• Please review our updated Terms and Rules here

Wang Series 700B Advanced Programming Calculator Refurb.

B

BlaBla1985

Experienced Member
Joined
Aug 29, 2021
Messages
174
Location
Zuid Holland, The Netherlands
Over the coming weeks (months) I am going to attempt to bring a Wang Series 700B from 1971 back to full operation. I've had this machine for a few years now and even though it mostly seems to work, there are a couple of things that i know of that need doing. And i'd like to go over everything in one go now instead of having to tear the machine apart over and over.
IMG_5713.jpg

IMG_5718.jpg

I know that at least something in the power supply is out of spec, one of the caps is going a little crusty and it takes around 20 to 30 minutes of slowly turning up a Variac to power the machine up into a stable state. It will not run stable otherwise.

IMG_5721.jpg

Another problem are the belts in the tapedrive, at least the main drive belt seems loose, and although the tape drive transport functions work and I can start the loading of a tape, it doesn't stop automaticly and it seems to load nothing. The playback seems a little jerky too so my first guess is the tape is not running smootly enough for the head to read the contents of the tape.
If anyone knows what belts the tapedrive take, i'm all ears.

There was also a little bit of crustiness visible on the memory board (no good picture of that yet but it did concern me a little)... The memory board does look pretty with it's core memory.

IMG_5727.jpg

Any help, pointers or suggestions along the way will be very appriciated. This machine isn't like any other i have in my little collection and I'd like to learn as i go along with this project.
I got a lot of documentation with this machine, including some of the programs the original user wrote on paper (and possibly on storage tapes aswell).

The background of this machine: It was used by an engineer in a water treatment facility. I guess there much have been a lot more to the system then just the calculator, but only the calculator survived. I think when the machine became obsolete this engineer purchased/obtained it for private use. I bought the calculator with documentation from this engineers son after he passed away and the machine was found in the attic under it's dust cover. This means that the machine is really clean inside for it's age.
 
Heey Doug, i did find you're emulator while Goolgling around. I have not had the time to do try it, but i'll try to do that in the coming days to see if mine reacts different to the emulator.
I'll try contacting Rick, thanks for the suggestion.

As far as i have tested it, the core functions of the machine are fine. I can do simple calculations on it. I can use the registers to do calculations both using DIR and INDIR methods. I'm still trying to wrap my brain around the programming, i've been reading into that using the reference manual. I hope to do a little test with that tomorrow. The main problem is i think some caps in the PSU, as it takes a lot of time to power it up. And some new belts for the tape deck, as that runs a little jerky.

and maybe one of the nixies is on it's way out, the colour of the digits are off.
 
Yes, using these are quite "mind bending". I "grew up" using a Wang 600, so it's fairly familiar to me. Let me know if you have any questions.

This nixies are also affected by the transistors that drive them, and associated passive components, so you might not need a new nixie tube (I'm not sure of their availability).
 
@BlaBla1985 You doing this will give me the motivation I need to refurb my wang 700 and matching selectric typewriter/printer. Keep up informed of all your progress and lots of photos. If you need info or pictures from my 700 please let me know.
 
VERAULT said:
@BlaBla1985 You doing this will give me the motivation I need to refurb my wang 700 and matching selectric typewriter/printer. Keep up informed of all your progress and lots of photos. If you need info or pictures from my 700 please let me know.
Click to expand...
That is the idea of this thread, to catalog and share as much of the internals as possible and get the machine in tip top shape again. I would love to have some of the peripherals for the 700. But all this stuff seems fairly hard to find. I did find a Selectric modified to be a Wang 601/701 on ebay which i may pick up.

durgadas311 said:
Yes, using these are quite "mind bending". I "grew up" using a Wang 600, so it's fairly familiar to me. Let me know if you have any questions.

This nixies are also affected by the transistors that drive them, and associated passive components, so you might not need a new nixie tube (I'm not sure of their availability).
Click to expand...
Good to know there is an expert on these on the forum. I want to try out a sample program from the Reference Manual tomorrow to test some basic programming.
After that I will be taking the machine apart to see how all the board look and what is happening underneath the cards.
 
BlaBla1985 said:
Good to know there is an expert on these on the forum. I want to try out a sample program from the Reference Manual tomorrow to test some basic programming.
After that I will be taking the machine apart to see how all the board look and what is happening underneath the cards.
Click to expand...
Be aware, each slot is custom wired for its board. You must make certain they always are inserted in the same place. Make detailed lists of the board/slot assignments. The slots do not form a "bus" of any sort. And be *VERY* careful if you expose the "wire weave" ROM underneath (mounted in the bottom of the case). The slightest tug on one of those wires can break it (I think they are #40 - so are ultra fine and fragile).
 
durgadas311 said:
And be *VERY* careful if you expose the "wire weave" ROM underneath (mounted in the bottom of the case). The slightest tug on one of those wires can break it (I think they are #40 - so are ultra fine and fragile).
Click to expand...
Yikes, do you have photos of this thing?
 
oh thats terrifying. The core memory cards are the reason I have held off on my machine all this time. I never thought about a whole bottom plate ROM core...
 
Yes, those two items are probably the most fragile. They both are protected by covers, but I worry about even vibration/shock damage. Definitely don't go dropping one of these, even if it's not on your toes!
 
durgadas311 said:
Be aware, each slot is custom wired for its board. You must make certain they always are inserted in the same place. Make detailed lists of the board/slot assignments.
Click to expand...
I was afraid of that, i plan on making some sort of excell thingy with a photo of the card area and the numbers on the cards. and also number them from back to front on the ESD bags (and on top of that try to store them in order).

since the calculation part of the machine seems to work i'm not going to deep into the core rom. I will give it a good visual inspection but wasn't planning on messing with it otherwise. I will take pictures though.

One of the goals of this whole project is to get a large amount of good resolution pictures on a USB stick or something to add to the 2 binders of documentation that I got with the machine. Basicly adding a chapter to all the stuff the original owner had with it.
 
durgadas311 said:
This nixies are also affected by the transistors that drive them, and associated passive components, so you might not need a new nixie tube (I'm not sure of their availability).
Click to expand...
The reason my thought is that this nixie is on it's way out is that all the digits have the same odd colour then the other digits. it glows more white then deep orange. I've seen a nixie tube die in a clock, instead of removing it I more or less observed it as an experiment. The colour changed from deep orange to more white/blueish colour and then slowly a larger area of the digits wouldn't strike at all.
Nixie tubes are still somewhat available but if this size is easy to find i don't know, here in Europe it's mostly old Sovjet stock so those will likely be incompatable.
 
Interestingly, the program i wanted to key in from the manual was already in memory. I guess the guy who found it in his dad attic did a bit of testing of his own before selling it (it was sold to me as working for as far as he understood what was going on)... or this data has been there for a while. It does mean that the core memory card works as a way to store a program.
 
Yes, the core memory is non-volatile. Although I don't know if the integrity is guaranteed. Even large mainframes using core memory might glitch during power off/on and a byte (or more) might get corrupted. At least those are the stories I've heard. The Wang 600's I used had semiconductor memory, but they rarely powered-up cleanly.
 
Empty.jpg
I got most of the top half of the machine out and stored away in ESD bags in a protective case. Seems really empty like this. I made a spreadsheetfile to keep track of what goes where, but i also noticed the cardnumbers are printed in between the sockets on the metal framework in the machine.

It looks like some rework was done to the memory board, it looks a bit messy to be honest. I'm not really sure what to do about it... My gut says to leave it alone.
5887-Rework.jpg

The only other obvious thing are 2 caps on on the PSU daughterboards.
336-PSU-BadCap.jpg
The larger one has some clear corrosion on it. One of the 2 smaller ones has some green on the legs aswell.
 
Except for maybe the electrolytic capacitors, I tend to take the "if it ain't broke, don't fix it" approach. Trying to reflow solder on those core memory wires would turn into a nightmare very quickly. Those (fine) wires are lacquered copper, I believe. Trying to "fix" the appearance of the solder could be dangerous (to the memory). Also, while it seems tempting to replace the core memory with semiconductor RAM, the microcode depends on core memory behavior (like destructive READ) and so you'd be facing a whole other set of issues.

From what I've heard about modern vs. ancient electrolytics, you can't necessarily just match the values and be good. Those older caps behaved differently (as I've been told). At the very least, you want to increase the voltage rating - possibly a lot.
 
durgadas311 said:
I tend to take the "if it ain't broke, don't fix it" approach.
Click to expand...
I tend to like that approach as well. At this point I would change the 2 caps with the green stuff on them and maybe test a few other to be sure. The large main cap for the 12V (still in the chassis) would be an interesting one to replace if needed.
And try to replace the belts in the tapedeck, although that's an easy thing to do at any point. I like how modular this machine was built.

I don't really understand how to read the value's on those caps. I get that the large one with the green on the legs is 16V DC. But at what capacitance? 200nF, uF... just F (i doubt that).
I'm used to the newer style with the nF or uF printed with the values.

By the way, am i correct in stating that at least a part of the core memory in my machine is working? As i can read out what is stored in the program steps and i can also execute that program over and over. If i understand core memory correct, each time you access it the data gets destroyed and then written back. Which seems to be the case in mine.
If that is correct, then I won't even clean that area on the memory board, it seems to be fine and messing with it seems far to dangerous.
 
Last edited:
"200-16" = 200uF at 16VDC. You can pretty much assume uF unless otherwise specified. Of course, once I say that I'm sure you'll find an exception. I'd be tempted to see how it is being used in the schematic, and probably at least bump it to 30VDC. Most modern caps are smaller than old ones at the same value, so you should be able to fit them in the space allowed.
 
durgadas311 said:
You can pretty much assume uF unless otherwise specified.
Click to expand...
Got it, thank you.

I have been bitten more then once "assuming" things that seemed logical. Schematics would be nice, but i don't think i have anything with the paperwork i got.

BlaBla1985 said:
By the way, am i correct in stating that at least a part of the core memory in my machine is working? As i can read out what is stored in the program steps and i can also execute that program over and over. If i understand core memory correct, each time you access it the data gets destroyed and then written back. Which seems to be the case in mine.
If that is correct, then I won't even clean that area on the memory board, it seems to be fine and messing with it seems far to dangerous.
Click to expand...
I edited this in later, is this assumption correct?
 
You must log in or register to reply here.
Back
Top