durgadas311
Veteran Member
Excellent progress. Still trying to view the attachment, but something is wrong. Sent a note to the administrators. Is that same diagram available on your HoneyPi web site?
- VCF South West - June 14 - 16, Davidson-Gundy Alumni Center at University of Texas at Dallas
- VCF West - Aug 2 - 3, Computer History Museum, Mountain View, CA
- VCF Midwest - Sept 7 - 8 2024, Schaumburg, IL
- VCF SoCal - Mid February 2025, Location TBD, Southern CA
- VCF East - April 2025, Infoage Museum, Wall NJ
-
Please review our updated Terms and Rules here
Honeywell 200 resurrection
- Thread starter RobS
- Start date
RobS
Experienced Member
I don't know what went wrong there. Is this okay?
I took a look at the emulator on your website, which is equally impressive. If only my project could be completed so quickly. Thanks for that video of my programme running. My rudimentary test emulator doesn't have graphics. I threw it together just to ensure that the Pi programme worked.
Last edited:
RobS
Experienced Member
I mentioned previously that I had to assemble some replica boards for the main memory, so here is a picture of one of them next to an original to illustrate the lengths to which I am prepared to go to maintain authenticity. It is worth doing this in the main memory section whereas the control memory is my own design, so only functionally similar to the original.
Board B on the left is an original manufactured in March 1969 according to the date stamp on it while board H on the right is a replica. The PCB was made by my friendly professional PCB manufacturer from my design files and so far as possible I used original components removed from spare boards. Certainly the transistors are all originals as they have proprietary Honeywell part numbers and I don't know their characteristics and the majority of the other components also are. The most obvious variation from the original is that I omitted unused edge connector pins to keep down the cost of gold plating them, which is a significant part of the manufacturing cost. Modern PCB design software lays out straight tracks, so I had to use a copper layer overlay in the design artwork to create an accurate copy of the original while still providing the manufacturer with industry standard files to feed into their automated machines.
Board B on the left is an original manufactured in March 1969 according to the date stamp on it while board H on the right is a replica. The PCB was made by my friendly professional PCB manufacturer from my design files and so far as possible I used original components removed from spare boards. Certainly the transistors are all originals as they have proprietary Honeywell part numbers and I don't know their characteristics and the majority of the other components also are. The most obvious variation from the original is that I omitted unused edge connector pins to keep down the cost of gold plating them, which is a significant part of the manufacturing cost. Modern PCB design software lays out straight tracks, so I had to use a copper layer overlay in the design artwork to create an accurate copy of the original while still providing the manufacturer with industry standard files to feed into their automated machines.
RobS
Experienced Member
Oh dear, I just noticed an error on my control memory diagram. It shows sixteen bits in the main memory address register whereas there are only fifteen just as in the other registers. The type 201 processor only catered for two and three character addresses, i.e. twelve and fifteen bits, supporting a maximum memory size of 32k characters.
The control memory always uses fifteen bit addresses within its registers and the only requirement to make twelve bit addressing work correctly is that the increment/decrement function turns off carries to the top three bits and the character loading function doesn't change them either. As a result operations while in address mode two are confined to a single 4k bank of memory with memory wrap-around occurring within it. Of course different registers could be pointing to different 4k banks, but personally I've never tried exploiting that "feature" in a programme. As whole registers get copied one to another during normal operations any differences in the bank bits would quickly get erased, so it was never intended that the registers be used like that, I suspect. I have written mixed mode programmes which switch from one mode to another in different sections of the code, but it had to be done carefully. Designing a replica machine or even a software emulator which exactly emulates the original machine isn't actually that easy because of such quirks, but that is all part of the fun (if that's the right word to use) in this type of work.
The control memory always uses fifteen bit addresses within its registers and the only requirement to make twelve bit addressing work correctly is that the increment/decrement function turns off carries to the top three bits and the character loading function doesn't change them either. As a result operations while in address mode two are confined to a single 4k bank of memory with memory wrap-around occurring within it. Of course different registers could be pointing to different 4k banks, but personally I've never tried exploiting that "feature" in a programme. As whole registers get copied one to another during normal operations any differences in the bank bits would quickly get erased, so it was never intended that the registers be used like that, I suspect. I have written mixed mode programmes which switch from one mode to another in different sections of the code, but it had to be done carefully. Designing a replica machine or even a software emulator which exactly emulates the original machine isn't actually that easy because of such quirks, but that is all part of the fun (if that's the right word to use) in this type of work.
durgadas311
Veteran Member
Thanks, Rob. I can view these latest attachments. It is great to see those old (new) boards up close! And the block diagrams, and details like this latest post on address handling, will help keep me honest with my "virtual hardware".
RobS
Experienced Member
I'm still well despite now being age 76 and living in the now high risk south-east corner of the UK, so I'm being careful. I have spent the last year fitting out a new work area in which to assemble the computer as my tiny workroom has become too crowded, so I haven't actually been working on the project itself but hope to start again very soon. I'll soon be installing LED lighting strips in the work area, which will be a great improvement over the temporary lights that I'm currently working under, there being no windows to provide natural light and the ceiling not yet being in a state to install the lights.
One concern that I have is that I haven't recently heard from the chap in California who promised to send me a control panel for the computer some six or seven years ago. We have been in contact occasionally since then but it seems he's gone off the grid again, so I hope that all is well with his people. I emailed him at Thanksgiving just to stay in contact but received no response. Given the situation around LA at present I won't try to contact him again until I really need the panel in case he has far more important issues to deal with.
That is the thing about these online encounters, that whatever the original reason for making contact one acquires concerns about each other's welfare. Hence thank you for your enquiry. As a former governor of California once said, I'll be back.
One concern that I have is that I haven't recently heard from the chap in California who promised to send me a control panel for the computer some six or seven years ago. We have been in contact occasionally since then but it seems he's gone off the grid again, so I hope that all is well with his people. I emailed him at Thanksgiving just to stay in contact but received no response. Given the situation around LA at present I won't try to contact him again until I really need the panel in case he has far more important issues to deal with.
That is the thing about these online encounters, that whatever the original reason for making contact one acquires concerns about each other's welfare. Hence thank you for your enquiry. As a former governor of California once said, I'll be back.
RobS
Experienced Member
Good news! After many attempts I have at last regained contact with the chap in California who has an original control panel and he has pulled it out of the shed where it has been for many many years and sent me these pictures of it. He originally contacted me through VCF as a result of reading this thread so joining this site has really paid off for me. Thanks VCF.
It looks like I should be able to restore it to its original glory and hopefully it needs little more than cosmetic work on it although I have a good stock of spare parts should anything need replacing. Now we just need to arrange shipping it to me in England. Getting this original control panel really gives me hope of resurrecting the Honeywell 200 in more than just spirit.
I am also about to put the final coat of paint on the walls of my new workspace where I intend to assemble the machine as my present workroom is too small to accommodate everything. Shifting the larger parts into this new space will give me enough room in my workroom to focus on assembling my newly designed logic boards and also getting the control panel operational will make it much easier for me to test them, so I will hopefully soon be making progress on the project again.
It looks like I should be able to restore it to its original glory and hopefully it needs little more than cosmetic work on it although I have a good stock of spare parts should anything need replacing. Now we just need to arrange shipping it to me in England. Getting this original control panel really gives me hope of resurrecting the Honeywell 200 in more than just spirit.
I am also about to put the final coat of paint on the walls of my new workspace where I intend to assemble the machine as my present workroom is too small to accommodate everything. Shifting the larger parts into this new space will give me enough room in my workroom to focus on assembling my newly designed logic boards and also getting the control panel operational will make it much easier for me to test them, so I will hopefully soon be making progress on the project again.
durgadas311
Veteran Member
Great news! Maybe setup a "Go Fund Me" page for the shipping! I'd donate!
RobS
Experienced Member
Thanks for the thought but I have already set aside a considerable amount of money to fund the project as the need arises. As the likelihood of success increases so I am more inclined to invest in it. What else can a bloke in his late seventies do with his spare cash that's legal? My wife and I have already cancelled our latest luxury foreign holiday because of the pandemic. Also I pointed out to this kind chap from the outset that I didn't intend FedEx to profit more from our arrangement than he did.
Once I get the machine installed in the new workspace it will be easier to take photos of it, so I will deal with that task as soon as I can, but I will wait for the control panel to arrive first. It really isn't much to look at yet though. As with much of this type of technology the real art is in the detailed logical design, an activity in which I revel. It is hardly surprising that I like compacting the logic as much as I do when each six inch square logic PCB that I design costs $70 to be custom manufactured by a professional company, half of that cost being for the gold plated edge connector. No wonder they don't make computers like that any more.
Once I get the machine installed in the new workspace it will be easier to take photos of it, so I will deal with that task as soon as I can, but I will wait for the control panel to arrive first. It really isn't much to look at yet though. As with much of this type of technology the real art is in the detailed logical design, an activity in which I revel. It is hardly surprising that I like compacting the logic as much as I do when each six inch square logic PCB that I design costs $70 to be custom manufactured by a professional company, half of that cost being for the gold plated edge connector. No wonder they don't make computers like that any more.
m_thompson
Veteran Member
The Rhode Island Computer Museum has a control panel that I think is identical to the image in yesterday's post. I will take a picture of it Wednesday and add it to the Museum's WWW page.
RobS
Experienced Member
Thanks for that info M_T. One of the problems that I had when compiling my own website HoneyPi.org.uk was that very few pictures of the H200 were available and virtually none in colour. In fact when I saw the recent pictures of the control panel in California I wrote to a former Honeywell engineer and asked him to confirm my recollections that the panels were black and white rather than black and yellow. I took a look at the RICM website and there doesn't appear to be any Honeywell equipment mentioned, which isn't unusual.
As with most museums, computer museums probably need to have flexible relationships with conservators as preservation and conservation can have different approaches to working technology as opposed to passive artifacts. Conservation often makes considerable additions of modern material to give an artifact a valid context and this is a good thing provided that the original artifact can be distinguished from those modern additions. My contact in California chose to give me his control panel specifically because I would put it in the right context, albeit a partially contrived one. This is the difficulty where only a few components of a vintage device exist. The main memory of my machine is built using almost all original logic boards with just a few carefully constructed replicas, which are clearly marked as such, and the core memory modules themselves are originals. Hopefully now the control panel will also be an original item. On the other hand I am designing a very different control memory system, but even there I am only using components that Honeywell themselves were using in the 1960s and the control memory will be functionally very similar to the original, especially with regard to the timing of events so that the machine performs exactly as the original would have.
Elsewhere here someone recently asked in another thread what he should do with an IBM 360 control panel that he found in his loft. He could keep it as a conservation piece for his lounge and configure the lights to flash like Christmas decorations or sell or give it to a collector or museum. On the thread suggestions were posted about its potential value, no doubt to encourage him to keep it in its original condition until it found a "good" home. What are the best homes for such items though? Is a glass case in a museum better than the Frankensteinian laboratory of a hardware hacker like myself? When I eventually stagger out of my workshop a much older man after years of toil gabbling "It lives!" will anyone thank me? I have little idea. My project was conceived to turn a pile of junk, that I would have to pay good money to get recycled under our British Waste Electrical and Electronic Equipment Regulations, into something that might have a positive value. I have joined our national Computer Conservation Society, which actually specialises in the conservation and replication of vintage British computers, to get some insight into the practicalities of good practice in this activity, but the reality is that there is a lot of Frankensteinian hacking necessary that museum curators might wince at. Our national CCS actually works in close connection with our National Museum of Computing at Bletchley Park, arguably the historical home of practical computing, if war can ever be viewed as a practical activity. It is an interesting question in general, what we are really trying to achieve here apart from getting some amusement out of old pieces of what others might see as junk.
Years ago I mentioned to someone who used the working IBM 1401 machine in the Computer History Museum in California that the daughter of the chief designer of the Honeywell 200 was sending her late father's personal papers, copies of some of which she sent to me, to the museum to put in their archives and his response was "Nobody will ever get to see them again then." The question is whether museums run the risk of preserving just the bodies of such past entities for posterity without also preserving their spirits as we try to do with our reconstructions and emulations.
Here endeth the sermon to the converted. Back to the work in hand then and thanks again for aiming to get that picture posted.
As with most museums, computer museums probably need to have flexible relationships with conservators as preservation and conservation can have different approaches to working technology as opposed to passive artifacts. Conservation often makes considerable additions of modern material to give an artifact a valid context and this is a good thing provided that the original artifact can be distinguished from those modern additions. My contact in California chose to give me his control panel specifically because I would put it in the right context, albeit a partially contrived one. This is the difficulty where only a few components of a vintage device exist. The main memory of my machine is built using almost all original logic boards with just a few carefully constructed replicas, which are clearly marked as such, and the core memory modules themselves are originals. Hopefully now the control panel will also be an original item. On the other hand I am designing a very different control memory system, but even there I am only using components that Honeywell themselves were using in the 1960s and the control memory will be functionally very similar to the original, especially with regard to the timing of events so that the machine performs exactly as the original would have.
Elsewhere here someone recently asked in another thread what he should do with an IBM 360 control panel that he found in his loft. He could keep it as a conservation piece for his lounge and configure the lights to flash like Christmas decorations or sell or give it to a collector or museum. On the thread suggestions were posted about its potential value, no doubt to encourage him to keep it in its original condition until it found a "good" home. What are the best homes for such items though? Is a glass case in a museum better than the Frankensteinian laboratory of a hardware hacker like myself? When I eventually stagger out of my workshop a much older man after years of toil gabbling "It lives!" will anyone thank me? I have little idea. My project was conceived to turn a pile of junk, that I would have to pay good money to get recycled under our British Waste Electrical and Electronic Equipment Regulations, into something that might have a positive value. I have joined our national Computer Conservation Society, which actually specialises in the conservation and replication of vintage British computers, to get some insight into the practicalities of good practice in this activity, but the reality is that there is a lot of Frankensteinian hacking necessary that museum curators might wince at. Our national CCS actually works in close connection with our National Museum of Computing at Bletchley Park, arguably the historical home of practical computing, if war can ever be viewed as a practical activity. It is an interesting question in general, what we are really trying to achieve here apart from getting some amusement out of old pieces of what others might see as junk.
Years ago I mentioned to someone who used the working IBM 1401 machine in the Computer History Museum in California that the daughter of the chief designer of the Honeywell 200 was sending her late father's personal papers, copies of some of which she sent to me, to the museum to put in their archives and his response was "Nobody will ever get to see them again then." The question is whether museums run the risk of preserving just the bodies of such past entities for posterity without also preserving their spirits as we try to do with our reconstructions and emulations.
Here endeth the sermon to the converted. Back to the work in hand then and thanks again for aiming to get that picture posted.
Last edited:
Al Kossow
Documentation Wizard
There is no "National Museum of Computing" in California
If you are talking about the Computer History Museum (computerhistory.org) we have a research center that is open to the public and we have facilities for researchers to view everything in the document collection. Material is scanned and put on line by request.
Information on access can be found here, but like most of the world's museums we are still closed due to Covid. We are still taking research requests by email, we just can't have any visitors.
https://computerhistory.org/access-and-research/
Last edited:
m_thompson
Veteran Member
I will take a picture of the Honeywell panel and add it to the WWW page tomorrow.
We did have two Honeywell DPS-6 systems, but they are now spares for nuclear reactor controllers in England.
RobS
Experienced Member
Sorry Al, I got my museum names mixed up. The National Museum of Computing (TNMOC.org) is of course the one at Bletchley Park in England, which I mentioned earlier in my post. When I wrote that later part to which you refer I had actually just looked at your Computer History Museum website, but nevertheless some gremlin in my mind made me type the wrong museum name. As to the remark that that chap made, I appreciate you setting us straight on that as I had no reason to believe him or not as he was very much a passing acquaintance. With the H200 having earned the nickname "IBM 1401 killer" in its heyday it is perhaps understandable that my acquaintances with present day 1401 users have only been in passing. In fact the objective of my project is to demonstrate that the H200 was far more than just a 1401 clone as some people have claimed. I would agree with them that it wasn't even a particularly good emulator of the 1401 in some circumstances and in his personal documents, now hopefully in the possession of your Computer History Museum, its chief designer, the late Dr. William L. Gordon, mentioned that his team primarily wanted to create the best machine that they could for the price even if it meant not being able to emulate the 1401 precisely. When I wrote my "PI Factory" program (described fully on my website) as a potential demonstration of the H200's capability I used features in it that a 1401 couldn't emulate within the same amount of memory space despite my choosing to restrict my program to the most basic instruction set as provided on the cheapest H200 marketed. This wasn't intentional but just the normal way that I programmed the machine back in the 1960s to get the maximum out of it. My company had never used an IBM computer and also never used any IBM compatible software, so we were able to use the H200 to the fullest extent of its native capabilities.
Funnily enough the launch of the H200 on the market forced IBM to hurry up the launch of their new 360 and the IBM 360 itself was also mentioned in that post of mine.
M_T, regarding DPS6 computers, I had several myself for many years but the power supply company in England that provides the electricity to my own home, as it happens, contacted me to buy them after I mentioned somewhere on the Internet, possibly here in fact, that I didn't want them any more. They wanted them as sources of spares and also possibly as training machines for their engineers because they were still using them in their network. Given that they do use nuclear reactors in their system I suspect that we were both approached by the same company as they seem to have been searching the world for DPS6 parts, which end up carefully preserved in a secure warehouse near Oxford in England. Of course I was in two minds about them paying me for some old junk given that they were effectively using the money that I paid them for my electricity to do so. On the other hand I could hardly object to them eking out the last functionality of their old DPS6s to save customers like myself some money, could I? It certainly proves that eventually what goes around comes around. I don't suppose many VCF members get to pay their electricity bills with unwanted computer parts though.
Funnily enough the launch of the H200 on the market forced IBM to hurry up the launch of their new 360 and the IBM 360 itself was also mentioned in that post of mine.
M_T, regarding DPS6 computers, I had several myself for many years but the power supply company in England that provides the electricity to my own home, as it happens, contacted me to buy them after I mentioned somewhere on the Internet, possibly here in fact, that I didn't want them any more. They wanted them as sources of spares and also possibly as training machines for their engineers because they were still using them in their network. Given that they do use nuclear reactors in their system I suspect that we were both approached by the same company as they seem to have been searching the world for DPS6 parts, which end up carefully preserved in a secure warehouse near Oxford in England. Of course I was in two minds about them paying me for some old junk given that they were effectively using the money that I paid them for my electricity to do so. On the other hand I could hardly object to them eking out the last functionality of their old DPS6s to save customers like myself some money, could I? It certainly proves that eventually what goes around comes around. I don't suppose many VCF members get to pay their electricity bills with unwanted computer parts though.
Last edited:
RobS
Experienced Member
Just from looking at the photos of the interior of the control panel sent to me I have noticed something interesting about the four 3IDA0 driver boards in it.
The eight driver transistors on each board are marked S1 to S8 and have terminals marked A, C and G. This means that they aren't transistors but SCRs, which in turn implies that the lights on the panel were powered by AC current, not DC. This surprised me as I have similar type 8IDA0 driver boards removed from Honeywell Keytape machines in my stock and these have transistors which controlled DC current to the lights on the panels for those machines.
I also noticed that in the general picture of the interior that I posted previously one of the 3IDA0 boards looks much darker than the others and has a piece missing. If that board has blown then I won't have identical parts to make a replacement board, which will be a nuisance. Er, would anybody notice if one of the boards inside the control panel in the Rhodes Island Computer Museum went missing or very dark and broken? No, I'm only joking, at least for now. It will probably be a while before the control panel gets to me for me to assess its condition completely. In the meantime I will ask the present owner in California to take a look at that dark board and visually check it for serious damage.
Given two similar pieces of equipment, one in pristine condition and the other the worse from wear, I would consider it appropriate for the pristine one to be kept in a museum as an example of its type and the dilapidated one to be hacked about by an amateur restorer like myself. The control panel is in any case a plug in unit which can easily be replaced by another better one in the future if the opportunity arises.
The eight driver transistors on each board are marked S1 to S8 and have terminals marked A, C and G. This means that they aren't transistors but SCRs, which in turn implies that the lights on the panel were powered by AC current, not DC. This surprised me as I have similar type 8IDA0 driver boards removed from Honeywell Keytape machines in my stock and these have transistors which controlled DC current to the lights on the panels for those machines.
I also noticed that in the general picture of the interior that I posted previously one of the 3IDA0 boards looks much darker than the others and has a piece missing. If that board has blown then I won't have identical parts to make a replacement board, which will be a nuisance. Er, would anybody notice if one of the boards inside the control panel in the Rhodes Island Computer Museum went missing or very dark and broken? No, I'm only joking, at least for now. It will probably be a while before the control panel gets to me for me to assess its condition completely. In the meantime I will ask the present owner in California to take a look at that dark board and visually check it for serious damage.
Given two similar pieces of equipment, one in pristine condition and the other the worse from wear, I would consider it appropriate for the pristine one to be kept in a museum as an example of its type and the dilapidated one to be hacked about by an amateur restorer like myself. The control panel is in any case a plug in unit which can easily be replaced by another better one in the future if the opportunity arises.
Last edited:
m_thompson
Veteran Member
The console at the RICM is from a Honeywell 120, but looks nearly identical to the 200 console.
I put a picture here: https://www.ricomputermuseum.org/col...ll-120-console
I put a picture here: https://www.ricomputermuseum.org/col...ll-120-console
Last edited:
RobS
Experienced Member
M_T, thanks for that picture but I am concerned about the description accompanying it. Below is an extract from a 1965 Honeywell advertisement which I can vouch is genuine as I have the original magazine page containing it in my collection of H200 ephemera and have included below it the identification of the magazine and issue from the back of the page containing the advertisement.
Now this was in an advertisement covering the entire series 200 range up to the 4200, so it is just possible that in 1965 it was vapourware and just intended to attract orders for machines not yet available, but the specification of the 120 shown differs from that on the RICM website. I recollect that one of the smaller machines in the 200 series was produced in two forms, one using transistors and the other using microchips, but I assumed that the latter used the same very primitive Honeywell custom chips that I am using to build my replica. As these clearly emulate their transistorised forerunner circuits I assume that they actually were designed by Honeywell. I know that there were foreign versions of the series 200 machines that may have utilised different technology, such as those made by NEC in Japan, so another version of the 120 may have existed but evidently it was already planned if not produced long before Honeywell took over GE. The GE range of machines that Honeywell acquired were distinctly different technically from their own 200 series, so I can't imagine that any GE technology would have been dressed up to look like a series 200 machine. I may be wrong about that, but your description seems unlikely to be the complete story of the H120.
P.S.
I just read on the Computer History Museum website (I got the name right this time Al.) that GE didn't sell its computer division to Honeywell until 1970.
Now this was in an advertisement covering the entire series 200 range up to the 4200, so it is just possible that in 1965 it was vapourware and just intended to attract orders for machines not yet available, but the specification of the 120 shown differs from that on the RICM website. I recollect that one of the smaller machines in the 200 series was produced in two forms, one using transistors and the other using microchips, but I assumed that the latter used the same very primitive Honeywell custom chips that I am using to build my replica. As these clearly emulate their transistorised forerunner circuits I assume that they actually were designed by Honeywell. I know that there were foreign versions of the series 200 machines that may have utilised different technology, such as those made by NEC in Japan, so another version of the 120 may have existed but evidently it was already planned if not produced long before Honeywell took over GE. The GE range of machines that Honeywell acquired were distinctly different technically from their own 200 series, so I can't imagine that any GE technology would have been dressed up to look like a series 200 machine. I may be wrong about that, but your description seems unlikely to be the complete story of the H120.
P.S.
I just read on the Computer History Museum website (I got the name right this time Al.) that GE didn't sell its computer division to Honeywell until 1970.
Last edited:
Dwight Elvey
Veteran Member
Just being SCRs does not always mean that they were driven by AC. They may have been intended as a latching function of something. They could then capture a short pulse and hold it until it was later cleared by either shunting the SCR or opening the current source. It can then work as a holding device instead of a two transistor latch.
Dwight
Dwight