The HP9845B 16 bit computer was revolutionary in 1979 and the early 1980's, find out why.
The 9845C colour version of this created the amazing war room images for the movie War Games
I wonder about that computer, at around the same time I used another HP computer that was much more user friendly.
It had a keyboard with a row of soft keys below the CRT. These keys was used to compose commands to the computer. This may not be anything special, but the computer uses a HPIP (Later named GPIB) connection to a harddisk unit to fetch the software needed to perform the command and the harddisk could serve multiple computers. I must admit that I did not really use that computer, a Univac 1100 computer was much more fun to use and I also bough my first home computer at about the same time (A NASCOM2 Z80 based one).
We also had a P-Code machine, that was another type of computer and it has gone completely out of fashion. It was fast at that point in time, but being a very advanced CISC it was sidelined by RISC computers and the base language never got popular enough to maintain it either.
Could someone with more EE skills explain the ferrite beads on capacitors?
My understanding:
- (ideal) capacitors block DC and likewise form a low pass filter, but inductance lets higher frequencies through.
- ferrite beads block very high (RF) frequencies, so it's a high frequency low pass filter.
So it's just blocking high, and low, frequencies but not a middle band? So it's a bandpass filter?
There must have been a very specific purpose for having them on every cap, since nobody else does it. And putting them _attached_ to each cap must be for one/both of two reasons:
- Package convenience instead of populating two slots on the board
- Extreme locality for minimum... I don't know the EE term... noise/interference between two coupled components.
Anyway, if I'm right/wrong, I'd love someone to give me some background information. Thanks!
--Chris
Found my answer!
"3-lead capacitor, 0.22UF/ 50V, with ferrite beads on two outer leads. The three-terminal structure
decreases residual inductance and greatly improves the noise suppression effect at high frequencies greater than several dozen MHz."https://www.allelectronics.com/item/223d50-fb/emi-suppression-capacitor-w/ferrites/1.htmlAlso interesting is that they can form resonates if you select the wrong values:
I would have loved to put a scope on that thing and see how incredibly noisy it is between the heavy # of jellybean logic ICs and the non-angled traces.
Guarantee those are spacers. Nice color and shape, wouldn't be surprised if they were stock necklace beads or something?
Dave, noticed a couple glitchy pictures, like around 3:49 the computer is on screen for 4 frames then the previous image for 1 then it's alright after that. Hope the rendering's not going funky on ya!
Tim
Wow that takes me back. I did contract programming in BASIC to read data from a HPIB connected (isotope measurement?) equipment that the company sold, to process to extract key information and plot a graph. The computer was installed in a medical lab at RPAH Sydney and I used to do my programming at night when the computer wasn't being used. Unfortunately I've forgotten many details, as I was a student at that time, and didn't practise taking notes then.
Dave, noticed a couple glitchy pictures, like around 3:49 the computer is on screen for 4 frames then the previous image for 1 then it's alright after that. Hope the rendering's not going funky on ya!
Pretty sure that's an edit goof, I was having issues editing.
Could someone with more EE skills explain the ferrite beads on capacitors?
My understanding:
- (ideal) capacitors block DC and likewise form a low pass filter, but inductance lets higher frequencies through.
- ferrite beads block very high (RF) frequencies, so it's a high frequency low pass filter.
So it's just blocking high, and low, frequencies but not a middle band? So it's a bandpass filter?
There must have been a very specific purpose for having them on every cap, since nobody else does it. And putting them _attached_ to each cap must be for one/both of two reasons:
- Package convenience instead of populating two slots on the board
- Extreme locality for minimum... I don't know the EE term... noise/interference between two coupled components.
Anyway, if I'm right/wrong, I'd love someone to give me some background information. Thanks!
On 2nd thought I think they are ceramic type spacers, used to prevent any top side shorts when reflow soldered.
Talk about over engineered...
So a few thoughs about this.
Not a single bodge wire, I think that might be unusual for the era, for low quantity productions such as this.
The CPU could possibly be bond wired onto a PCB. But I dont think there was a way to attach it to the heatsink and the PCB the same time.
That keyboard itself looks great, I really like how the keys are in good shape, color after 40 years.
I think the power supply was a bit over the top though.
I also wonder what was the reliability of these computers. And if IT had a supply of replacement boards at hand so they could fix a computer on the moment's notice. I mean these things would be used by the dozens at eg. NASA.
You should desolder those Rifas and then connect them to 240v outside, fun fireworks for the whole family to enjoy!
Damn, there's a lot of stuff in that box...
Absolutely amazing engineering. No wonder HP had that great reputation back in the day. Everything about this machines just oozes quality.
Modern hardware is just too much commodity to justify a build quality like that. No one would be willing to pay even half the prices.
Those memory modules are weird in these early 16bit HP machines. They would double up 8 bit RAM chips to get 16bit space. So no, that's not 64K of memory on that one board. That's 32K of 16bit memory of two chips each and there would be custom memory drive circuits. "But a byte is 8-bits, that makes no sense!" It's not Bytes...it's kilowords.
See Curiousmarc's series on the earlier single text line 9825 for more details on this wacky stuff.
Modern hardware is just too much commodity to justify a build quality like that. No one would be willing to pay even half the prices.
It would double the price for no perceivable benefit. The level of integration in modern PCs means there's not much you can repair anyway.
You can spend a bit extra for Japanese Capacitors(TM) in your PC power supply if you want:
OTOH nobody needs to put their RAM modules on little riser boards so they can probe them with an oscilloscope while the PC is running.
did anyone notice the very early PCB layout tool shown on that brochure cover ? Anyone got a copy of that laying around ... i'd like to see that run . Probably most of the HP pcb's were designed on that.
https://www.hp9845.net/9845/hardware/9845c/images/9845C_EGS.jpgsure enough :
98302-12111 9845
98302A EGS/45 Schematic Drawing Manual
82/07
98302-12501 9845
98302A EGS-45 Schematic Drawing Chapter Manual
82/01
98303-12111 9845
EGS/45 Printed Circuit Layout Chapter Manual
82/07
they had a schematic/ pcb tool for this machine !!! i wo wanna see this.
Man, I learned Basic on one of those machines. It came with a tutorial in one of the tape cartridges. The first program I wrote took the numerical output of another program by Brüel Kjær and plotted a simple graph on the screen. The whole shebang cost 20kUS$ back then. That'd be 65k$ in today's money.
They built a 2m² brick and mortar cubicle with locking doors and windows just to house the precious piece of equipment. The company was an automaker. I guess they feared it could end up being part of Johnny Cash's Cadillac.
did anyone notice the very early PCB layout tool shown on that brochure cover ? Anyone got a copy of that laying around ... i'd like to see that run . Probably most of the HP pcb's were designed on that
Oh wow, yeah.
Surely this has to happen...