TP9. And this one only showed 5.76V (6.22V) !
So there seems to be clearly something wrong here!


The 12V supply is only turned on when needed -- that is when the disk drive is in operation (the 12V line is used for the motors). When it's turned off, TP9 will show the battery voltages less a little bit. Connect the meter to TP9 again and see if the voltage goes up to 12V when it's using the disk (e.g. during the self test or a CAT operation).

Quote:


d) I also did the test you suggested with 1) writing something on the disk with the 9114, then 2) run a test, startup, initialize etc with the 9114a and 3) try to read the disk in the 9114B again. This test worked. cat :1 showed the very same files I had copied onto it.

So, it seems that something with the 12V circuit is broken. Is


Given that the thing can read a disk, I think not. The motors must be turning.

More likely it's a problem with the write circuit or the head selection circuit. This is on the PCB attached to the drive itself, and is mostly discrete SMD components. Actually, I've had a 9114A in for repair where a couple of those transistors had failed. I replaced them with pretty generic types and had no problems.

One thing that might not be obvious. On my schematics, a transistor with a circuit round it is just that -- a transistor. One with a square box round it is a 'digital transistor' -- it includes base resistors, etc.

Quote:
there any chance that we can figure out what without the scope/logic probe? It so happens that I was able to find a HP-IL


Maybe, but it's a lot easier with the test gear.

Quote:
1631D Logic Analyzer and Digital Scope for rather cheap (I believe), yet i have not used a Scope in ages and never was very good at it at all. This was just the last standard HP-IL device I did not have yet and it was a good bargain.

Ooooh very nice. I hate to say this, but the way to get good at using test gear is to use it -- to practice. And this is a fairly good thing to practice on.


#10

Tony,

Thanks for your kind reply. makes a lot of sense why the 9100B were so much worse. yet I am still very very impressed...

Quote:
The 12V supply is only turned on when needed -- that is when the disk drive is in operation (the 12V line is used for the motors). When it's turned off, TP9 will show the battery voltages less a little bit. Connect the meter to TP9 again and see if the voltage goes up to 12V when it's using the disk (e.g. during the self test or a CAT operation).

As for the 12V - I did measure it while the drive was spinning (during the selftest on turn-on with a disk in there and during the tried Initialize) and it did not go to the 12V...


Quote:
Given that the thing can read a disk, I think not. The motors must be turning

Sorry, I was unclear. The thing CAN NOT read the disk! What I did was write to a disk in the 9114b, then try to run some tests on the 9114A and then put it back into the 9114B to see if the 9114B can still read it. Your idea here was that the 9114A has constant power to the head and this would have made the disk non-readable for the 9114B afterwards. Which, alas, is not the case.

Okay, so I will most definitely wait for the 'scope and then try to learn how to use it, maybe there will be some specific tasks related to fixing this 9114A you can think of.

In the meantime, is there anything with regards to the 12V circuit you can think of? As I mentioned, even when the drive is spinning and the head is moving during the self-test and the tried initialize, the voltage does not go to 12V...

Thanks a lot, I know already what I will be doing tonight :-)

Cheers

Peter


#11

Quote:
In the meantime, is there anything with regards to the 12V circuit you can think of? As I mentioned, even when the drive is spinning and the head is moving during the self-test and the tried initialize, the voltage does not go to 12V...

OK, you have a problem with the 12V step-up converter, this is essentailly sheet 2 of the schematics on the Australian Site.

U108d and U107 form an oscillator. It can be controlled by U108c which compares divided-down version of the 12V line with the reference voltage. The osciallator also feeds a monostable U106 via U108b. The output of that drives the swtiching MOSFET Q105. The 12V supply really comes from the back EMF on L101 when Q105 turns off.

Now, since you have battery voltage at the 12V TP, L101 (switching coil), L102 (output filter coil) and CR103 are most likely fine.

I would start by looking at the output (pin 3) of U106. There should be oscillations here. If you can't use a 'scope, maybe connecting a meter here will show something around half the battery voltage, but I'd prefer to check it was actually oscillating.

My _guess_, and it's only a guess, is that Q105 has failed. At least I've given it a standard type number

#12

Tony, made a few more measurements tonight, but nothing big to add to my other post ('12V missing')

1) when i do the Amp measurement, the drive does come up and does the self-test (and only draws about 0.75-0.85 amp), the battery led is blinking. I could not find what this means. Also, the fault light does NOT go off in this mode. Other than that, the drive does all the same things.

2) I doubble checked the 12v point and it does not go to 12V when the motor is running and the head is moving. As a matter of fact, it never goes to 12V or even close, no matter what command I send or what the drive does

3) I also measured the other TPs but only TP6 and TP7 had a fixed voltage (around 5v, just a taf below around 5.97 or so)

I tried to find the other components of the 12v circuit on the board but did not get very far... One thing that puzzles me - the spindle motor needs 12V (it is even marked on it) yet the 12V TP does not show 12v. How come the motor still spins?

Anyway, I'm awaiting further instructions if at all possible, yet I will be out until Friday next week when I can test them. In the meantime I was able to find the manual and even the service manual of the 1631D online - Agilent is amazingly well organized. I was able to download the manual and service manual for the 3468 A/B DMM as well!

Cheers

Peter


#13

Quote:
Tony, made a few more measurements tonight, but nothing big to add to my other post ('12V missing')

1) when i do the Amp measurement, the drive does come up and does the self-test (and only draws about 0.75-0.85 amp), the battery led is blinking. I could not find what this means. Also, the fault light does NOT go off in this mode. Other than that, the drive does all the same things.


A blinking power LED means low battery voltage -- see the lower part of page 2 of the schematics. If the battery voltage is too low (detected by U104A), then Q104 turns on, connecting C104 in parallel with C105. This slows down U102, so you can see the LED blink (actually, in a 9114A, the power LED is always blinking, it's just normally too fast to see).

Quote:

I tried to find the other components of the 12v circuit on the board but did not get very far... One thing that puzzles me - the spindle motor needs 12V (it is even marked on it) yet the 12V TP does not show 12v. How come the motor still spins?


Well, it is getting _some_ voltage -- about 6V from the battery. It'll probably run on that, albeit probably too slowly. But it's not suprising the whole thing doesn't work properly without the 12V supply.


Possibly Related Threads…
Thread Author Replies Views Last Post
  OT--Nat'l Semiconductor NS-7100 pics anywhere? Matt Agajanian 25 11,393 12-15-2013, 08:39 AM
Last Post: gene wright
  Strange HP 25 Problem (Repair) Onur Ilkorur 5 2,978 12-06-2013, 05:13 PM
Last Post: Onur Ilkorur
  Strange Battery Icon during updaate of Prime Firmware. Harold A Climer 7 3,605 12-05-2013, 04:40 PM
Last Post: Michael de Estrada
  MS advert shows spreadsheet with obvious error BruceH 3 2,253 11-14-2013, 09:50 AM
Last Post: Bill (Smithville, NJ)
  HP PRIME : strange behavior when trying user key capability Damien 12 4,027 11-03-2013, 11:02 AM
Last Post: Joe Horn
  HP Prime: Rounding error in determinant Stephan Matthys 3 1,704 10-25-2013, 09:29 PM
Last Post: Walter B
  Prime Error or Mine? toml_12953 12 3,808 10-22-2013, 10:35 AM
Last Post: toml_12953
  Explaination on How to Reset Caculator in Users guie: error Harold A Climer 5 2,652 10-15-2013, 02:11 AM
Last Post: cyrille de Brébisson
  Repair of HP-34C - Error 0 and Error 9 Jeff Kearns 3 1,754 10-11-2013, 12:29 PM
Last Post: Randy
  HP Prime function APP - Strange limitation ! :o( dg1969 2 1,632 10-04-2013, 12:10 PM
Last Post: dg1969

Forum Jump: