Any 9100 owners out there?



#2

Just got a 9100B off TAS; hopefully, it's as functional as it appears to be. I'm crossing my fingers.

I'm having Larry Atherton rebuild the power supply and generally check it out; he regaled me with a half-hour lecture on how insanely complex the 9100 is; how it shouldn't even have been possible to build one with the technology of the time, and so forth.

He asked if I knew anyone else that owned one of these machines that he could talk to if need be. Well, I imagine Tony has one...anyone else?


#3

No, I don't have _one_. I have 2 :-) (both B's)

The design is simple in terms of component count, but complex in terms of how it's all intertwined. There are something like 44 flip-flops (only) in there -- 20 on each of 2 PCBs (these include the program counters for the firmware and microcode, the arithmetic digit registers, etc), 2 for the clock divider, 1 for the core memory page select and finally one for the error flag. The last is an SR, all the others are JKs.

There really is no need to 'rebuild the power supply'. In all the 9100s I've worked on, I think I've had one PSU problem (rectifiers for the main -15V line shorted, which blew the mains fuse). I've never had the regulator transistor fail. And I've never had to replace any capacitors.

You're welcome to talk to me about it. I've taken mine into rather more pieces than HP ever intended me to do, and I've traced out schematics...


#4

In the playing with ancient electronics I've done over the past year or so, I've found old capacitors to be a fairly common problem, in everything from an NEC Turbo Express hand held game to the 9835 I acquired recently.

Now, the only way I can tell if a cap is marginal or bad is if it's visually obvious: bulging ends or visible leakage or corrosion. But I understand they can be electrically bad, too.

That said, I have many old HP desktops with untouched power supplies that work perfectly. If my 9815 or 9825 die, they're fairly easy to replace, even today. But if the 9100 croaks, well, those are a lot harder to find!

Of course Larry Atherton knows about a zillion times more about this stuff than I do, so I'll trust his judgment.


#5

I think the Guru is Tony in this case and not Larry ;-P
I never met a person with more HP(and DEC and.... ) related hardware knowledge. To understand what I mean look at cctalk/cctech at classiccmp.org

;-) Rik


#6

Hmmm. I am not sure I'm ever going to be a 'guru' of anything. And quite a number over on cctalk seem to object to my postings.

But anyway. I will admit I've taken a 9100 apart. Really apart, meaning I desoldered the 4 parts of the ROM assembly from each other. I do have a fair idea as to how they work.

Feel free to ask...

#7

My experience (based on a _lot_ of vintage hardware, not just HP) is that capacitors can fail, but not as often as some will have us believe, and that the 'witch hunt' on old electrolytics in power supplies is unjustified.

The other issue is : why would you have to replace an entire 9100 just because a capacitor in the PSU has failed? It's unlikely to do much (if any) more damage when it goes. So you'd _then_ replace the capacitor and get the machine running again.

#8

Well done finding this unit David... it never appeared on my occasional searches, so I'll probably need to do some tweaking there.

Your 9100B looks to be in very good condition. Just be careful with the display CRT if you have to transport it, as I understand that is the only part that is pretty much impossible to find a replacement for.

Congrats and enjoy!


#9

Yeah, the one thing that makes me nervous is the odd appearance of the exponent in the first line of the display. On my monitor it looks as if a piece of black tape or something is covering the bottom segments of the X register exponent and part of the last digit of the mantissa.


#10

You know, your guess of 'bad capacitors' might just be right here...

The 'display task' -- the firmware that produces the display -- is synchronised to the mains to prevent flicker. There's a resistor connected to one end of the transformer secondary winding which drives a transistor on the right sideboard in the processor section for this.

Anyway, ripple on one of the CRT electrodes will blank out part of the display and it'll be pretty stable as a result of the above. I think my schematic of the PSU board gives the voltage _and ripples_ that I measured in my working machine. Use a 'scope to check the ones in your machine.

And then check the associated smoothing capacitors, etc


#11

Thanks for the pointers, Tony! It's not clear from the one photo of the display if there's really a problem or if it's some artifact of the image, but I should know shortly.

#12

I was watching it on E-bay, I didn't intend to buy it, but I wanted to see how much it went for. Way more than I could ever afford....

The CRT is impossible to find now. It was a custom part for this machine, it's an electrostatically-deflected one with a larger than normal deflection angle (I suspect this leads to some non-linearity in the display, but this doesn't matter for displaying digits).

Worse than that, the electrodes in the electron gun are supported by fragile glass rods. If these break (due to rough handling), it not only ruins the CRT but if the electrodes then short together it can apply high voltages to the logic side of the machine, taking out transistors and diodes all over the processor section. Ouch.

The HP service manual is mostly a boardswapper guide, but it's worth reading nonetheless. It has some information on the PSU and CRT circuitry, and a warning to remove and inspect the CRT if you suspect the machine may have thrown about. I do it anyway, but then I am not afraid of taking thing apart ;-)

#13

Hi David,

I own one HP9100A recovered just before dumping by a local research lab. It had a pair of shorted rectifier diodes in the power supply that I replaced quite easily (along with the related fuse) and it is now working fine. The magnetic card reader is ok too. I too wonder how many operative such machines are around nowadays, it would be nice to set up a sort of census about that.

Regards

Paolo


#14

Ah, so you got yours for free!


#15

Yes, actually it was bit of luck.
When I recovered it from dumping I even didn't know that the 9100A was such an interesting historical item, I simply liked the old-fashioned appearance. Then while looking for repair suggestions I discovered this forum and learned some more.

Still I would like to know how many such 9100A or B units are preserved today: I know about five or six working machines that were mentioned in the forum but I would guess there are a lot more.

Sometime I wonder how to start a census within the forum, or possibly even within a larger target including science museums and so on.


Paolo


#16

You know, it would not be difficult to design a MySQL database schema, with a simple PHP front end input form and report module, that would allow forum members to input whatever calcs they have so that we could have some statistics.

Of course we'd need support from Dave Hicks. Is he still alive?

#17

I have an HP9100A and I look forward to know how many of us have one.
I get it in 1987 from HP by a HP2000 service engineer back then and still works ok.


Possibly Related Threads...
Thread Author Replies Views Last Post
  HP-9100 used in computer-based education (video) Juergen Keller 5 306 12-07-2012, 11:04 AM
Last Post: hpnut
  9100 Documentation Charlie Brett 4 232 08-24-2012, 02:59 AM
Last Post: Bruce Larrabee
  HP 9100 display trick Juergen Keller 4 217 05-01-2012, 03:15 AM
Last Post: Bill Wiese
  HP-12C spiral bound Owners Handbook Ron S 3 186 02-29-2012, 07:02 AM
Last Post: Gerson W. Barbosa
  9100 Keyboard Rebuild (with pics!) David Ramsey 4 415 11-08-2011, 05:48 PM
Last Post: David Ramsey
  9100 printer paper David Ramsey 2 172 11-07-2011, 06:08 PM
Last Post: David Ramsey
  Is that an HP 9100 in the 1970 version of the Andromeda Strain movie? Gene Wright 1 120 01-01-2011, 10:06 PM
Last Post: Katie Wasserman
  9100 B on TAS goes for $5600.00 Donald Williams 10 359 10-12-2010, 12:30 AM
Last Post: Keith Midson
  HP owners actually use them? Dimitri Simitas 36 1,183 03-14-2010, 04:01 PM
Last Post: Egan Ford
  MLDL2000, Clonix/NoVRAM/NoV32, PILBox ... Users/Owners? New stuff? Vieira, Luiz C. (Brazil) 6 262 10-26-2009, 02:11 AM
Last Post: Ángel Martin

Forum Jump: